DOING RESEARCH IN BUSINESS AND MANAGEMENT

132 views 10:03 am 0 Comments May 12, 2023

DOING RESEARCH
IN BUSINESS AND
MANAGEMENT
An essential guide to planning your project
Mark Saunders
Philip Lewis
Second Edition
Doing Research in
Business and Management

At Pearson, we have a simple mission: to help people
make more of their lives through learning.
We combine innovative learning technology with trusted
content and educational expertise to provide engaging
and effective learning experiences that serve people
wherever and whenever they are learning.
From classroom to boardroom, our curriculum materials, digital
learning tools and testing programmes help to educate millions
of people worldwide – more than any other private enterprise.
Every day our work helps learning flourish, and
wherever learning flourishes, so do people.
To learn more, please visit us at
www.pearson.com/uk
Doing Research in
Business and Management
An essential guide to planning your project
Mark Saunders and Philip Lewis
Second Edition
Harlow, England • London • New York • Boston • San Francisco • Toronto • Sydney
Dubai • Singapore • Hong Kong • Tokyo • Seoul • Taipei • New Delhi
Cape Town • São Paulo • Mexico City • Madrid • Amsterdam • Munich • Paris • Milan

Pearson Education Limited
KAo Two
KAo Park
Harlow CM17 9NA
United Kingdom
Tel: +44 (0)1279 623623
Web: www.pearson.com/uk
First published 2012 (print and electronic)
Second edition published 2018 (print and electronic)
© Mark Saunders and Philip Lewis 2012, 2018 (print and electronic)
The rights of Mark Saunders and Philip Lewis to be identified as authors of this work have been asserted by them in
accordance with the Copyright, Designs and Patents Act 1988.
The print publication is protected by copyright. Prior to any prohibited reproduction, storage in a retrieval
system, distribution or transmission in any form or by any means, electronic, mechanical, recording or otherwise,
permission should be obtained from the publisher or, where applicable, a licence permitting restricted copying
in the United Kingdom should be obtained from the Copyright Licensing Agency Ltd, Barnard’s Inn, 86 Fetter
Lane, London EC4A 1EN.
The ePublication is protected by copyright and must not be copied, reproduced, transferred, distributed, leased,
licensed or publicly performed or used in any way except as specifically permitted in writing by the publishers, as
allowed under the terms and conditions under which it was purchased, or as strictly permitted by applicable copyright
law. Any unauthorised distribution or use of this text may be a direct infringement of the authors’ and the publisher’s
rights and those responsible may be liable in law accordingly.
All trademarks used herein are the property of their respective owners. The use of any trademark in this text does not
vest in the author or publisher any trademark ownership rights in such trademarks, nor does the use of such trademarks
imply any affiliation with or endorsement of this book by such owners.
Pearson Education is not responsible for the content of third-party Internet sites.
ISBN: 978-1-292-13352-2 (print)
978-1-292-13353-9 (PDF)
978-1-292-22231-8 (ePub)
British Library Cataloguing-in-Publication Data
A catalogue record for the print edition is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Saunders, M. N. K., author. | Lewis, Philip, 1945– author.
Title: Doing research in business and management / Mark Saunders and Philip
Lewis.
Description: 2nd edition. | Harlow, England : Pearson, 2018. | Includes
bibliographical references and index.
Identifiers: LCCN 2017023723 | ISBN 9781292133522 (print) | ISBN 9781292133539
(pdf) | ISBN 9781292222318 (epub)
Subjects: LCSH: Business—Research.
Classification: LCC HD30.4 .S277 2018 | DDC 650.072—dc23
LC record available at https://lccn.loc.gov/2017023723
10 9 8 7 6 5 4 3 2 1
22 21 20 19 18
Print edition typeset in 9/13pt Stone Serif ITC Pro by iEnergizer Aptara
®, Ltd.
Printed and bound in Malaysia
NoTE THAT ANY PAGE CRoSS REFERENCES REFER To THE PRINT EDITIoN

About this text
About the authors
Acknowledgements
xi
xv
xvi

1 Choosing your research topic 1
2 Reviewing the literature critically 31
3 Managing the research process 58
4 Using secondary data 85
5 Choosing your research design 104
6 Collecting data 137
7 Analysing data 179
8 Writing and presenting the research proposal 215
Appendix 1: How to reference 239
Bibliography 247
Index 250
Brief contents
This page intentionally left blank
About this text xi
About the authors xv
Acknowledgements xvi
1 Choosing your research topic 1
1.1 Why you should read this chapter 1
1.2 Why choosing the right research topic is so important 2
1.3 Why choosing a research topic is difficult 4
1.4 Ten ways to generate ideas for a research topic 5
1.5 How to refine research topic ideas 11
1.6 What makes a good research topic? 15
1.7 How to turn a research idea into a research project 18
Summary 28
Thinking about your research topic 29
References 30
2 Reviewing the literature critically 31
2.1 Why you should read this chapter 31
2.2 What a critical literature review is 32
2.3 Why it is important to review the literature critically 34
2.4 The types of literature available to you 37
2.5 Searching for and obtaining literature 39
2.6 Evaluating the usefulness of literature to your research 46
2.7 Reading, noting and correctly referencing useful literature 47
2.8 Drafting your critical literature review 52
Summary 55
Thinking about your critical literature review 56
References 57
3 Managing the research process 58
3.1 Why you should read this chapter 58
3.2 Getting access to your research organisation, respondents
and participants 59
3.3 What about access to information? 61
3.4 Six strategies for making sure that you get the organisational
access you want 62
3.5 Managing yourself 69
3.6 Managing your supervisor 71
Contents
viii Contents
3.7 Managing your university 74
3.8 The ethics of doing research 75
Summary 83
Thinking about your research process 84
References 84
4 Using secondary data 85
4.1 Why you should read this chapter 85
4.2 Forms secondary data can take 86
4.3 The potential of secondary data 93
4.4 Possible pitfalls of using secondary data 96
4.5 Assessing the suitability of secondary data 98
4.6 Where and how to find secondary data 100
Summary 101
Thinking about using secondary data 102
References 103
5 Choosing your research design 104
5.1 Why you should read this chapter 104
5.2 The importance of research philosophy 106
5.3 Differing approaches to theory development: deduction,
induction and abduction 111
5.4 Differing purposes: exploratory, descriptive and explanatory studies 115
5.5 Differing strategies 119
5.6 Making sure your research conclusions are believable 131
Summary 135
Thinking about your research design 136
References 136
6 Collecting data 137
6.1 Why you should read this chapter 137
6.2 Selecting samples 138
6.3 Collecting data using questionnaires 148
6.4 Collecting data using semi-structured or unstructured interviews 158
6.5 Collecting data using observation 168
Summary 176
Thinking about collecting data 177
References 177
7 Analysing data 179
7.1 Why you should read this chapter 179
7.2 Different types of data 180

Contents ix
7.3 Analysing data quantitatively 183
7.4 Analysing data qualitatively 202
Summary 212
Thinking about analysing data 213
References 214
8 Writing and presenting the research proposal 215
8.1 Why you should read this chapter 215
8.2 The importance of the research proposal 216
8.3 When you should write your research proposal 220
8.4 What you should include in your research proposal 220
8.5 The style you should use to write your research proposal 227
8.6 How your research proposal will be judged 231
Summary 237
Thinking about your research proposal 237
References 238
Appendix 1: How to reference 239
Bibliography 247
Index 250
Supporting resources
Visit www.pearsoned.co.uk/saunders to fnd valuable online resources:
Companion Website for students
Multiple-choice questions to test your learning
True or false questions to test your understanding
Additional case studies
Guides to SPSS and Qualitrics
For instructors
Complete, downloadable Instructor’s Manual
PowerPoint slides that can be downloaded and used for presentations
Also: The regularly maintained Companion Website provides the following features:
Search tool to help locate specific items of content
Email results and profile tools to send results of quizzes to instructors
online help and support to assist with website usage and troubleshooting
For more information please contact your local Pearson Education sales
representative or visit
www.pearsoned.co.uk/saunders
This page intentionally left blank
It is now 20 years since we collaborated in the writing of our first research methods
book and six years since we wrote the first edition of
Doing Research in Business and
Management
. The success of both books suggests that research methods is a popular subject with business and management students. This may be so. But we think that it has
more to do with the fact that research methods is a complex area – one where it is easy
to do things, but much less easy to do things right.
In writing the second edition of
Doing Research in Business and Management, we have
responded to the many comments we have received regarding the previous edition as
well as recent developments in research methods and methodology. In particular this
has led us to revise Chapter 4 to take account fully of the numerous sources of secondary data available online; Chapter 5 to incorporate fully revised sections on philosophical underpinnings of management research including a discussion of postmodernism,
and on different approaches to theory development including a discussion of abduction; Chapter 6 to incorporate a new section on observation; and an appendix on how
to reference. Alongside this we have taken the opportunity to also update examples and
references as well as revise tables and figures. Inevitably, the body of knowledge of
research methods has developed further since 2012, and we have revised all chapters
accordingly. Our experiences of teaching and supervising students and working
through the methods with them have suggested alternative ways to explain concepts,
and we have incorporated these where appropriate. However, the basic structure
remains the same as the first edition.
When we wrote the first edition of
Doing Research in Business and Management, we
had one overall mission in mind. That was to write a text that was clear and straightforward and explained things in a way that lost none of the complexity, or academic rigour, of the subject. In writing the second edition this mission has not altered. We still
feel just as passionate about clear communication.
In fully revising
Doing Research in Business and Management, we have taken into
account that although some degree programmes require students to complete an
assessed research project, they may be told that, rather than collect their own data, they
should use only data that have already been collected for some other purpose (secondary data) or, alternatively, write an extended review of the literature. For a second category of students on undergraduate programmes, the extent of their research work is a
research methods module which is assessed by a research proposal. There is a third category, those business and management students who opt not to do a research project at
all. For those undertaking research to be assessed by a written project report, we aim to
help in all aspects of the research process: from thinking of a topic through to writing
the final submission. We therefore include material, in Chapter 3, on managing the
research process, as well as chapters on using secondary data (Chapter 4) and reviewing
About this text
xii About this text
the literature critically (Chapter 2). There are also two chapters (6 and 7) on collecting
and analysing data, as an understanding of these is important for all types of research
projects as well as preparing a research proposal. If you’re taking a research methods
module which is assessed by a research proposal, you will find that there is considerable
emphasis on the preparation of a research proposal. Indeed, Chapter 8 deals specifically
with writing the research proposal.
It may sound strange, but we think that business and management students in the
third category, those who opt not to do a research project at all, can gain just as much
from this book as those in the other two categories. As a student, you will spend much
of your time studying material which is the result of careful research that has been scrutinised by the research community prior to publication. This scrutiny is a guarantee of
good quality: that you should put your faith in what you have read. However, some of
what you read may not have been through quite such a rigorous process. Knowing
something about the research process enables you to ask the right questions of the
material you are studying. It gives you the sense of healthy scepticism that is the hallmark of a university education.
How you might use this text
We don’t anticipate that you will read this text progressively from Chapter 1 through
to Chapter 8. In fact, you may not read all the chapters, although we certainly hope
that you will! The reason, we suspect, is that you will choose those chapters that
meet your own needs. This may be because you are in one of the categories we mentioned earlier, have specific questions about the research process you need to answer,
or it may be that your research methods lecturers specify certain chapters. We’ve
written the chapters in such a way that they can be read on their own without
recourse to the other chapters. To some extent, they draw inevitably on material
from other chapters directly. Where this is so, we have cross-referenced to the relevant chapter. But the point remains that you can pick up any chapter in isolation
and make sense of it.
This book is not a self-study text in the truest sense; there are no questions with
model answers! However, we have included points in each chapter which facilitate an
element of independent learning. Each chapter begins with a summary of content
which we call ‘Why read this chapter?’ This gives you some idea of the chapter content
and the approach we have taken on the topic being discussed. Each chapter contains a
small number of examples of research called ‘Research in practice’. These serve to illustrate in a practical manner some of the points being made in the chapter, in much the
same way as a lecturer would give practical examples in a research methods lecture.
Every chapter ends with a summary of the main points in the chapter and a section
called ‘Thinking about . . .’ Here we make suggestions as to how you may test and reinforce the learning you have achieved during the reading of the chapter. Throughout the
book, key research terms we use are isolated and placed in ‘Definition’ boxes to make it
easy for you to refresh your understanding of these terms as you read through each
chapter.

About this text xiii
What’s in the text?
Chapter 1 deals with the first issue you will encounter in the research process: choosing
the right research topic. We suggest some novel ways in which you may decide upon
your topic, offer guidance in deciding what constitutes an effective research topic and
consider some topics which may be problematic. In the latter part of the chapter, we
deal with the issue of defining suitable research questions and objectives. The chapter
ends with a discussion on what is meant by the all-important term ‘theory’.
In Chapter 2, we approach the subject of the critical literature review. We offer some
practical suggestions on the way you may go about approaching your literature review
and actually conducting it, using full-text databases of academic articles. The chapter
also explains what constitutes an effective critical literature review and offers guidance
in how it may be structured.
Chapter 3 is concerned with practical issues regarding gaining access to organisations from which you may collect your own research data. In this chapter, we also consider the issues of self-management you may face in conducting your research,
particularly the effective use of resources such as time. The management of other
aspects of the research process is also discussed, such as your supervisor, university and
those from whom you collect your data. We also help you to think about the ways in
which you adhere to the code of research ethics that you will be required to observe.
In Chapter 4 we consider the use of secondary data. We discuss the valuable role
which secondary data may play in your research and the reasons you may use secondary data. The ready availability of a wealth of secondary data online is considered. We
also warn you about some of the pitfalls inherent in the use of secondary data and how
to assess its value to your own research project.
The subject of Chapter 5 is research strategy. This involves a consideration of the
main philosophies you may adopt and the ways in which they affect choice of strategy.
We discuss the different types of research strategy, with an emphasis on the possibility
of mixing strategies in one research project. We end the chapter with a discussion of the
importance of validity and reliability: ensuring that your research results and conclusions are believable.
Chapter 6 gets to the core of the research process: the collection of data. We first
explain how to choose a sample. We then consider three frequently used methods of
collecting primary data in more detail. In this we look at how to design and distribute
effective questionnaires, including the use of Internet questionnaires; conduct face-toface, telephone and Internet-mediated interviews; and undertake structured and
unstructured observations.
In Chapter 7 we deal with the process of data analysis. We discuss the two types of
data – quantitative and qualitative – and the ways in which these data may be prepared
for analysis and actually analysed. The use of statistics in both the presentation and
analysis of data is explained with particular emphasis on the use of different software
packages. We also discuss ways in which qualitative data may be prepared for analysis
and analysed. As with the analysis of quantitative data, we emphasise the way in which
you may develop theory from the analysed data.

xiv About this text
Chapter 8 is devoted to the writing of your research proposal. We explain how the
process of writing clarifies your ideas, and we emphasise the importance of treating the
research proposal as an item of ‘work in progress’ by constantly revising it. The chapter
also includes a discussion on what content the proposal should contain, how it may be
structured and the appropriate writing style to be adopted. Finally, we suggest some of
the criteria against which the quality of your research proposal may be assessed.
We hope you will learn a lot from this book: that’s why it exists! But we also hope
that you will enjoy reading it. Doing your research project should be fun!
Mark and Phil
July 2017

Mark N.K. Saunders BA, MSc, PGCE, PhD, Chartered FCIPD is Professor of Business
Research Methods at Birmingham Business School and Director of Postgraduate
Research Methods Training for the College of Social Sciences at the University of
Birmingham. He is a Fellow of the British Academy of Management and member of the
Fellows’ College. Mark currently holds visiting professorships at the Universities of
Surrey and of Worcester. He teaches research methods to master’s and doctoral students
as well as supervising master’s dissertations and research degrees. Mark has published
articles on research methods, and human resource aspects of the management of
change including trust and organisational learning, in a range of journals including
British Journal of Management, Human Relations, Journal of Small Business Management,
Management Learning, R and D Management and Social Science and Medicine. He is coauthor with Phil Lewis and Adrian Thornhill of Research Methods for Business Students,
currently in its seventh edition and, with Bill Lee, co-author of
Doing Case Study Research
for Business and Management Students.
He is also co-editor with Bill Lee and Vadake
Narayanan of the
Mastering Business Research Methods book series and editor of the
Handbooks of Research Methods in Management book series. Mark undertakes research
and consultancy in public, private and not-for-profit sectors. Prior to becoming an academic, he had a variety of research jobs in local government.
Philip J. Lewis BA, PhD, MSc, MCIPD, PGDipM, Cert Ed began his career in HR as a
training adviser with the Distributive Industry Training Board. He then taught HRM
and research methods in three UK universities. He studied part-time for degrees with
the Open University and the University of Bath, from which he gained an MSc in
Industrial Relations and a PhD for his research on performance pay in retail financial
services. He is co-author with Mark Saunders and Adrian Thornhill of
Research Methods
for Business Students
, currently in its seventh edition, and of Employee Relations:
Understanding the Employment Relationship
; and with Adrian Thornhill, Mike Millmore
and Mark Saunders of
Managing Change: A Human Resource Strategy Approach; and with
Mark Saunders, Adrian Thornhill, Mike Millmore and Trevor Morrow of
Strategic Human
Resource Management
. He has undertaken consultancy in both the public and private
sectors.
About the authors
We are grateful to the following for permission to reproduce copyright material:
Figures
Figure 1.1 from Competitive Advantage Maps, Working Paper, Centre for Strategy &
Performance, Institute for Manufacturing (Platts, K., and Khater, M., 2006), Reproduced
with permission by Dr M. Khater from the Institute for Manufacturing; Figure on
page 95 from Eurostat (2016) Copyright European Communities, 2016, Source: Eurostat,
http://epp.eurostat.ec.europa.eu, © European Union, 1995–2017; Figure 7.3 from
Organic farming. Organic crop area on the rise in the EU. Two million hectares more
since 2010,
Eurostat News Release, Source: Eurostat, http://epp.eurostat.ec.europa.eu,
© European Union, 1995–2017.
Screenshots
Screenshot 2.2 from www.emeraldinsight.com, www.emeraldinsight.com; Screenshot
2.3 from https://www.ebsco.com, reproduced with permission of EBSCO Information
Services; Screenshots on page 197, 7.7 from IBM SPSS Statistics Viewer, screenshots
reprinted courtesy of International Business Machines Corporation, © International
Business Machines Corporation. SPSS was acquired by IBM in October, 2009. IBM, the
IBM logo, ibm.com, and SPSS are trademarks of International Business Machines Corp.,
registered in many jurisdictions worldwide. Other product and service names might be
trademarks of IBM or other companies. A current list of IBM trademarks is available on
the Web at “IBM Copyright and trademark information” at www.ibm.com/legal/
copytrade.shtml.
Tables
Table 2.6 adapted from Critical Reading and Writing for Postgraduates, 3rd ed., Sage
Publications Ltd (Wallace, M., and Wray, A., 2016)
Text
Extract 1.1 adapted from Rural broadband and digital-only services, Seventh Report of
Session 2014–15,
House of Commons Environment, Food and Rural Affairs, p. 3, Contains
Parliamentary information licensed under the Open Parliament Licence v3.0;
Extract 2.2 from SME innovation and learning: the role of networks and crisis events,
European Journal of Training and Development, Vol. 38, Issue 1/2, pp. 136–49 (Saunders,
M. N. K., Gray, D. E., and Goregaokar, H., 2014), doi: 10.1108/EJTD-07-2013-0073; Article
3.1 from Sweden leads the race to become cashless society,
www.theguardian.com,
04/06/2016 (Henley, J.), Copyright Guardian News & Media Ltd. 2017; Extract on page
76 from
Research Ethics for Research involving Human Participants: Code of Practice (2016),
Acknowledgements
Acknowledgements xvii
https://www.brookes.ac.uk/Documents/Research/Policies-and-codes-of-practice/
ethics_codeofpractice/, Oxford Brookes University Research Ethics Committee; Extract
on page 94 from Eurostat (2016) Copyright European Communities, 2016, Source:
Eurostat, http://epp.eurostat.ec.europa.eu, © European Union, 1995–2017; Extract on
page 95 from
Methodological Manual for Tourism Statistics, Version 3.1, p. 59 (2014),
Source: Eurostat, http://epp.eurostat.ec.europa.eu, © European Union, 1995–2017;
Extract 4.1 from http://www.edelman.com/insights/intellectual-property/2016-
edelman-trust-barometer/global-results/, Daniel J. Edelman, Inc. and/or its subsidiaries
and affiliates are the owners of the TRUSTBAROMETER trademark and copyright in the
TRUSTBAROMETER surveys worldwide, which are used under license. These materials
are not sponsored or endorsed by Daniel J. Edelman, Inc.; Article 5.1 from Emergence of
‘serial returners’ – online shoppers who habitually over order and take advantage of free
returns – hinders growth of UK businesses,
Barclaycard, Barclaycard 2017; Extract 7.6
from The influence of culture on trust judgments in customer relationship development by ethnic minority small businesses,
Journal of Small Business Management, Vol.
52, Issue 1, pp. 67–8 (Altinay, L., Saunders, M. N. K., Wang, C. L.), Copyright © 2013,
John Wiley and Sons; Extract on page 288 from
The Sociological Imagination, OXFORD
UNIVERSITY PRESS, INCORPORATED (MILLS, C. WRIGHT 1959) pp. 218–19,
Reproduced with permission of OXFORD UNIVERSITY PRESS, INCORPORATED in the
format Book via Copyright Clearance Center Inc.
Picture Credits
The publisher would like to thank the following for their kind permission to reproduce
their photographs:
Mark N. K. Saunders: 205
Cover image:
Front: Mark N. K. Saunders
All other images © Pearson Education
In some instances we have been unable to trace the owners of copyright material, and
we would appreciate any information that would enable us to do so.

This page intentionally left blank
Chapter 1
Choosing your research topic
This is a big moment in your life. You are about to embark on a voyage of discovery. You
will discover a lot about the research process, the topic you are going to research and, we
hope, a lot about yourself too. For many of you, the research project is the one part of your
course where you have an opportunity to choose what you are going to study and the way
in which you study it. We hope that you see this as an exciting opportunity because we
believe strongly that’s what it is. It’s your chance to express your individuality, your ingenuity and imagination, your resourcefulness and, above all, your personal organisational skills.
These attributes have always been important. But in the twenty-first century, they are
more important than ever. Why do we say this? It’s because we think that all the social,
economic and technological changes of the last few years have empowered us all to take
charge of our own lives to a greater extent than ever before. So take charge of your
research topic now!
The overall purpose of this chapter is to enable you to get your research project off to a
good start by choosing a topic to research that will give you the best chance of succeeding
and passing this important component of your course. It’s worth bearing in mind that however good you may be at all the relevant skills that go into producing a good research project,
you will give yourself a better chance of succeeding if you have chosen your topic wisely.
In this chapter, we talk about why choosing the right research topic is so important. We
then explain why, for many of us, the choice of topic is so difficult. The choice is made
easier if this decision-making process is tackled in a systematic way. In the chapter, we
outline some of the procedures for adopting a systematic approach. Then, having got to
the stage where you have chosen a topic, we examine ways in which the topic you have
chosen can be refined in such a way that it is acceptable to your assessors and will provide
you with the maximum amount of satisfaction.
We end the chapter with a consideration of what makes a good research topic, and
some help on writing research questions and objectives.
1.1 Why you should read this chapter
2 Chapter 1 Choosing your research topic
It is possible, of course, that you may be constrained in your choice of research topic.
Your university may define strict limits outside which you may not stray. Alternatively, it
is possible that an employer has asked you to undertake a piece of research. In either
case, some of the points in the next two sections on choosing the right research topic
and generating research ideas may not apply to you directly. However, we encourage you
not to ignore the points made. It may be that although you may not have a free choice of
the general topic, the way in which you approach it may be entirely your decision.
Now let’s look at some of the reasons why choosing the right research topic is so
important.
You have to live with it
The decision about which research topic to choose is something you will have to live
with, maybe for as long as a year or more. We mean this in two senses. First, it is a topic
you will become intimately familiar with, so it makes sense to choose something that
you will enjoy. Ask yourself: what am I really interested in? OK, so it may be that football is your passion. Well, football nowadays is big business, particularly in the major
leagues in Europe such as the English Premier League. So if your passion has a business
dimension, then maybe there is a research topic to be pursued. One of our student’s
undergraduate projects explored the reasons why attendances at matches were generally lower when the match was screened live on television. This led into interesting
areas such as the affiliation aspect of motivation theory, where some people, for example a football team’s fans, have a need to be affiliated with like-minded people and are
motivated towards interaction with these people, leading them to attend matches. This
posed a question about the extent to which a comfortable armchair and the economic
benefits of not attending the match in person overrode the need for affiliation!
The second sense in which we mean that the decision about your choice of research
topic is something you will have to live with is that when you have chosen it, normally
there is no going back. You will make life much more difficult for yourself if you find out
after one half of the time period allotted to your research that you have chosen the
wrong topic. It may be possible to change even at that stage, but you spend the rest of the
allotted time playing catch-up. We talk later in this chapter about problems in choosing
a research topic, and the consequences of making the wrong choice. So it is better to
spend the time at the outset making sure it’s right for you. It is a time-consuming process. Many students have remarked to us in the past that they thought that choosing
their topic would be easy. It may be easy to choose one that interests you, but turning it
into a viable proposition for your course may take much longer than you thought.
It will be better choosing a topic that will both exploit and develop
your knowledge and skills
As well as choosing a topic that you will enjoy, it obviously makes sense to choose a
topic that you are capable of doing well. Making a list of your skills and knowledge
1.2 Why choosing the right research topic is so important
1.2 Why choosing the right research topic is so important 3
seems a good starting point. Here are a few questions you can ask yourself to help you
prepare the list.
What are your personal strengths and weaknesses?
We are all better at some things than others. You may have discovered strengths on your
course that you may want to exploit. These may relate to your background experience or
particular skills that you can practise in the data collection and analysis stages of the
research. The knowledge gained in previous or current work experience is a good knowledge source for many students. You are more likely to know your way around some of the
areas that need specialist knowledge. In addition, this specialist knowledge will lead more
easily to an informed research question that needs answering. Alternatively, you may be
keen to learn about an industry that’s new to you. For example, you may be fascinated by
software design and want to learn more about that industry with a view to possible
employment. In this case, you will need to ask yourself whether locating a research project in that industry will give you an equal chance of success compared to an industry
with which you may be familiar. It’s also a good idea to think about the modules you have
studied and those in which you have had success. This will give you a confident start.
The data collection and analysis methods you adopt offer slightly different options.
Here, you may be experienced at interviewing but less so at designing questionnaires.
Do you exploit your expertise or decide to learn the skill of questionnaire design? Of
course, you can include an element of both. The choices you make may be based on
practical as well as personal development considerations. It’s not much good having
learned a lot but not passed the module!
What knowledge and skills do you think you will need in the future?
This may be quite a difficult question to answer for many of us. Few of us could have
predicted 25 years ago the extent to which we all now need information technology
skills in both our work and home lives. Yet it may be possible for you to predict some of
the generic skills you will need for effective personal performance. Some of these, such
as influencing others and conducting meetings, you may have encountered during
your course. The opportunity may present itself in your research to practise some of
these skills, particularly in the data collection stage. In the same way, your choice of
topic could help develop your specialist knowledge of an aspect of your chosen area
of employment.
What resources can you draw upon to help?
Perhaps the value of this question is most evident when it reveals the absence of
resources. Most of you will have access to key people such as lecturers, managers and
colleagues. The extent to which these can be of assistance will, of course, vary. It will be
very valuable to be able to consult an ‘expert’ in the subject area you are studying for
your research but, if you have chosen a fairly specialist field, the absence of such assistance may be a considerable block to your progress. You will, of course, have access to
information technology, but such issues as the processing of questionnaire answers
and the analysis of questionnaire data is a complex and demanding affair if you have no
experience in this. Don’t be afraid to ask for help!

4 Chapter 1 Choosing your research topic
Will your choice of topic help you pass the whole course?
Although we have put the emphasis here upon self-development, the point remains
that you must choose a topic that will allow you to meet the assessment requirements
and will give you the best possible chance of ultimate success. If there is one fundamental lesson that we have learned as a result of supervising many research projects over the
years, it is this: the earlier you start deciding upon your topic, the more likely you will be
to choose the right one and ensure final success!
There is no question that for many of us, choosing the right topic is one of the most difficult aspects of the whole research process. At this stage, you are on your own! It has to
come from you; and making decisions, which have important consequences, is often
difficult for most of us. So, what are some of the reasons why this one may be particularly difficult? Let’s have a look at some of them.
There is simply too much choice
It’s wonderful living in an age when so much information is available at the end of our
fingertips. But this can lead us to think that there is no question that has not been
asked and no problem which has not been solved. Whatever it is we are interested in,
there appears to be a vast amount of information available, much of which there is
never enough time to read. This results in the inevitable feeling that whatever it is you
have in mind will have been done before. Well, it probably has, but maybe not in the
way you intend to do it. But that doesn’t mean that you can’t tackle it. One of the most
popular undergraduate research topics is worker motivation. ‘What is that workers
value most about their working lives?’ is a frequently asked question. The textbooks
are full of generic answers to this question, and your university library will be full of
project reports which have asked the same question. Yet there are lots of managers in
organisations who need to know the answer in respect of their own employees. In
other words, even the most familiar topic can be applied to many different, specific
situations.
The fear it will be too difficult
The research process is challenging enough without making it more difficult by choosing a topic that stretches you too far. How will you know if that is the case? Have a look
at the literature on a topic that interests you as a possible choice. Maybe the way in
which the topic is covered has an overly theoretical approach which makes it too difficult for you to ‘think your way in’. The motivation of people to work has a distinctly
practical feel. But researching the way in which the brain operates to direct our enthusiasm to one interest rather than another seems to emphasise the biological aspects
rather than the business perspective, for which your course has prepared you.
1.3 Why choosing a research topic is difficult
1.4 Ten ways to generate ideas for a research topic 5
The fear that it will be insufficiently theoretical
It’s quite understandable that you should feel a mild sense of panic when you go to your
supervisor to explain with great enthusiasm your choice of research topic only to be
greeted by ‘yes, very interesting, but what role will theory play?’. This is where it starts
to get tricky. But don’t despair, because theory has a role to play in all project reports.
It’s just a question of how and where you use it. Later in this chapter we explore the role
of theory in writing research questions and objectives. And in Chapter 5 we explain
that theory can be used as a ‘way in’ to your research by setting up theoretical propositions which can be tested. It can also be used as a lens through which you can study
your data, or a structure against which you can perform your data analysis.
In case you are unconvinced, just consider this statement: ‘students read research
methods textbooks in the hope that they perform more effectively in their research module’. That’s a theory, and what you are doing right now is evidence that it is accurate!
The temptation to re-use work you have already done
It is tempting to take the easy way out and use an assignment which you have written for
a previous purpose and just enlarge it to make it into a research project. There is a similar
temptation for part-time students who perhaps have produced a research report at work.
The trouble is that it never quite ‘fits’. It’s a bit like fitting a wheel to a bike that’s not quite
the right size. It may be OK, but it’s never more than that. You are likely to spend more
time getting it to fit than you would spend on thinking through a purpose-made topic.
Now that we have given you some general guidelines on choosing a suitable research
topic, let’s look at some techniques for deciding on the topic itself. We have listed our
favourite 10 of these in Table 1.1. They are in no particular order.
1.4 Ten ways to generate ideas for a research topic
Table 1.1 Ten techniques for generating research ideas
1 Thinking
2 Looking at past project titles
3 Using past projects from the university library
4 Using past course assignments
5 Using relevant literature
6 Following the news media
7 Brainstorming
8 Concept mapping
9 Making a note of ideas
10 Discussion with helpers
6 Chapter 1 Choosing your research topic
Thinking
We are bound to start with this one, because there is no escaping it! By thinking, we
mean really thinking. Keep the need to select a topic in your mind all the time – when
you watch TV, read a newspaper, browse relevant web pages, talk to colleagues, talk and
listen in seminars, discuss issues with your lecturers. Probably, like us, prompted by
your study of business and management, questions often go through your mind when
you are in stores, airports and buses – questions such as ‘How are the work rotas for
these people organised?’; ‘How can the cost of delivering this service be reduced while
maintaining quality of service?’ This is what you may call ‘background thinking’. Now
let’s look at a series of more specific techniques for generating ideas.
Looking at past project titles
You are not the first person to have trodden the path, so it’s a good idea to look at what
those before you have attempted. You may find a list of past research project titles for
your course in your university library; alternatively, your course leader may have one.
This will give you an idea of the sort of topic that may be suitable. It will also fire your
imagination and help you to start thinking about how that title may relate to something
that you have thought about. Don’t worry about what you think may not be a particularly well-written title. And do bear in mind that in some universities, all past projects
are placed in the library whether they are bare passes or distinctions. So the fact that a
project is in your library does not necessarily mean that it’s a good piece of work. The
point is to generate ideas for your project, not work out what makes a good project.
Using past projects
Having delved into the university library to look for past project titles, why not spend
some more time checking the projects which have caught your attention? Raimond (1993)
suggests a useful method for generating research topic ideas this way (see Table 1.2).
Table 1.2 Four steps for generating research topic ideas using past projects from the
university library
1 Select six projects that you like.
2 For each of these six projects, note down your first thoughts to answer these three questions
(if responses for different projects are the same, this does not matter):
(a) What appeals to you about the project?
(b) What is good about the project?
(c) Why is the project good?
3 Select three projects that you do not like.
4 For each of these three projects, note down your first thoughts to answer these three
questions (if responses for different projects are the same, or cannot be clearly expressed, this
does not matter; note them down anyway):
(a) What do you dislike about the project?
(b) What is bad about the project?
(c) Why is the project bad?

1.4 Ten ways to generate ideas for a research topic 7
Having completed these four steps, you will have a note of the things that you like
and dislike about projects and, of equal importance, what you consider makes good and
poor projects. What’s more, that list will be personal to you. It’s your opinion, and that
is what you can use to guide your choice for your own research topic.
Using past course assignments
Many students find that past course assignments serve as a good starting point for
research topic ideas. It seems logical to develop the work of an assignment in which you
have had some choice of topic, particularly when you have enjoyed it and received a
good grade. However, do bear in mind the maxim that ‘all research ends with ideas for
more research’. Look hard at what you have done and ask yourself: ‘Are there questions
posed by my work which I have not yet answered?’
Beware of developing the assignment because you got a good grade! There must be
scope for development, and it must meet the assessment criteria. For example, many
universities have self-plagiarism rules to prevent students from re-using work that they
have already submitted as an assignment.
Using relevant literature
Since what you are going to tackle should use the literature relevant to your topic, it seems
sensible to start examining that literature. Let’s assume that you have decided to look at a
possible research topic that is a development of a module you have enjoyed. Go back to
your lecture notes and course textbooks on that topic and make a note of the names of
relevant authors. This will give you a basis on which to undertake a preliminary search.
This should help you to produce a list of articles, books, reports and other items.
A particularly valuable literature source of research topic ideas is academic review articles. They are valuable because they contain both a thorough review of the state of
knowledge in that topic area and pointers towards areas where further research needs to
be undertaken. Browsing recent journals in your field is also a good source of possible
research ideas. For many subject areas, your project supervisor will be able to suggest possible recent review articles, or articles that contain recommendations for further work.
Books, by contrast, are often less up to date than journal articles. But they do often
contain a good overview of research that has been undertaken, which may suggest ideas
to you. Reports may also be of use. The most recently published are usually up to date
and, again, often contain recommendations that may form the basis of your research
idea. An example of such a report is shown as Research in practice 1.1.
Using reports to generate research topic ideas
Access to broadband in the UK is inconsistent, the impact being particularly in rural
areas where speeds are unacceptably slow. Written evidence to the House of Commons
Environment, Food and Rural Affairs Committee explains how poor broadband can lead
Research in practice 1.1

8 Chapter 1 Choosing your research topic
Table 1.3 lists some useful questions to ask when searching articles and reports for
possible research topic ideas. The answers to these can help progress your choice of
topic.
Table 1.3 Useful questions to ask when searching articles and reports for possible
research topic ideas
•  What did the authors conclude?
•  What alternative conceptual models, explanations or hypotheses did the authors consider?
•  What methods did the authors use to approach the problem?
•    Do you accept the authors’ conclusions? If not, are there other methods that could allow you
to test their conclusion?
•  Does the authors’ research suggest new ways to interpret a different problem?
•  Are there other problems that could be studied using the same methods?
Following the news media
We would always encourage you to go to the academic literature as your first port of call,
but don’t ignore the value of keeping up to date with items in the news. News media can
be a very rich source of ideas. The stories which occur every day in the ‘quality’ newspapers (e.g.
The Times, The Financial Times, The Guardian, The Independent and The Daily
Telegraph
) in both print and online versions may provide ideas which relate directly to a
possible research topic. Don’t forget, there may be other ideas which flow from the
main story. On the morning of writing this section in 2016, it was announced that
to a range of problems: from reduced access to online learning resources for students,
families being unable to use everyday online services and the effectiveness of rural
businesses being severely affected. Slow broadband can produce a feeling of a two-tier
society, with rural communities suffering markedly due to infrastructure problems which
make them harder to reach.
The committee expressed concern that BT, the infrastructure developer, told it that
the 2015 target of 95% of premises receiving superfast broadband by 2017 may slip.
The committee recommended that the government unit overseeing broadband development should insist upon the 2017 target being met. Moreover, a target date for when
the last 5% of premises will obtain access to superfast broadband coverage must be
published.
The importance of good broadband for all is highlighted by the government’s policy
of providing its services ‘digital-by-default’. This policy has clear ramifications when
broadband access is limited or non-existent.
The report emphasised that it is vital that the last premises in the UK to have access
to basic and superfast broadband are treated just as well as the first premises and are not
left behind or forgotten.
Source: House of Commons Environment, Food and Rural Affairs (2015) Rural broadband and digital-only
services, Seventh Report of Session 2014–15
. London: The Stationery Office Limited.
1.4 Ten ways to generate ideas for a research topic 9
the film Batman v Superman: Dawn of Justice had taken $424m (£300m) at the box office
worldwide in its first five days, a record for a March debut and the sixth-highest US
opening weekend. This suggests a research topic exploring the enormous amount of
marketing spin-off opportunities presented by such a high-profile brand.
Brainstorming
Brainstorming can be a useful and fun way of generating research topic ideas. It is particularly valuable when you brainstorm with a group of people, ideally those who really understand why you are doing this. You can brainstorm on your own – but it’s much less fun!
To brainstorm, you start off by defining the general field you are interested in. Try to
make this as precise as possible. In the early stages of formulating a topic, you may have
to be pretty imprecise, such as ‘I am interested the effects of the weather on food retailing, but I am not sure how I can turn this into a research topic’.
The next stage is to ask the other members of the group for suggestions, relating to
the imprecise topic you have suggested. It’s a good idea to arm yourself with a pen and a
large sheet of paper and note down all the suggestions you receive. The following five
rules are very important:
1 Do record as many suggestions as possible.
2 Do record all suggestions, however ‘wacky’ or ‘off the wall’ they may appear at first sight.
3 Don’t criticise or evaluate any ideas until they have been considered.
4 Do consider all the suggestions and explore the precise meaning of each of them.
5 Do analyse the list of suggestions and decide which appeal to you most as research
ideas and why.
Definition
brainstorming: a technique that can be used to generate and refine research ideas. It is best undertaken with a group of people.
Concept mapping
Having completed your brainstorming, you may move on to use another technique
which we find very useful in many contexts –
concept mapping. This is a process which
moves from a general idea that may have been the outcome of a brainstorm, or other
idea-generating technique, to the creation of a map which represents visually the
organisation of your thinking. Concept maps may be elaborate or simple; indeed, they
Definition
concept map: a diagram which represents visually the way we organize our thoughts about a set of
related ideas.

10 Chapter 1 Choosing your research topic
may start out very simple and end up quite detailed. The purpose is to help you organize
your thinking about a topic and get them into a coherent state. Concept maps are very
useful in helping you recognise where you have gaps in your knowledge. This can help
with planning further searches in the literature. They are also a good base from which
you can move on to develop research questions and objectives. An example of a concept
map is shown in Figure 1.1.
Competitive
advantage
Product
differentiation
improves improves improves
Technical
superiority
helps
Effective
R&D
Effective
supply chain
Low-cost
production
Efficient
operations
drives
Perceived
quality
Brand
awareness
Strong brand
equity
Brand
loyalty
requires
requires
requires
Product
quality
Service
quality
depends on depends on
requires requires
Figure 1.1 A simple concept map showing representation of competitive advantage
Source: Competitive Advantage Maps, Platts, K. & Khater, M., Working Paper, Centre for Strategy & Performance,
Institute for Manufacturing, 2006. Reproduced with permission from the Institute for Manufacturing.
Making a note of ideas
Many of you will look at this title heading and say ‘that’s all very well for them but I
never get any ideas!’ You do; all of us do. Think positively and forget any notions that
ideas are reserved just for the creative few rather than for the rest of us. It’s perfectly
normal to go through several days without having any ideas at all. Then, perhaps when
you are thinking of something else entirely, an idea will come to you followed by more
ideas. Sometimes, the ideas will come pretty quickly, so you will have to make sure you
capture them before they are forgotten. When you start getting ideas, you will learn not
to worry about the times when the ideas are not coming. It’s just important to take full
advantage of those idea flows when they happen.
So what about capturing those precious ideas when they come? It’s important to
voice-record or note ideas on your mobile phone or keep a notebook for this purpose.
All you need to do is to note any interesting research ideas as you think of them and, of
equal importance, what started you thinking of the idea. You don’t have to go as far as
Mark, who takes his notebook on holiday with him so he can jot down any flashes of
inspiration that occur to him while sitting in a cafe!

1.5  How to refne research topic ideas 11
Discussion with colleagues, friends and lecturers
Don’t forget the more usual sources of ideas and advice: your colleagues, friends and
university lecturers. Project supervisors will often have ideas for possible student projects which they will be pleased to discuss with you. These may be based upon work done
by other students which may be developed, or taken in a different direction. One of our
students conducted an analysis of a food manufacturer’s advertising strategy for launching a new brand with a view to discovering the optimum advertising expenditure level.
We suggested a similar project using an insurance company. Each of the two students
tackled the project in a different way. Both delivered very successful pieces of work.
You can also pick up useful ideas from talking to practitioners and professional
groups. Find out if your local professional association has a students’ evening. Here you
can meet local managers who will be more than willing to discuss ideas with you. Most
professional groups are pleased to welcome students to meetings where often interesting presentations are given which may spark off your ideas. Once again, do bear in
mind the value of discussing possible ideas and make a note of them.
Refining topics given by your employing organisation
The choosing of a research topic may be less problematic than we have suggested above.
At least, on the face of it, this may be so. We say this because you may be given a topic.
This may apply particularly to you if you are a part-time student and your manager has
given you a topic to research. The problem here is that it is not ‘your’ topic, and you
may not be wildly enthusiastic about it. If you find yourself in this situation, you will
have to weigh the advantage of doing something useful to the organisation against the
disadvantage of a potential lack of personal motivation. Aim for a balance.
One of the problems in this situation is that the research project your manager
wishes you to undertake is larger than that which is suitable for your course. This may
not be an insurmountable problem. Within the larger project there may well be a
smaller element which you can concentrate upon for your university research project.
In such cases, it may be possible to complete both by isolating an element of the
larger organisational project that you find interesting and treating this as the project
for your course.
1.5 How to refine research topic ideas
Refining a topic given by an organisation
Peng, a student of Phil’s, worked for a large manufacturing organisation which felt it
had a problem with communicating with its employees. She was asked by the HR manager to carry out an employee survey to establish what employees felt about the way in
Research in practice 1.2

12 Chapter 1 Choosing your research topic
One final point about the potentially tricky job of ‘serving two masters’: the biggest
potential problem may be one of your own making: to promise to deliver research outcomes to your employer and not do so. After all, don’t forget that in this situation, as an
employee, you get paid for what your employer asks of you.
The preliminary study
Even if you have been given a research idea, it is still necessary to refine it in order to
turn it into a research project. This is often called a
preliminary study.
For some research ideas, this study need only be an initial review of some of the
literature. This is similar to ‘using relevant literature’ and ‘following the news media’
in the ‘ten techniques for generating research ideas’ section (Table 1.1). You may
think of this as the first attempt at drafting your critical review of the literature, a
topic we deal with in detail in Chapter 2. Other research ideas may benefit from
revisiting the techniques discussed earlier in this section, in particular looking at
past project titles, using past course assignments and using past projects from
the university library. Discussions with lecturers and managers may also be very
valuable.
which managers communicated with their staff, and the way the company in general
delivered important news to employees. One of the motives behind this managerial initiative was the feeling of the HR director that the company should have a regular newsletter. (Not the least reason for this motive was the thought that this would promote the
profile of the HR function!) This project was not one which Peng would have chosen.
However, she was interested in the theoretical proposition that employees would feel
more committed to the company if they felt trouble was being taken to communicate
effectively with them. Peng talked this over with Phil and, with the approval of the HR
manager, included some questions designed specifically to test this theoretical
proposition.
Peng decided to treat the company research as one major piece of work with a separate element which was ‘hers’ for her university research project. This involved writing a
large report for her employer and a smaller report for her university work.
Peng was very aware of the political dimension to her work. She knew that the HR
director had a vested interest in the outcomes of her work. Phil advised her to keep this
in mind but not let it compromise the element of the work she was doing for her university course. Peng and Phil agreed that it was important to have a clear stance with
regard to her personal objectives, and to stick to them.
Definition
preliminary study: the process by which a research idea is refined in order to turn it into a research
project.

1.5  How to refne research topic ideas 13
Your university may require you to complete a research project which restricts you to
collecting secondary data, data that already exists and was originally collected for some
other purpose (see Chapter 4). If so, it is important that you establish that the data you
require are available.
If your research is going to be conducted in an organisation, it is essential to gain a
good understanding of that organisation. One good way of doing this is to shadow
employees who are likely to be important in your research. They will usually provide
useful insights into the way the organisation works as well as specific guidance for your
research. Whatever way you choose to familiarise yourself with the organisation, don’t
forget that the underlying purpose is to gain a greater understanding so that your
research question can be refined.
At this early stage you will be testing your research ideas. However, it may be that
after a preliminary study, or after discussing your ideas with colleagues, you decide that
the research idea is no longer viable. Don’t despair too much if this is the case. It is
much better to revise your research ideas at this stage than to have to do it later, when
you have undertaken far more work.
The Delphi technique
Another approach you may use for refining your research ideas is the Delphi technique.
This technique involves using a group of people who are either involved or interested in
the research idea to generate and choose a more specific research idea. To use this technique, you need to:
1 explain your research idea to the members of the group (they can make notes if they
wish);
2 encourage group members at the end of your explanation to seek clarification and
more information as appropriate;
3 ask each member of the group, including the originator of the research idea, to generate independently up to three specific research ideas based on the idea that has been
described (they can also be asked to provide a justification for their specific ideas);
4 collect the research ideas in an unedited and non-attributable form and to distribute
them to all members of the group;
5 conduct a second cycle of the process (steps 2 to 4) in which individuals comment
on the research ideas and revise their own contributions in the light of what others
have said;
6 undertake subsequent cycles of the process until a consensus is reached. These
either follow a similar pattern (steps 2 to 4) or use discussion, voting or some other
method.
Definition
Delphi technique: a technique which can be used with a group of people who are either involved or
interested in the research topic to generate and select a more specific research idea.

14 Chapter 1 Choosing your research topic
Using a Delphi Group
Kumar worked part time at The Sizzling Wok, a restaurant specialising in Asian fusion
cooking. The owner of the restaurant, Virat, explained to him that while business in
general was good, it was important that the restaurant generated more turnover in the
traditionally slacker periods, i.e. during the day and on evenings in the early part of the
week.
Kumar thought that a research project involving some consumer research and business costings would be useful for Virat and provide a project for his university course
which would involve marketing, accounting and business planning.
Kumar was part of a small group of course colleagues which met frequently to discuss
ideas and help each other with the more challenging parts of the study programme. He
explained his research idea to the group members. As a result the group generated a
series of research ideas for Kumar to explore. These were:
Having ‘special offers or special menus’ at ‘quiet’ times, meaning that customers
could eat more cheaply at these times.
A loyalty scheme with points awarded to customers leading to free meals and extra
points awarded at ‘quiet’ times.
Promotions including features such as ‘show cooking’, where the chefs prepared special dishes responding to customers’ wishes in the restaurant itself, thus generating
interest and a sense of ‘bonding’ with the customer.
Restricting the opening times of the restaurant to avoid opening for business at
‘quiet’ times.
Examining the costs of opening at ‘quiet’ times (e.g. staffing costs) with a view to
making these times more profitable.
Following on from the list of general ideas, the group then developed some more specific research ideas, among which were the following:
how the restaurant may generate more business during the ‘quiet’ times of the week,
and
examination of the profitability of operating during the ‘quiet’ times of the week.
At the end of this process, Kumar was able to refine his thinking and decide upon a
research idea with which Virat was pleased and his research supervisor was impressed.
It was: ‘How can The Sizzling Wok operate during the less busy times of the week in
order to meet the demand of maintaining a consistent level of profitability at all trading
times?’
Kumar was delighted with the outcome of the meetings and resolved to help his colleagues with the same level of commitment with which they had helped him.
Research in practice 1.3
People generally enjoy trying to help one another, so this technique can be very effective. It is also very good for forging team spirit among groups of people. An example of
the developing of a research question is shown in Research in practice 1.3.

1.6 What makes a good research topic? 15
Narrowing down
The Delphi technique illustrates what is perhaps the most difficult aspect of refining
your research ideas: the move from the general to the specific. Many of us start out with
some idea of what we want to do for our research but find it tricky to narrow down.
One useful way of moving from the general to the specific is to think of research
ideas as developing through a process which goes through three stages of idea classification. These are:
1 The general area.
2 The more specific field.
3 The precise focus of the research.
One of our students recently was following the marketing route through his degree. He
was also very interested in environmental issues. These can be termed his general area
of interest. More specifically, he was also very enthusiastic about exploring the different
ways in which marketeers can learn about ‘greener’ ways of conducting marketing for
their organisations. After much thinking and discussion, he developed the research
questions: ‘What methods do marketeers use to learn about “greener” ways of conducting marketing for their organisations?’ and ‘Why do they decide upon the methods
chosen?’
The next section draws together some of the thoughts in the section ‘Why choosing the
right research topic is so important’ as well as introducing ideas about what goes into
making a good research topic choice. Some of the most significant of these ideas are
listed below.
It is a topic about which you are enthusiastic and which matches
your career goals
Let’s start with by repeating two of the points we made earlier. A good research topic is
one in which you have a genuine interest because, as we said earlier, you have to live
with it for possibly a few months or longer and you want to give yourself the best chance
of success. A well-chosen topic is also one that matches your career goals.
There are resources available, particularly data and time
We also referred to the question of resources, in particular those of your skills and
knowledge and individuals, such as lecturers, who may be able to help you. But there
are two other key resources to think about. The first of these is data. It is all very well
thinking up and precisely defining an intriguing research idea, but it’s of little use if you
cannot get access to the data to explore it. You may, for example, be very interested in
1.6 What makes a good research topic?
16 Chapter 1 Choosing your research topic
the effect on patients of specific management initiatives in the UK National Health
Service. However, getting primary data from patients may be far from easy for all but
researchers appointed by the Service. Secondary data may be available, but will it
answer precisely the research question you have set?
The second question to ask yourself is: ‘Is the research topic achievable within the
available time?’ This may be quite a difficult question to answer at the beginning of
your research project because you cannot anticipate the potential delays. If you are
dependent on data from other people such as managers, customers or colleagues, it is
reasonable to say that this will not be as high a priority for them as it is for you. It is
always a good idea to be realistic (or even pessimistic!) in your assessment of the timescale necessary to gather data. Even the collection of secondary data may be less
straightforward than you envisage, particularly that from non-official sources. The
more realistic you are in setting your timescales at the beginning, the less stressful you
will find the latter stages of the research process.
The subject is topical
In the winter of 2015–16, Apple was locked in a bitter dispute with the FBI over the latter’s
desire to compel Apple to access the information on the iPhone of a mass killer. Apple
obviously expressed the deep concern felt by most people about the reasons for the FBI
request but refused, arguing that if the US government could make it easier to unlock an
iPhone, it would have the power to reach into anyone’s device to capture their data. The
government could then extend this breach of privacy and demand that Apple build surveillance software to intercept an individual’s messages, health records or financial data,
location, or even access the phone’s microphone or camera without the user’s knowledge.
At the time of writing this dispute remained unresolved. But it is likely to be repeated
in other cases, particularly as other technology organisations such as Amazon and
Google came out in support of Apple.
This raises an interesting question about the ownership of digital data and the extent
to which the privacy, which is at the heart of our relationship with many of the digital
companies with which we deal, may be compromised in particular circumstances. This
may be the basis of a marketing investigation about the degree to which technology
companies can stress data privacy as part of their market appeal.
The Apple–FBI case may have been the catalyst for this project and serve as a useful
example from which another research project may be developed.
Whatever the outcome, you still have a worthwhile project
Let’s say that you develop a research question such as ‘How do severe winter weather
conditions affect retail food supermarket trading patterns?’ On the face of it, this
sounds a reasonable question. It could be expected that there would be an effect, not
only on the volume of trade but on the type of food purchased. However, it is possible
that the answer may be ‘there is no effect’. The consequence of the ‘no effect’ result for
you is not only that it is rather dispiriting, but you don’t have an interesting story to tell
in your project report.

1.6 What makes a good research topic? 17
The trick is to define a research question that gives you results of similar value whatever you find out. In the example here, ‘To what extent does the weather affect retail
food supermarket trading patterns and why?’ is a much wider question that will guarantee that you have the scope to write an interesting project report, whatever the
outcome.
The topic fits the specifications and will meet the assessment
criteria set by the university
Your lecturers are likely to issue a set of guidelines to help you with completion of your
research project. This will give you general ideas of what constitutes a successful project. A good place to start in your study of this is the assessment criteria. It’s a good idea
to check through these criteria to make sure that your choice of research topic will enable you to meet those criteria. So, for example, if there is a criterion that your research
must aim to solve an organisational problem, it’s of little use pursuing a study that analyses consumer expenditure data over a given time period. Similarly, if the assessment
criteria state that you should only use secondary data, you should not design a questionnaire or interview people.
It is taken for granted that the chosen research topic should match the content of
the course being undertaken. Yet we have seen a number of cases where this is questionable. An interest in the under-representation of women in management, for example,
could easily develop into a study of the role of women in society in general, rather than
in organisations specifically. This is an easy trap to fall into. One of the problems with a
study chosen by you, and pursued by you, is that you become so close to it that you lose
sight of the fact that you are wandering away from the initial brief. That is why talking
to people is such an important part of the process.
As with all concerns you may have about the progress of your research, talk to your
lecturer if you are not sure about the appropriateness of your topic choice.
There is a clear link to the relevant literature
It is almost certain that your assessment criteria will specify the need to link your work
to the relevant academic literature. It is essential here to ask yourself: ‘Why am I doing
this?’ Too often the literature chapter in project reports we read is of the ‘he said this,
she said that and they said the other’ variety (section 2.2). You should not use the literature in this way. Rather, an alternative way of thinking about the review of the literature
is to call it a ‘review of previous research on my topic’. From this you can extend this
work. This can be done, for example, by setting your study in the context of the literature reviewed. A student of Mark’s, for example, reviewed studies of workers who have
some autonomy over the way in which their work is organised and applied this to her
project on the work organisation of courier drivers. As well as locating your study in a
different work context to that which has been covered in your literature review, you can
set it in a different time context or national culture. Many of the classic management
theories are several decades old. You may well ask the question: ‘Does this still apply in
the twenty-first-century post-industrial age?’

18 Chapter 1 Choosing your research topic
Fresh insights into the topic are provided
We hope it will have occurred to you that the previous points about the extension of
your literature search will guarantee that your research will be able to provide fresh
insights into your chosen topic. A fresh context leads to fresh insights. What is less easy
to guarantee is the degree to which your project report’s discussion conclusions are
insightful.
Research question(s) and objectives are capable of being
stated clearly
Perhaps the best test of the clarity of your research question(s) and objectives is the
extent to which those who know nothing about your topic can understand what it is
you plan to do. If you pass this test, and the others listed in Table 1.4, you are well on
your way to developing a successful research topic. We say more about writing clear
research questions and objectives in the next section.
Table 1.4 Summary: what makes a good research topic?
•  It is a topic about which you are enthusiastic and it matches the career goals.
•  There are resources available, particularly data and time.
•  The subject is topical.
•  Whatever the outcome, you still have a worthwhile project.
•    The topic fits the specifications and meets the standards set by the examining institution.
•  There is a clear link to the relevant literature.
•  Fresh insights into the topic are provided.
•  The research question(s) and objectives are capable of being stated clearly.
Settling on a suitable research idea is a great step forward in the research process. It is
challenging and time consuming. But the early stages are not complete yet. The task of
turning your big idea into precise and meaningful research question(s) and objectives is
vital if you are to end up with reliable and valid data from which you can draw valid
conclusions.
Do you need research questions and objectives? Well, we think that you do. We see
them as a progression: from research questions, you then move on to define objectives.
Perhaps this is clearer if we define what we mean by these two key terms. Research questions are those questions that the research process will address. They are often the forerunner of research objectives. On the other hand, research objectives are clear, precise
statements that identify what you wish to accomplish as a result of doing the research.
And research objectives are vital. Why? You’ve guessed it – because if you don’t know
where you’re going, you’re likely to end up somewhere else!
1.7 How to turn a research idea into a research project
1.7 How to turn a research idea into a research project 19
Developing research questions
Rather like generating research ideas, defining research questions is not a simple, straightforward matter. Here again, we come up against what is a suitable research question. For
the answer to this, we go back to the last section, ‘What makes a good research topic?’ A
suitable research question is one that reflects the that fact you have thought about what
fits the specifications and meets the standards set by the examining institution, provides a clear link to the relevant literature and promises fresh insights into the topic you
have chosen. Let us concentrate on that word ‘insights’. Here we raise the point about
two types of questions – those questions that are too easy and those that are likely to
provide insights. An example will illustrate the difference. Let us say that your research
idea involves examining the impact that Internet banking has had upon local bank
branch closure. Asking the question ‘How many local bank branch closures have there
been in the period since Internet banking became widespread in my country?’ would be
too simple. It would merely describe a situation and not provide insights into the link
between the growth of Internet banking and local branch closures. Questions that
would promise more insights would be ‘What effect has the growth of Internet banking
had upon the uses customers make of branch facilities?’ and ‘Why do customers prefer
to bank online in preference to visiting their local branch?’ The answers to these questions may give some clue as to the eventual effect on branch closures of Internet banking. If, for example, you found that the overwhelming response to the question ‘Do you
still value your local branch?’ put to Internet banking users was negative, then the obvious conclusion would be that the future for local bank branches is bleak.
Avoiding the asking of questions that promote descriptive answers, because they are
too easy, is one way of ensuring that you do not ask unsuitable research questions.
Another is asking questions that are too difficult. Earlier in this chapter, we cited the
example of a research interest that many researchers have had in recent years in the
under-representation of women in management. There are several ways you may be able
to gain insights into the research question ‘Why are women under-represented in
senior management positions?’ but it is a difficult question to answer definitively. In
this case, as with most research questions, there is a simple response to the question of
how to find out the answer: ask those who select the appointees to senior management
positions. But the reality here is that it would probably be very difficult, if not impossible, to gain access to those key decision makers in large organisations. Without such
access it would not be possible to get a good understanding of the subtle ‘unofficial’
processes that go on at staff selection which may favour one type of candidate over
another. Overreaching yourself in the definition of research questions is just as big a
danger as asking simple, descriptive questions.
To end up with a research question that is suitable, try subjecting it to the ‘Goldilocks
test’. This is what Clough and Nutbrown (2012) use to decide if research questions are
either ‘too big’, ‘too small’, ‘too hot’ or ‘just right’. By ‘too big’ they meant those that
Definition
research question: the one overall question or a number of key questions that the research process
will address. These are often the precursor of research objectives.

20 Chapter 1 Choosing your research topic
probably need significant research funding because they need too many resources.
Questions that are too small are like those that we mentioned earlier – the ones that just
promote descriptive answers. The ‘too hot’ questions are so called because they stray
into sensitive areas, rather like our example of asking senior managers why they tend
not to recruit women to senior management positions. There are other reasons why the
context may be too sensitive. Many of these are likely to involve issues of status and
power among subjects whom you wish to play a part in your research. Alternatively, the
timing of the research may be inappropriate. One of our students was conducting
research on redundancy. His requests to organisations were often met with refusals
because the organisations that refused to participate were going through difficult times.
Outside people talking about redundancies would have sparked off suspicion among
employees that these organisations were planning redundancies. On the other hand,
research questions that are ‘just right’ (Clough and Nutbrown, 2012: 43) are those that
are ‘just right for investigation at this time, by this researcher in this setting’.
Defining the research question
Rebecca was studying for a BSc in Management Studies and taking her placement year
in Forsure, a multinational insurance company. The company’s overall mission was: ‘to
help our customers live their lives with more peace of mind by protecting them, their
relatives and their property against risks and by managing their savings and assets’. One
of the company’s strategic objectives was:
During her degree Rebecca had become particularly interested in corporate strategy.
She was particularly interested in the way in which the often rather ambitious mission
statements were put into practice by organisations.
Rebecca was rather concerned that the strategic objective of her placement company
quoted above contained two areas of her course-customer focussing and employee
involvement- neither of which she had studied in depth on her course.
She confessed her concern to her research supervisor, Dr Hunt, who suggested that it
might be useful for her to concentrate upon the process of delivering the strategic objective rather than the objective itself. This would lend the project an element of generalisability which would not be the case were it to focus on either customer focussing and
employee involvement.
Together with Dr Hunt, Rebecca developed the following research question:
Research in practice 1.4
To become the preferred company for all our stakeholders by strengthening our focus on
the customer and fostering employee involvement through building a culture of trust
and achievement.
How do the procedures and processes the senior management of Forsure use to cascade
downwards support their strategic objective of becoming the preferred company for all
their stakeholders by strengthening the focus on the customer and fostering employee
involvement through building a culture of trust and achievement?

1.7 How to turn a research idea into a research project 21
Another useful metaphor to help you think about clarifying your research question
is Clough and Nutbrown’s (2012) Russian doll principle. Here your research can be broken down from the original statement to something which strips away all the complicated layers and obscurities until the heart of the question can be expressed, in the
same way as the Russian doll is taken apart to reveal a tiny doll at the centre. Table 1.5
has some examples of research ideas and the general focus research questions they
suggest.
Table 1.5 Examples of research ideas and the focus research questions they suggest
Research idea General focus research questions
1 The marketing of security in credit cards To what extent does a credit card company
market the measures it takes to ensure
consumer security in order to gain competitive
advantage?
2 Organisations’ employee newsletters How effective are organisations’ newsletters at
gaining employee identification with the
organisation in geographically diverse
organisation structures, and why?
3 The use of shelf display point-of-sale
material in retail supermarkets
How does the use of shelf display point-ofsale material in retail supermarkets affect
buyer behaviour?
4 Sustainable accountancy To what extent are organisations ensuring that
environmental and social performance is
better connected with strategy and financial
performance, and why?
In the examples in Research in practice 1.4 and Table 1.5, there is one general focus
research question generated from the central research idea. Of course, this may lead to
several more detailed questions or the definition of research objectives. In the next section, we deal with the development of research objectives.
Developing research objectives
Having written your research questions, you may be asking yourself why it’s necessary
to develop
research objectives. The answer is that it may not be. Your research project
assessment brief may state that it’s acceptable to write research questions or research
objectives. In our view, it’s better to do both. Why? Because we think that research
objectives add an element of precision to research questions. In Table 1.6 we illustrate
Definition
research objectives: clear, specific statements that identify what the research process seeks to
achieve as a result of doing the research.

22 Chapter 1 Choosing your research topic
this point by showing research objectives alongside the research questions from
which they have been developed. In general, research objectives tend to be more
acceptable to the research community. Research in practice 1.5 is an example of how an
official research report prepared by the UK government expresses an overall research
aim and then expands this into specific objectives.
Developing research objectives from an overall research aim
The UK Department for Work and Pensions (DWP) commissioned a leading public
affairs research agency to explore how employers approach the recruitment and hiring
process for unskilled and semi-skilled workers. This included the factors that influence
recruitment decisions and how hiring decisions are made in practice. The purpose of the
research was to support Jobcentre Plus* customers seeking work and further inform the
Department’s employer engagement practices.
A qualitative research study was carried out between May and July 2013 based on
focus groups with small employers reflecting a range of industry sectors, and a small
number of employer case studies which provided the opportunity to observe any differences in how employers describe their recruitment process and what they do in practice.
Research aim
The DWP wished to explore how employers approach the recruitment and hiring process, what influences recruitment decisions, and to understand how hiring decisions are
made in practice.
Research in practice 1.5
Table 1.6 Phrasing research questions as research objectives
Research question Research objective
1 Why have organisations introduced
employee communication schemes?
1 To identify organisations’ objectives for
employee communication schemes.
2 How can the effectiveness of employee
communication schemes be measured?
2 To establish suitable effectiveness criteria
for employee communication schemes.
3 Has employee communication been
effective?
3 To assess the extent to which the
effectiveness criteria for employee
communication have been met in
published studies.
4 How can the effectiveness of employee
communication be explained?
4a To determine the factors associated with
the effectiveness criteria for employee
communication schemes being met.
4b To estimate whether some of those factors
are more influential than other factors.
5 Can the explanation be generalised? 5 To develop an explanatory theory that
associates certain factors with the
effectiveness of employee communication
schemes.

1.7 How to turn a research idea into a research project 23
Table 1.6 comes from research we did on an organisation’s employee communication scheme. This was the forerunner of research on the scheme of a particular organisation who wanted to establish if their scheme was ‘working’. This prompted us to
explore a highly debatable question to which the literature offered no satisfactory
answers.
Writing research question 1 as an objective prompted a consideration of the objectives of the organisations. This was useful because it led to the finding that there often
were no clear objectives. This in itself was an interesting theoretical discovery. (How
often management initiatives are introduced with no clear objectives in mind!)
Objectives 2 and 3 operationalise the matching research questions by introducing
the notion of explicit effectiveness criteria. This is particularly important if measurement of success is to be attempted. Similarly, objectives 4a, 4b and 5 introduce an element of precision about factors that lead to effectiveness in question 4. The most
significant difference between these questions and objectives is illustrated by the way
in which question 5 becomes objective 5. Although similar, they differ in the way that
the objective makes clear that a theory will be developed that will make a causal link
between two sets of variables: effectiveness factors and employee communication
scheme success.
You may find it easier to write specific research questions than objectives. But, as
Table 1.6 demonstrates, research objectives add a further level of precision. Our view is
With this key aim, research was designed to consider how DWP might firstly, support
its customers who are seeking work and secondly, further inform the Department’s
employer engagement practices.
Within this overarching aim, there were four key objectives:
to explore how employers approach the recruitment process, including:
– the sources they draw on to identify potential candidates;
– the recruitment processes and practices they use;
– the candidate features that influence an employer’s decision about whether to
employ; and
– the rationality of the recruitment process and the circumstances in which employers change their practices, either explicitly or implicitly;
the effect of labour market and sector requirements on the recruitment process;
experiences and views about Jobcentre Plus and third party agencies as recruitment
organisations; and
the impact of government initiatives on recruitment decision-making.
* Job Centre Plus is a DWP agency providing services to those attempting to find employment.
Source: Department for Work and Pensions (2014) Small Employer Recruitment Practices: Qualitative Research
into How Small and Medium-Sized Enterprises Select Candidates for Employment.
Report No 855, July 2014.
24 Chapter 1 Choosing your research topic
that research objectives require more rigorous thinking, which derives from the use of
more formal language.
We make one final point on the development of your research objectives. Think
about adding personal objectives to those specific research objectives. These may be
concerned with your specific learning objectives from your research or more general
personal objectives such as enhancing your career prospects. You will be familiar with
the SMART test, listed in Table 1.7. Try this when you have thought about your personal
research objectives.
Table 1.7 The SMART test
Check that the objectives are:

Specific What precisely do you hope to achieve from undertaking the research?
Measurable What measures will you use to determine whether you have achieved your
objectives?
Are the targets you have set for yourself achievable given all the possible constraints?
Achievable
Realistic Given all the other demands upon your time, will you have the time and energy to
complete the research on time?
Will you have time to accomplish all your objectives in the time frame you have set?
Timely

The part that theory plays in developing research questions
and objectives
There is no more misunderstood word in the research vocabulary than ‘theory’. In this
section, we clarify what
theory is by explaining what it is not. It’s important that you get
this clear. In Chapter 5, we explain the importance of theory in research design. We
emphasise that the role of theory will play a key part in your study, as all projects will
need to link to theory in some way. The most likely link will be to an existing theory
explained in the literature relevant to your research topic.
Theory may be broadly defined as an explanation of the relationship between two or
more concepts or variables. So, for example, the fact that you are reading this text is
based (we hope!) on the theory that the outcome will be that you are better equipped to
pursue your study.
We said we would clarify what theory is by explaining what it is not. Sutton and Staw
(1995) help here and, although their article was written in the last century, it is still
really helpful! In their view, theory is not:
lots of data;
lists of variables;
Definition
theory: an explanation of the relationship between two or more concepts or variables.
1.7 How to turn a research idea into a research project 25
numerous hypotheses;
pages of references;
frequent use of diagrams.
Theory is not lots of data
We warned earlier in this chapter about developing research questions that are too
easy. They often allow only for description of phenomena. Such questions lead to project reports which are full of
data and little else. Data only describe what you have
found out and, possibly, report on the patterns observed: theory explains the relationships between these patterns. As Sutton and Staw (1995: 372) say, ‘data do not generate
theory – only researchers do that’.
Theory is not lists of variables
You may be tempted to list a series of variables in an attempt to explain the determinants of a given process or outcome. But simply listing variables which may predict an
outcome is insufficient. For example, in a study of effective management decisionmaking, merely listing those factors which seem to be associated with effectiveness is
insufficient. What is required for theory development is an explanation of why the
factors associated with effectiveness are likely to be strong predictors of effectiveness.
Theory is not numerous hypotheses
In section 5.5 we explain the role of hypotheses in a research design, and in section 7.3
we talk about
hypothesis testing. Testing possible relationships between variables certainly has a part to play in setting up research projects. But they do not clarify why
phenomena have happened and therefore do not constitute theory.
Definitions
data: facts, opinions and statistics that have been collected together and recorded for reference or
for analysis.
variable: individual element or attribute upon which data have been collected.
hypothesis: tentative (usually testable) statement about the ‘relationship’ between two or more
variables.
Theory is not pages of references
‘How many references should I include?’ is a question you have probably been asked by
student colleagues; you may have asked it yourself. It’s as if the more references you
include, the more ‘theoretical’ your work will be. OK, so it may (only may) indicate that
you have done lots of reading, but that isn’t theory. Theory is present when you outline
why the things you describe occur.
26 Chapter 1 Choosing your research topic
Theory is not frequent use of diagrams
Diagrams, boxes and arrows may help in clarifying patterns and causal relationships
between variables, but they don’t usually explain why the relationships have
occurred.
So, to sum up, the key word to stress in establishing the presence of theory in your
project report is ‘why’. Why do certain things happen when particular variables are present and others absent? Why does A happen and B does not? Why do people behave in
this way rather than the opposite way?
We made the point earlier that the fact that you are reading this text is based on the
theory that it better equips you to complete successfully your research study. In that
case, every purposive decision we take is based on theory: that certain consequences
will flow from the decision. So, if your decision to read this text is theory-based, then so
is, for example, every manager’s meeting that is devoted to making decisions. Yet managers are unlikely to realise this. They are equally unlikely to want to acknowledge this!
After all, theory is something that belongs in the university; it is not something which
is part of the ‘real world’. Yet, as Kurt Lewin (1945: 129) said, ‘there is nothing as practical as a good theory’.
So, if theory plays an important part in our everyday lives and is implicit in the
decision-making processes which go on, it’s not something that you should be too
scared about. What is important is that the element of theory in your research is made
explicit, not left implicit.
How ‘grand’ does my theory need to be?
Still feeling a bit scared about the idea of theory development? Well, let’s try to put your
mind at rest by looking at a three-type categorisation of theory developed by Creswell
(2008). Creswell talks of ‘grand theories’, ‘middle-range theories’ and ‘substantive theories’ (Figure 1.2).
Grand
theories
Middle-range theories
Increasing restrictions
in terms of general
applicability
Increasing capacity
to change the way
we think about
the world
Substantive theories
Figure 1.2 Grand, middle-range and substantive theories
Source: Saunders et al., 2016; developed from Creswell, 2008.
1.7 How to turn a research idea into a research project 27
Grand theories are those that are so significant that they make a major contribution
to the way in which we understand the world around us. An example of one such grand
theory is Darwin’s theory of evolution. Don’t worry; there is no need for you to think
about following in Darwin’s footsteps! Neither is there the need for you to be so ambitious as thinking of
middle-range theories. These are not as world-changing in their
scope as grand theories, yet they are clearly of great importance in their particular field
of study. You will have studied many of these in your course. Examples are those concerned with motivation, leadership or creative thinking. Of course, you may apply one
of these to your particular situation. But you wouldn’t be expected to come up with an
original theory of such significance, which could be generalised to many other
contexts.
However, most of us are concerned with developing rather more modest theories –
substantive theories. These are restricted to a specific context: a particular time, research
setting, population or problem. So you may be concerned with a study of the reason
why, for example, the introduction of a new computer system in a particular organisation proved to be a failure. The conclusion you draw would be prefaced by the word
‘because’ – a term which serves as a ‘trigger word’ for a theoretical statement (e.g. ‘it
failed because there was too little of A and not enough B and C’. Alternatively, ‘it failed
because management did not explain the reason for the introduction to staff concerned
which led to staff discontent’. Either of these theoretical explanations may stand alone,
or you may link them to well-established middle-range theories developed by other
researchers.
Substantive theories may be more modest, but they may lead to ‘middle-range theories’. And by developing substantive theories, we are doing our bit as researchers to
develop our understanding of the world about us.
Definitions
grand theories: theories that make a major contribution to the way in which we understand the
world around us.
middle-range theories: theories which are not as world-changing in their scope as grand theories
but are of great importance in their particular field of study.
substantive theories: theories which are restricted to a specific context: a particular time, research
setting, population or problem
Writing a research question based on a middle-range theory
Aafreen was a part-time management student at her local university. Her full-time job
was a practice manager at a large city health centre in which she was responsible for a
team of over 30 administrative personnel whose main duties included reception work,
patient record-keeping, financial control, stock control, prescription monitoring and
general clerical duties.
Research in practice 1.6

28 Chapter 1 Choosing your research topic
In most respects, the local Health Service Trust and general practitioner doctors in the
practice were happy with the operation of the practice, including the way it was administered by Aafreen and her team. However, a recent customer satisfaction survey concluded that the way in which the practice’s performance was perceived by its clients was
not as good as it could have been.
In management meetings this survey conclusion was discussed at some length, and a
number of suggestions were made about how things could be improved. One central
theme ran through those suggestions: that the practice needed to improve its rather
dreary and pedestrian image. The management team, including, Aafreen, decided that
the way in which all staff members, including the professional medical staff, presented
themselves needed to improve.
Consequently, Aafreen was charged with the responsibility to implement a staff selfpresentation programme with the intention that all staff should present an image consistent with a modern, professional health centre. She knew that this would have to be
handled sensitively because a number of staff members, although excellent performers
in their jobs, presented a somewhat dowdy image.
Aafreen had studied motivation theory in the early part of her course and guessed
that the sure way for the initiative to fail was to impose it upon staff. She felt that the
route to programme success was to use empowerment theory, giving the ownership of
the programme to the staff members themselves, starting with a meeting where ideas
could be invited.
Aafreen decided that she would implement her investigation and keep detailed notes
of every aspect of it, and then write it up as her course project.
Aafreen felt that an initial research question would help her focus on both the practical project to be undertaken at the practice and her course project. After some thought,
and trial and error, she wrote the following research question:
‘How should a staff self-presentation programme be implemented effectively using
the principles of empowerment theory?’
Choosing the right research topic is important because: you have to live with it for a
considerable period of time; it should exploit and develop your knowledge and skills,
and enable you to draw upon existing resources and help you to pass the course you
are studying.
Choosing an appropriate research topic is difficult because: there is too much choice;
there are fears that it will be too difficult and insufficiently theoretical and there is a
temptation to choose a topic that you have done before for another purpose resulting in repetition.
Ten techniques for generating ideas for research topics are: thinking; looking at past
project titles; using past course assignments; using past projects from the university
library; using relevant literature; following the news media; brainstorming; concept
mapping; making a note of ideas and discussion with helpers.
Summary
Thinking about your research topic 29
Pursuing a research topic assigned to you by your employer poses its own problems.
In such cases, you should aim for a balance between the competing demands of your
employer and your university by, for example, isolating an element of the larger
organisational project that you find interesting and treating this as the project for
your course.
Research topic ideas may be refined by the conduct of a preliminary study; application of the Delphi technique and a process of narrowing down.
A good research topic is one: that you are enthusiastic about and it matches your
career goals; for which resources are available; that is topical; that, whatever the outcome you still have a worthwhile project; where the topic fits the specifications and
meets the standards set by the examining institution; that links to the relevant literature; that provides fresh insights into the topic and for which research question(s)
and objectives are capable of being stated clearly.
The next stage from choosing the research topic is the development of research questions and objectives. A suitable research question is one that reflects the that fact you
have thought about what fits the specifications and meet the standards set by the
examining institution; provides a clear link to the relevant literature and promises
fresh insights into the topic you have chosen. Research objectives add an element of
precision to research questions.
All projects will need to link to theory in some way. The development of a ‘grand theory’ or ‘middle-range’ is unnecessary; ‘substantive theory’ development is sufficient.
Look again at the 10 techniques available for generating research ideas (Table 1.1).
Choose those that appeal to you most. Use these to try to generate a research idea or
ideas. Once you have some research ideas, or if you have been unable to find an idea,
talk to your project supervisor.
Evaluate your research ideas against the checklist ‘what makes a good research topic?’
(Table 1.4).
Choose a research report contained in an article in an academic journal relevant to
your topic. Revisit the useful questions to ask when searching articles and reports for
possible research topic ideas (Table 1.3) and apply these to the article.
Write a general focus research question related to your research topic. Try to ensure
this is ‘Why?’ or ‘How?’ rather than a ‘What?’ question.
Use the general focus research question to write more detailed research questions and
your research objectives.
Go back to the research report contained in an article in an academic journal relevant to your topic that you used with the questions in Table 1.3. Assuming that the
research question and objectives are not made explicit, try to establish from the content of the article what the research question and objectives may have been.
Thinking about your research topic
30 Chapter 1 Choosing your research topic
Creswell, J. (2008).
Qualitative, Quantitative, and Mixed Methods Approaches (3rd ed.).
Thousand Oaks, CA: Sage.
Clough, P. and Nutbrown, C. (2012).
A Student’s Guide to Methodology (3rd ed.). London: Sage.
Department for Work and Pensions. (2014).
Small Employer Recruitment Practices: Qualitative
Research into How Small and Medium-Sized Enterprises Select Candidates for Employment.
Report No 855, July 2014.
House of Commons Environment, Food and Rural Affairs. (2015).
Rural Broadband and DigitalOnly Services, Seventh Report of Session 2014–15. London: The Stationery Office Limited.
Lewin, K. (1945). The Research Centre for Group Dynamics at Massachusetts Institute
of Technology
. Sociometry, 8(2), 126–36.
Raimond, P. (1993).
Management Projects. London: Chapman & Hall.
Saunders, M., Lewis, P. and Thornhill, A. (2016).
Research Methods for Business Students
(7th ed.). Harlow: Pearson Education Ltd.
Sutton, R. and Staw, B. (1995). What theory is not.
Administrative Science Quarterly,
40(3), 371–84.
References
Chapter 2
Reviewing the literature critically
Since you started your programme, your lecturers have expected you to read textbooks
and academic journal articles as part of your studies. Sometimes your lecturers will have
told you exactly which book chapters and articles they want you to read. At other times,
such as when undertaking an assignment, you will have been expected to find at least
some books and articles yourself, deciding whether or not they were relevant. This will
have meant you searching in your university’s library catalogue to see if there were any
potentially useful books, as well as searching one or more of your university’s online databases of academic journal articles. You will probably also have searched the Internet for
relevant reports and articles using a general search engine such as Google or Bing. Having
found some books and articles, you would then have read them quickly to work out
whether or not they were relevant before reading them more carefully and using them in
your assignment. Well done! You have already started to develop some of the skills you
need to review the literature critically.
The overall purpose of this chapter is to help you further develop the skills and understanding you need to critically review the literature so that you can complete your research
project. The skills you develop will also be helpful for your assignments. We explore the
skills by looking at critically reviewing the literature as an iterative process consisting of four
activities which you will need to go through a number of times to develop your literature
review chapter (Figure 2.1).
However, before we do this, it is important that you understand what a critical literature
review is, why you need to review the literature for your research project and are aware of
the variety of literature available to you.
In this chapter, we therefore start by talking about what a critical literature review is and
its purpose. We then go on to explain why it is important that you review the literature
critically as part of your research project and describe the variety of literature available to
you. Next, we consider the skills and understanding you will need to conduct a critical literature review. First, we look at how you will search for and obtain the literature using your
2.1 Why you should read this chapter
32 Chapter 2 Reviewing the literature critically
university’s online databases of academic articles. As part of this we also consider briefly
your use of the general search engine Google and the online encyclopaedia Wikipedia.
Next, we talk about how you can evaluate the usefulness of the literature you have found
for your own research project. We then explain how to read, note and correctly reference
this useful literature, emphasising the importance of not passing off others’ work as your
own by mistake and so being accused of plagiarism. We end this chapter by talking about
how you can structure and draft your literature review.
Evaluate usefulness
to your research
Draft your
literature review
Search for and
obtain literature
Read, note and
correctly reference
useful literature
Figure 2.1 The process of critically reviewing the literature
You will find that, although there are numerous definitions of a critical literature review,
these usually emphasise four aspects. In particular, it:
offers an overview of significant literature available in your chosen topic;
includes relevant items such as academic journal articles, books and other sources;
provides a discussion and critical evaluation of these, the level of detail reflecting the
significance of each item to your research questions;
develops a clear argument to contextualise and justify your research.
This means your literature review is not like a book review, or series of book reviews,
simply describing and summarising what each book or article is about. First, it includes
only significant literature published on your research topic (Table 2.1. You will therefore have to decide what is significant to your research topic and why it is significant.
This will involve you in assessing how relevant, if at all, each item you read is to your
2.2 What a critical literature review is
Definition
(critical) literature review: a detailed overview of the significant literature available about your chosen topic, providing a discussion and critical evaluation, and using clear argument to contextualise
and justify your research.

2.2 What a critical literature review is 33
research and, on the basis of your assessment, deciding whether or not to include it. We
talk about this more in section 2.6. For those items (mainly articles and books) you
decide to include, you then need to ensure that your review is more than just a description of their contents. You will need to discuss those aspects of what has been said that
are relevant to your research, and within your discussion you must be critical. We would
be surprised if this was the first time you had heard the term ‘critical’. It, or something
similar such as the phrase ‘critically evaluate’, is often used in assignments and exam
questions. However, it is worth pausing for a moment to explain what is meant by ‘critical’ in relation to your literature review. This word is crucial as it summarises what you
must do when writing your literature review.
Within your literature review, being critical simply means you provide clearly justified, reasoned judgements about what you have read and are now writing about. To do
this, you need to have a good knowledge of the literature on your topic that both supports and opposes your ideas, and show through your writing that you have this knowledge. This is often referred to as ‘demonstrating familiarity with the topic’ in assessment
criteria and means that you also need to show you know whom the recognised experts
in your topic area are. When you read and note the items that relate to your research
topic, you will need to do so with some scepticism, questioning what you read. If you
think an author’s arguments, ideas or research findings are unclear, biased, poorly justified, inconsistent or need to be tested further, this may well mean that they are.
However, it is not enough for you to just write that the author’s arguments are unclear,
biased, poorly justified, inconsistent or need to be tested further. In your critical literature review, you must also include clear reasons why you think this. By justifying the
points you make, you are showing your critical judgement. Through this, you will also
provide a reasonably detailed justified analysis of the key literature in your research
topic area, including research that both supports and opposes your ideas.
Table 2.1 What a critical literature review is and is not
A critical literature review . . . A literature review that is not critical . . .
•    identifies and includes the most relevant and
significant research to your topic
•    includes all research that may possibly be
relevant to your topic
•    discusses and evaluates this research •    just summarises and describes this research
•    identifies the recognised experts (authors) •    fails to mention recognised experts (authors)
•    contextualises and justifies your research
questions
•    fails to justify or mention your research
questions
•    highlights areas where new or further
research is needed
•    does not highlight where new or further
research is needed
•    considers and discusses research that
supports and opposes your ideas
•    only considers and discusses research that
supports your ideas
•    justifies points made logically with reference
to published research and clear argument
•    makes unjustified or poorly justified points
•    distinguishes between fact and opinion •    mixes up fact and opinion
•    includes recent relevant research •    misses out on recent relevant research
•    references all items referred to fully •    fails to reference all or some items referred to

34 Chapter 2 Reviewing the literature critically
Now you know what a critical literature review is, it is important to understand why you
need to undertake one and include it in your project report. While we have already provided you with some hints about the reasons for this in the previous section, we still
believe it is worth talking about these and others in more detail.
Provides the base on which you will build your research project
Reviewing the literature critically is one of the first things you need to do when you
start your research as it will provide the base on which your research project is built. By
reviewing the literature, you will be able to develop a good understanding of what
research has already been undertaken on your topic area. This means that you will be
clear about what is known about your topic, and also the context in which it is known,
and will be able to write about this in your literature review chapter or chapters. Let’s
say your research topic is personalisation of marketing. As you review the literature, you
will begin to find out not only how online marketers are using technology to provide
potential customers with unique and customisable content, but also the reasons why
organisations are personalising their marketing. You will also become aware of the
countries and industry sectors in which this research was conducted – in other words,
the context of the research. Eventually you will have sufficient knowledge from what
you have read to discuss the broader context of your own research on personalisation of
marketing. You will also be able to explain and justify your own research questions in
relation to what is already known (and what is not known!) about viral marketing and
in what contexts (section 1.6).
Helps you decide on the topic you want to research for your project
You will have read about how reviewing the literature can help you decide on your
research topic in section 1.4. In that section, you read about the importance of academic review articles in providing you with a thorough review of the state of knowledge
in a topic area as well as with pointers towards areas where further research needs to be
undertaken. You also read how browsing peer-reviewed journals could provide you with
a good source of possible research ideas. You will find that such research ideas are often
introduced under the heading ‘Directions for future research’ or with the phrase
‘Further research is needed.’ These articles are useful not only for the ideas, but also
because they highlight areas where research is needed and give you a justification for
undertaking your research to which you can refer. It is also likely that the theories these
articles identify will be those you will explore further in your own research.
Although books may be less up to date than journals, they can also suggest research
topics to you. In particular, edited books based on papers given at a conference or seminar often contain a final chapter that summarises the key findings from the earlier
chapters and draws out key themes, including future directions for research (Research
in practice 1.1).
2.3 Why it is important to review the literature critically
2.3 Why it is important to review the literature critically 35
In addition to reviewing academic journal articles and textbooks, you will also be
reading trade and professional magazines. Reviewing this type of literature will provide
you with an indication of whether your research topic is ‘hot’. Returning to our example of personalisation of marketing, if this has been discussed in recent issues of marketing magazines such as
digital marketing magazine, it is newsworthy and likely to be a
‘hot’ topic. This can be helpful, particularly if you are going to collect your own data.
People are often keener to be involved in research they believe is topical and that will
have immediate benefits for them or their organisation (section 3.2).
Can give you insights into the secondary data that are likely
to be available
It is likely that you will make at least some use of data that have been collected by others
for some other purpose in your research project. Such secondary data are useful, not
least because they give you access to far larger data sets than you could collect yourself
and can also provide you with comparative and contextual data (section 4.3).
As you review the literature critically, you will become aware of the data that other
researchers looking at your topic have used. In some articles and reports (and occasionally books) the data collected will be presented as tables. You can use these as secondary
data for your own research, providing of course you reference the article, report or book
as the source of the data! In other articles, reports and books the authors will explain
how they have used data sets collected originally by someone else for some other purpose. Where these data sets are in the public domain, you will also be able to find,
download and use them in your research, providing of course there is a full reference to
the source of these data.
Can give you insights into possible ways of collecting
your own data
The literature you review will also give you ideas about how you might collect your own
data. This is important because, if a method has been used before to collect similar data
and worked well, it will hopefully work if you decide to use it. In particular, articles and
book chapters that report research using data that the researcher has collected will
nearly always describe how those data were collected in a section headed ‘Method’ or
‘Methodology’. While you would hope that the description is in sufficient detail for you
to be able to understand fully and repeat the same method, unfortunately this is not
always true. However, even if the description is not particularly detailed, it can still provide you with an indication of how you might collect your own data. In addition, it is
likely to include references to other items which you can read to find out more about
the method used.
Let’s say that you read an article on your research topic that has used a questionnaire
to collect the data. This article provides both a detailed description of the method and,
as an appendix, a blank copy of the questionnaire that the authors used to collect their
data. This gives you a clear idea of the questions you need to ask if you decide to also use
a questionnaire to collect your own data. If you think the questions are suitable for the

36 Chapter 2 Reviewing the literature critically
people you are intending to collect your data from, you may decide to use the same
questions. However, if you do this, you must give credit to the people who originally
designed the questions and include the full reference to their article in your methods
chapter. If you do not do this, you may be accused of plagiarism, as you are passing off
the work of others as your own.
Alternatively, you may read an article in an academic journal that focuses on
research methods, such as
Organizational Research Methods or Field Methods, which
explains the use of a particular technique for collecting data such as a form of interviewing or observation or a specific use of questionnaires. Such articles will provide
you with helpful insights about how to conduct your own research and what to avoid.
They are likely to contain more detail about particular techniques than we have been
able to include in Chapter 6. Where you use the ideas from one of these articles for
your own method, you should refer to that article in your methods chapter (Research
in practice 2.1).
Can give you insights into possible ways of analysing
your own data
Academic journal articles can provide insights into how to analyse both data you collect yourself and secondary data. The techniques used to analyse data are usually
named in the ‘findings’ or ‘results’ section. Where these are widely used techniques, in
particular statistical tests, it is likely that, although the test is named, it will not be
explained in any detail as it will have been assumed that the reader (you) will already
be aware of and understand the technique. In addition, you will not see a reference to
a journal article or a statistics textbook. This is not a problem as you will almost
Insights into research methods from the literature
An article by Mark (Saunders, 2012) in Field Methods concludes by offering a series of
recommendations about the use of web-based questionnaires for organisational research
which may be helpful for your research. These include:
1 For research in organisations, the use of web-based questionnaires should be considered
only where respondents are IT-literate and have ready access to the Internet at work.
2 Web-based questionnaires should be designed to allow the impact of people not
responding to be assessed.
3 Caution should be exercised regarding the inclusion of questions on topics related to
the use of the web or associated technologies.
4 Care should be taken when comparing responses to open (write in) questions from
web-based questionnaires with those from questionnaires delivered using other
methods, owing to the impact of the technology on response length.
Research in practice 2.1
2.4 The types of literature available to you 37
certainly be able to find sufficient information about the test in your own statistics
textbook or by using the Help feature of a statistics software programme such as IBM
SPSS Statistics.
Where an academic journal article uses data analysis techniques that are not widely
known, it is likely that the author or authors will have explained how to use the analysis
technique. Hopefully, this will be in sufficient detail for you to be able to understand
fully and use the same technique, although unfortunately this is not always true.
However, even if the description is not particularly detailed, it can still provide you with
an indication of how you might analyse your own data. In addition, it is likely to include
references to other books and articles which you can read to find out more about the
technique. Such articles are likely to contain far more detail than we have been able to
include in Chapter 7, and where appropriate should be referred to in your method or
findings chapter.
The amount of literature available to you is expanding rapidly as more sources are made
available online. In addition to the library catalogue, your university library’s web pages
will provide a comprehensive list of the other sources you can access and the types of
literature they contain. Most of these other sources accessed are databases, so your
library’s web pages will usually include direct web links and details of any passwords
needed.
We have listed the main types of literature you are likely to access through your
library catalogue and databases for your critical literature review in Table 2.2, along
with a brief description of their contents and an indication of their likely use for your
review. Textbooks, such as those that have been recommended to you for your modules,
will be very helpful as you start your literature review, providing an overview of your
research topic and the recognised experts. However, academic journal articles which
have been
peer-reviewed (also referred to as ‘refereed’) will be the most useful to you in
your literature review and will form the majority of items you use. These articles will
have been read and evaluated by academic experts, called peer reviewers, to ensure they
meet the quality criteria of the journal before they are published. Peer reviewers are usually anonymous and ensure quality by both detecting errors, which have to be corrected
before the article is published, and providing a detailed assessment of the article. Where
peer reviewers consider the quality is not sufficient or there are too many errors, the
article will not be published.
2.4 The types of literature available to you
Definition
peer review: the process of evaluating an article by experts to establish whether it meets quality
criteria and is suitable for publication.

38 Chapter 2 Reviewing the literature critically
Table 2.2 Main types of literature available to you
Type Contents
Use for your critical literature
review
Textbooks Written specifically for audiences such as
students or professionals. Material usually
presented in an ordered and relatively
accessible form. Often draw on a wide
range of sources including peer-reviewed
academic journal articles.
Useful, particularly as an
introductory source for an
overview of your research
topic and to find the
recognised experts.
Peer-reviewed
(refereed) academic
journal articles
Provide detailed reports of research. Articles
written by experts in the field and evaluated
by other academics (peer reviewers) to
assess quality and suitability. Pay rigorous
attention to detail and verification of
information. Usually contain an extensive
list of references. Before publication, have
usually been revised in response to
comments. Not all academic journal articles
are peer-reviewed (see below).
The most useful type for your
literature review.
Non-refereed
academic journal
articles
Articles may provide detailed reports of
research. Articles selected by an editor or
editorial board with subject knowledge.
Relevance and usefulness
varies considerably. Beware of
possible bias.
Professional and
trade journal articles
Articles written for members of
professional or trade organisations, so
related to their needs. Consist of a mix of
news items and more detailed accounts of
a practical nature. Articles rarely based on
research, although some provide
summaries of research.
Can provide useful insights
into practice, although may
be biased. Need to be used
with considerable caution.
Newspaper articles Articles written for members of public,
most newspapers addressing a particular
market segment. News presented is
filtered dependent on events, priority
being given to headline-grabbing stories
that are likely to appeal to that
newspaper’s readers.
Good source of topical events
and developments. May
contain bias in reporting and
coverage.
Conference
proceedings
Selected papers presented at a conference,
often published online, as a book or special
edition of a journal. Usually peer-reviewed.
Sometimes difficult to find.
Very useful if the theme of
the conference matches your
research.
Reports Reports on specific topics written by
academics and various organisations,
including market research organisations
and government departments. Often
available online. Beware of possible bias.
May not have gone through same review
process as peer-reviewed academic journal
articles, but those from established
organisations often of high quality.
Often difficult to access or
expensive to purchase. Can
be a useful source of
information when the topic
matches your research.

2.5 Searching for and obtaining literature 39
Although your skills in using general search engines such as Google and Bing are likely
to be excellent, we have found that many of our students are unaware of at least some of
the search features of full text databases of academic articles such as
Business Source
Premier
and Emerald Insight. While you will have already used these or similar databases
of academic journal articles for assignments, there are relatively simple things you can
do to make your searches as fruitful as possible.
The process of searching for and obtaining the literature consists of five stages:
1 Decide on your topic.
2 Identify and note the search terms and phrases you will use.
3 Choose your online databases.
4 Undertake your search.
5 Download relevant publications.
We will now look at each stage in more detail.
Decide on your topic
You have probably already got a reasonably clear idea of your research topic. However, if
you haven’t, don’t worry. Just turn back to Chapter 1 and look at sections 1.4 and 1.5,
which talk about generating ideas for your research topic and refining these ideas. This
process of refining these ideas is important because, unless you have a reasonably clear
and specific idea of the topic you wish to search for in the literature, most of the items
you find are not going to be of use for your research project!
Let’s say you’ve defined your research topic as ‘financial management’ and decide to
search for items on this topic in one of the online databases to which your university subscribes. Because the topic of financial management is so broad, an initial search for ‘financial management’ will probably find more than 100,000 items – far too many for you to
ever download, let alone read! However, if you define your topic as ‘the impact of recession
on financial management’, you can be more precise in your choice of keywords with
which to search the literature. By using keywords and phrases such as ‘financial management’ and ‘recession’ as search terms, you would be likely to find about 1,000 items, a
slightly more manageable number for you to begin to look at. You might have been even
more specific, defining your topic as ‘the impact of recession on financial management
budgeting’. Searching for this topic in the online database using keywords and phrases
such as ‘financial management’, ‘recession’ and ‘budget’ as search terms would find fewer
than 100 items. Not only is this a far more manageable number to look at, but it is also
likely that a greater proportion will be directly relevant to your research topic.
Identify and note the search terms and phrases you will use
Databases find relevant items by matching specific words and phrases with either the
full text or part of the publication such as the abstract (summary) or title. You will have
2.5 Searching for and obtaining literature
40 Chapter 2 Reviewing the literature critically
found when searching with Google or Bing that the more specific the word or phrase
you type in, the more likely you are to find relevant web pages. You will have also found
that if the word or phrase you type in is too specific, you don’t find the web pages you
need. The same principle applies to searching databases, including those of academic
journal articles. You therefore need to think carefully about the
search terms you will
use while searching, and, as when searching using Google or Bing, you need to be prepared to try a variety of keywords and phrases.
Definition
Search term: a word or phrase that describes your research topic, question(s) or objectives and can
be used either on its own or in combination with other phrases in databases.
Choose your databases
Your choice of databases will, not surprisingly, be dictated primarily by those available
through subscriptions paid by your university. We have listed those most frequently
available in Table 2.4, along with a brief description of what they cover and therefore
what you can search for using your search terms and phrases. Of these, the most widely
used by business and management students such as you are Business Source Premier
and Emerald Insight. We would recommend that you start your literature search using
these. However, these databases do not have such a good coverage of older items, particularly those published over three to four decades ago. You will find the database
JSTOR a useful source for journal articles, particularly when you need to obtain a copy
of that seminal article recommended by your supervisor.
Your ideas for possible search terms and phrases, such as ‘financial management’ and
‘recession’, are likely to be based upon what you already know about your research topic
and any reading you have already done. In addition, you will probably find some of the
techniques to generate research topics outlined in section 1.4 helpful. Techniques such as
looking at past project titles, brainstorming, concept mapping and discussing your ideas
can all help you clarify your thoughts and come up with possible search terms and phrases.
You will also find that dictionaries and encyclopaedias (including Wikipedia) are often
helpful for suggesting alternative words and phrases. We have listed six things to think
about when identifying your search terms in Table 2.3. They are in no particular order.
Table 2.3 Six things to think about when identifying your search terms
Using: Example

Terms appearing regularly in literature you’ve read
Both broad and narrow terms
Different words with the same meaning (synonyms)
Alternative spellings of the same word
Abbreviations and the full term
Old and new place names where these have changed
Downsizing and Redundancy
Services and Restaurant
Motorcycle and Motorbike
Organi
sation and Organization
EU and European Union
Peking and Beijing

2.5 Searching for and obtaining literature 41
You will notice that we have included Google Scholar in Table 2.4. This is because it has
a number of useful features to support your search of the literature and is readily available. In particular, once you know the names of recognised experts researching your
topic, you can find out who has referred to (cited) their work by simply searching for
their names. Google Scholar will rank all the recognised experts’ publications and, for
each, list the full references of other documents that have cited (referenced) the publication. This means you can see which of your experts’ publications have been cited
most, and you can find these other documents that have cited these publications
subsequently.
You will have also noticed that, in contrast, we have not included Wikipedia in our
table. There is no doubt in our minds that Wikipedia is useful if you want to find out
something quickly. However, the level of detail provided is unlikely to be sufficient for
your research project. As you know, Wikipedia articles can be edited by anyone and are
allowed to be imperfect. Despite Wikipedia’s clear editing policy and their request that
Table 2.4 Frequently used online databases of academic articles in business and
management
Database Description
Business Source Premier
(sometimes called EBSCO)
Full text articles for over 2,900 English-language journals published
worldwide. Covers all areas of business and management with access
for some titles back to 1886. Embargo of up to a year for full text of
articles from the most recent issues of some journals. Also includes
Datamonitor company profiles for the world’s 10,000 largest
companies, Datamonitor industry profiles for various industries and
Economist Intelligence Unit country reports.
Emerald Insight Full text articles for over 160 English-language management journals
and reviews from over 300 management journals. Covers all areas of
business and management.
JSTOR Full text articles for science, social science, arts and humanities
journals. Coverage usually extends back to volume 1, issue 1 of
journals and more of the current issues of journals are also becoming
available. Often the best place to find ‘old articles’.
Blackwell Reference
Online
Blackwell Encyclopaedia of Management, Blackwell ‘Handbooks’ and
‘Companions’ in business and management.
Wiley Online Includes over 1,100 full text journals covering the sciences, business,
law, humanities, psychology and social sciences.
Google Scholar Scholarly literature consisting of articles, theses, books, abstracts or
court opinions from many disciplines and sources. Usually provides a
web link, but access can depend on subscription. Literature ranked
according to the full text of each document, where published,
author(s), how often and how recently cited in other scholarly
literature.
Nexis Full text of articles in newspapers across the world and UK national
and regional newspapers as well as company information and
reports.

42 Chapter 2 Reviewing the literature critically
contributors ensure the information they add is verifiable, errors and misinformation
can and do occur. While these errors are usually put right quickly, an article may be
wrong at the time you read it. Finally, although Wikipedia expects articles to be phrased
to reflect the present consensus on a subject, people with rival opinions do compete to
change what is written, resulting in the information you find changing overnight. We
therefore recommend that you do not use Wikipedia as a source for your critical literature review.
Undertake your search
Now you have identified the search terms and phrases you are going to use and chosen
your databases, all you need to do is click on the links on your university library’s web
page and you are ready to start searching.
Most databases allow you to undertake a variety of searches ranging from ‘basic’ or
‘quick’ to ‘advanced’ in their complexity. Basic searches, as their name suggests, are
relatively simple, often limiting the number of search terms you can type in. More
advanced searches, as we will see shortly, allow you to specify your search more precisely. In addition, many databases have a browse facility allowing you to look through
electronic copies of journals by title or by subject.
The first thing you need to do for a basic search is to type in the search term. If you
look at the basic search screen for the online database Emerald, you will see that you
can type in one term or phrase as a search term (Figure 2.2, point 1) and have only one
option, selecting the content you want to search, in this case ‘articles and chapters’
(Figure 2.2, point 2). When you click on ‘Search’, this will retrieve all relevant items in
the database containing the search term, in this example ‘articles and chapters’ articles
about ‘service quality’. For this example, this is likely to be over 100,000, as the search
term is very broad!
1. Type in search
term
2. Select content
you wish to
search
3. Note ability to
browse content
by subject area
Figure 2.2 Quick search using Emerald Insight
Source: Emerald Insight, reproduced with permission.
2.5 Searching for and obtaining literature 43
For more advance searches, again the first thing you need to do is to type in your
search terms (Figure 2.3, point 1) using Boolean operators to link two or more search
terms (Figure 2.3, point 2). Boolean operators are the words you use to join together
your keywords and phrases for your search. If you look at the advanced search screen for
the online database Business Source Premier (Figure 2.3), you will see that the search
terms ‘supply chain’ and ‘power’ are joined by the Boolean operator ‘AND’. This
means that when you undertake this search, you will retrieve only items that contain
both these search terms. In other words, the word ‘AND’ is narrowing your search to
retrieve items that contains both search terms. Other Boolean operators that you may
find useful are ‘OR’ and ‘NOT’. The word ‘OR’ broadens your search by retrieving items
that, in this example, contain either the search term ‘supply chain’ or the search term
‘power’ or both search terms. The word ‘NOT’ narrows your search by excluding items.
This means that if you searched for ‘supply chain’ NOT ‘power’, you would retrieve only
items that included the search term ‘supply chain’ but that did not also include the
term ‘power’.
In many databases, you can amend your search by truncating your search terms or
using wildcards. For most databases, the truncation character is *. If you truncate the
1. Type in the search
term(s)
6. Select publication
type to constrain
your search
5. Select dates to
time-constrain
your search
4. Select if you want
only peer-reviewed
publications
3. Select where you
will search for
each term
2. Use Boolean
operators to link
search terms
Figure 2.3 Advanced search using Business Source Premier
Source: EBSCO Information Services, reproduced with permission.
44 Chapter 2 Reviewing the literature critically
keyword ‘learning’ to ‘learn’ by typing in the search term ‘learn*’, the database will
retrieve all items that include the word ‘learn’, including ‘learn’, ‘learning’ and ‘learner’.
Wildcards are useful if you have to deal with spelling variations, particularly between
US and UK English. The wild-card character differs between databases, but is often ‘?’.
So, if you type in ‘organi?ation’ as a search term, you will retrieve items with both the
US spelling organi
zation and the UK spelling organisation.
You will find it is best to start most searches by looking for your search term or terms
only in the abstract field (Figure 2.3, point 3).
Abstracts are summaries of articles, books
or reports that normally provide sufficient detail for you to decide whether or not they
are likely to be useful for your research. By only searching for your search terms in the
abstract, you will exclude literature that does not focus on your topic, even if your
search terms appear elsewhere in the main text.
Most online databases, including Business Source Premier, allow you to check

whether you want to retrieve all items or only those where the full text is available
and you have access to it. Similarly, you can often check whether you want to retrieve
all items, or only those that have been peer-reviewed (Figure 2.3, point 4). You can
specify the dates for which you want to retrieve publications (Figure 2.3, point 5) and
the type of publication (Figure 2.3, point 6). This can be useful if, say, you want to
retrieve only items that have been published in the last five years in academic
journals.
Definition
abstract: a summary of an article, book or report, providing an overview of what it contains and sufficient information for the original to be located.
Download the relevant publications
Before you obtain a copy of a relevant publication, usually by downloading an article in
an academic journal, it is worth making a quick assessment of its likely usefulness to
your research project. You can usually do this by reading the abstract on the screen
(Research in practice 2.2).
Obtaining articles is very easy if you have checked the box for only retrieving articles
where full text is available. For most journal articles, you will only have to click on an
icon to download an electronic ‘PDF Full Text’ copy which you can read using Adobe
Reader (if you have not already downloaded this, you can do so for free). For articles
where an electronic copy is not stored in the database, you will often be able to click on
a different icon and download the ‘Linked Full Text’. Again, you will usually be able to
be read this using Adobe Reader. For those articles not available electronically, your
library may well have a print copy. If not, you may be able to order it from another
library as an inter-library loan. You will probably be charged for this service, so use it
very sparingly.

2.5 Searching for and obtaining literature 45
Using an abstract to assess likely use of an article
Jaimie’s research project was about entrepreneurial learning in small and medium-sized
Enterprises (SMEs). In a search using the Emerald Insight database, she had found an
article in the peer-reviewed academic journal
European Journal of Training and
Development
by Saunders and colleagues (2014) that she thought would be useful. She
decided to read the abstract online to check.
Research in practice 2.2
Abstract
Purpose: To contribute to the literature on innovation and entrepreneurial learning by
exploring how SMEs learn and innovate, how they use of both formal and informal
learning and in particular the role of networks and crisis events within their learning
experience.
Design/methodology/approach: Mixed method study, comprising 13 focus groups,
over 1000 questionnaire responses from SME mangers, 13 focus groups and 20 case
studies derived from semi-structured interviews.
Findings: SMEs have a strong commitment to learning, and a shared vision. Much of this
learning is informal through network events, mentoring or coaching. SMEs that are
innovative are significantly more committed to learning than those which are less
innovative, seeing employee learning as an investment. Innovative SMEs are more likely
to have a shared vision, be open-minded and to learn from crises, being able to reflect on
their experiences.
Implications for research: There is a need for further process driven qualitative research
to understand the interrelationship between, particularly informal, learning, crisis events
and SME innovation.
Implications for practice: SME owners need opportunities and time for reflection as a
means of stimulating personal learning – particularly the opportunity to learn from crisis
events. Access to mentors (often outside the business) can be important here, as are
informal networks.
Originality/value: This is one of the first mixed method large scale studies to explore the
relationship between SME innovation and learning, highlighting the importance of
informal learning to innovation and the need for SME leaders to foster this learning as
part of a shared organisational vision.
Source: Saunders et al. (2014). Copyright © 2014 Emerald Group Publishing Limited Emerald Group
Publishing Limited 2046-9012 DOI 10.1108/EJTD-07-2013-0073. Reproduced by permission of the
publisher
The abstract confirmed that the Purpose of the article was to explore how SMEs learn
and innovate using both formal and informal networking. More detail was given in the
Findings and Research limitations/implications sections of the abstract. The Design/
methodology/approach
indicated that the research had been undertaken in the United
Kingdom using a questionnaire answered by over 1,000 SMEs’ managers as well as 13
focus groups.
Based on this information, Jaimie decided that the article was likely to be useful to her
research project, so she downloaded and saved the PDF file.

46 Chapter 2 Reviewing the literature critically
You have already started evaluating the usefulness of the literature to your research
through reading the abstract (Research in practice 2.2). Inevitably, the usefulness of any
item will depend on your research question and your aim and objectives. This means
that an article, which may be very useful for your friend’s research project, could be of
little or no use for your research project. As you review the literature, you will need to
read as much of the literature that is closely related and of value to your research question as time permits. Let’s say your topic is one where research has been ongoing for
some years. If this is the case, you should find plenty of literature of value that relates
directly to your research question. However, if your research question is about a new
topic, you may find that not much of value that is closely related has been published.
You will therefore have to review the literature more widely and set your own research
in a broader context of what is already known.
Each time you evaluate the usefulness of publication for your research project, you
are asking yourself questions about two interrelated aspects: its relevance and its value.
At the same time, you are also probably asking yourself ‘Have I read enough articles
yet?’ Any item’s relevance to your research project depends on the appropriateness of
the research reported and the connections you can make with your own research project. A publication’s value to you depends on the quality of the research that is reported.
This includes things such as how well methods and theory have been used, as well as
the quality of the argument. In contrast, the answer to the question ‘Have I read enough
articles yet?’ is about you having read sufficient publications to allow you to position
your research project in the wider context of what is already known, and to cite the
main writers with confidence.
Our students have found the questions in Table 2.5 helpful when they have been
evaluating the relevance of a publication to their research, and we hope that you will
also find them helpful for your own research.
2.6 Evaluating the usefulness of literature to your research
Table 2.5 Questions to help evaluate the relevance and value of literature
Question Comment
Relevance
1
Is the item recent? Although more recently published items are
often more relevant, do not discount items just
because they are old!
2 Is the topic of the research similar to your
own?
If it is, you may be able to compare and
contrast your findings.
3 Is the context of the research similar to
your own?
Beware: do not discount items just because the
context is different; you can often still get
useful insights.
4 Is this item cited (referenced) in other
items that you have also found useful?
If it is often referred to in other items, it is
probably important.

2.7  Reading, noting and correctly referencing useful literature 47
Question Comment
5 Is the author cited (referenced) in other
items you have found useful?
If yes, then the author is probably a recognised
expert.
6 Does the author contradict or support the
arguments you will make?
In either case, it is likely to be useful!
Value
7
Does the author use emotional words,
illogical argument or seem to only choose
examples that support the points being
made?
If the author does, the item is probably biased
and needs to be used with caution. If you do
use it, you should comment on the possible
bias and give reasons why you think this is so.
8 Does the author describe theory clearly
and in sufficient detail?
Even if there are gaps in the description of
theory, the item may be useful for the
references to the original theory.
9 Is the method used described clearly and in
sufficient detail?
Even if there are gaps in the method, the item
may still be of use. If you do use it, you may
need to comment on those aspects of the
method that are less clear.
10 Does the author highlight gaps in current
research or offer suggestions for future
research?
If the gaps or suggestions relate to your
research topic, they could provide a useful
justification for what you are doing.
11 Is there an extensive list of references at
the end?
A long list of references will often include
relevant publications that you have not yet
discovered – a great help.
Sufficiency
12
Do you recognise the author and her/his
ideas from other things you have read?
If you do, you are getting closer to having read
sufficient publications.
13 Have you read the publications by those
considered to be key researchers in the
field?
If you have, you are getting closer to having
read sufficient publications.
14 Can you write about the topic and
associated research with confidence?
Confidence normally only comes if you know
the topic well.
Source: Developed from Saunders et al. (2016).
Table 2.5 Continued
Reading and noting
You may be one of those people who, as they read an article or book, only uses a highlighter pen to mark those pieces they think are useful. If you are, beware! Just highlighting sentences and paragraphs that are likely to be useful will not help you concentrate
nearly as well as making notes in the margin or on a separate piece of paper. Your notes
Reading, noting and correctly referencing
2.7 useful literature
48 Chapter 2 Reviewing the literature critically
can remind you why you thought the passage in the article was important, how an idea
fits with your own research topic or even why you disagree with what you have just
read. Making notes will help keep your mind focused on what you are reading and to
remember what you have read.
Harvard College Library (2011) suggests that as you read, your note taking should
focus on:
1 summarising;
2 comparing and contrasting.
When you write a summary of what you have just read in your own words, you probably
find the process quite difficult. This is not surprising, as by summarising in your own
words, you have to understand fully what you have read. In contrast, copying sentences
just shows that you can copy. Copying sentences can also be a problem if you later forget they were copied and include them in your project report as your own words. Most
plagiarism detection software will highlight your copied sentences as actual direct
quotes and suggest their original source. If you have not put the copied sentences in
quotation marks and also stated their source as a reference, you will, quite rightly, be
accused of plagiarism. You can minimise this problem by making notes in your own
words that summarise what you have read and always acknowledging the source of the
ideas. If you decide you must copy a sentence exactly, enclose it in quotation marks and
note the full reference and page number next to it.
As you make your notes, ask yourself how the item you are reading compares or
contrasts with other articles and books you have read and your own thoughts upon
reading it. Within this you need to think carefully about the clarity of the arguments and whether they are justified by the evidence presented. Let’s say you have
already read Hofstede et al.’s (2010) book
Culture and Organizations about national
cultures. The article you are now reading by McSweeny (2002) is very critical of
Hofstede’s research and in particular the assumptions on which he bases much of
the research in this book. You are not sure who to believe, but you no longer feel so
positive about Hofstede’s theories! You therefore emphasise in your notes that
McSweeny believes Hofstede’s arguments are flawed, the reasons why and that you
don’t feel so positive towards Hofstede’s theories about national cultures. You also
find your notes on Hofstede’s book and add a brief comment, ‘See McSweeny (2002)
for critique’, to help ensure you do not forget to include this information in your
literature review.
Five questions that can help you to read critically and make useful notes are listed in
Table 2.6, along with a brief comment on each.
Definition
plagiarism: presenting the work and ideas of other people as it were your own, without acknowledging and referencing the original source.
2.7  Reading, noting and correctly referencing useful literature 49
Referencing
As you write your project report, you will need to record the sources of all the information, research findings, theories and other ideas to which you refer in your writing. This
process is known as referencing and allows you to acknowledge and give credit to the
work of others, also helping ensure you are not accused of plagiarism. It allows you and
anyone else who reads your project report to find the original source of this work. You
reference work by identifying the source of the ideas briefly in your main text and then
provide full details as a
list of references or bibliography at the end of your project report.
As a general rule, it is better to avoid quoting directly. However, if you do use a quotation, you must enclose this in quotation marks and state the number of the page from
which it was copied. So, if you were quoting directly from a book Mark and Phil wrote
with their colleague Adrian Thornhill, you would write:
Table 2.6 Five questions to help critical reading and noting
Question Comment
1 Why am I reading this? This question will help you focus on the reason or
reasons you are reading the item, rather than be sidetracked by the author’s agenda.
2 What is the author (or what are the
authors) trying to do in writing this?
Your answer to this question will help you to decide
how useful the material that you are reading may be
for your research project.
3 What is the author (or what are the
authors) saying that is relevant to my
research topic?
Your answer to this question will give you the focus
of your summarising notes.
4 To what extent am I convinced by
what is being said and why?
Your answer to this question will help you make
notes that compare and contrast what you are
reading with the other items you have already read.
5 What use can I make of what I have
read in my research project?
This question forces you to think about how you will
write about what you have read in your literature
review.
Source: Developed from Wallace and Wray (2016).
“For some project reports you will be required to include a bibliography. Convention dictates
that this should include all the relevant items you have consulted for your project, including
those not referred to directly in the text. For others you will be asked to include only a list of
references for those items referred to directly in the text.” (Saunders et al., 2016: 107)
Definitions
(list of) references: list of those items you have consulted, used and referred to directly in the text.
Your university will specify the precise style.
bibliography: list of all relevant items you have consulted and used, including those you do not refer
to directly in the text. Usually in alphabetical order. Your university will specify the precise style.

50 Chapter 2 Reviewing the literature critically
You would then provide the full reference at the end of your project report. We have
provided a guide in Appendix 1 to a version of each of the most commonly used styles
of referencing in business schools: Harvard and the American Psychological Association
(APA).
For books
• Title of book
• Edition (unless 1st)
• Place of publication
• Publisher
For all items
• Surname and initials of
author(s)
• Year of publication
For book chapters
• Title of chapter
• Surname and initials of
author(s)’ of the book
• Title of book
• Edition (unless 1st)
• Place of publication
• Publisher
• Page numbers of
chapter
For journal articles
• Title of article
• Title of journal
• Volume number
• Part or issue number
• Page numbers of
article
Figure 2.4 Information you need to reference publications
However, rather than just using one of the two formats we suggest, please check to
see if your own university or faculty has a preferred referencing style for recording articles, books and chapters in books as well as other items you have referred to in your literature review and elsewhere in your project report. If it does, it is essential that you
obtain a copy of this and follow it exactly. Sloppy referencing suggests you do not care
about your work and is likely to lose you marks. Whatever referencing style you use, the
information you will need to record for the most common publications you reference is
listed in Figure 2.4.
The most common referencing style in Business and Management is the author–date
system. When using this system, you usually use the author’s (or authors’) surname(s)
and the year of publication to identify documents in the main text as you refer to them.
All these references are then recorded in full in one alphabetical list by author’s surname at the end of your project report under the heading ‘References’ (Research in
practice 2.3). Another referencing system you may come across in your reading is the
footnotes system. Footnotes systems (such as Vancouver referencing), identify references in the main text with a number that is usually in superscript. All these references
are recorded in full in the order to which they were first referred, usually in one sequential list at the end of your project report under the heading ‘References’. This means
that, if you use the footnotes system, your list is unlikely to be in alphabetical order
(Research in practice 2.3).

2.7  Reading, noting and correctly referencing useful literature 51
Footnotes and author–date (Harvard) systems
Amina had used the footnotes system to reference the books and journal articles she
referred to in the first draft of her literature review on strategic change. This is shown in
the following two extracts from her literature review and list of references:
Research in practice 2.3
2.3 Barriers to strategic change
. . . The cultural web22 offers a way of auditing an organisation’s culture and the barriers
to change that culture can present. It has been argued that the web can be used to help
build a vision for the new organisation, which managers can then compare with the web
for their existing organisation.
23 Despite the numerous obstacles to strategic change,
summarised by Franken and colleagues
24 as . . .
References
22. Johnson, G., Whittington, R., Scholes, K., Angwin, D. and Regnér, P. (2014) Exploring
Corporate Strategy
(10th ed.), Harlow: Pearson.
23. Balogun, J., Hope Hailey, V. and Gustafsson, S. (2016)
Exploring Strategic Change
(4th ed.), Harlow: Pearson.
24. Franken, A., Edwards, C. and Lambert, R. (2009) Executing strategic change
,
California Management Review, 51(3), pp. 49–73.
In his feedback her supervisor commented that, although she had used the footnotes
system correctly, the University’s assessment criteria asked for the Harvard author–date
system. Amina checked the guidance notes on referencing provided by her university
and amended her draft so that the references were in Harvard format. This is shown in
the following two extracts:
2.3 Barriers to strategic change
. . . The cultural web (Johnson et al., 2014) offers a way of auditing an organisation’s
culture and the barriers to change that culture can present. It has been argued that the
web can be used to help build a vision for the new organisation, which managers can
then compare with the web for their existing organisation (Balogun et al., 2016). Despite
the numerous obstacles to strategic change, summarised by Franken et al. (2009) . . .
References
Balogun, J., Hope Hailey, V. and Gustafsson, S. (2016) Exploring Strategic Change
(4th ed.), Harlow: Pearson.
Franken, A., Edwards, C. and Lambert, R. (2009) Executing strategic change
, California
Management Review
, 51(3), pp. 49–73.
Johnson, G., Whittington, R., Scholes, K., Angwin, D. and Regnér, P. (2014)
Exploring
Corporate Strategy
(10th ed.), Harlow: Pearson.
52 Chapter 2 Reviewing the literature critically
Theme A Compare and contrast authors’ ideas about each theme
Dickinson Gers McBrain
Authors
Murray Smith Wapram
Theme B
Theme C
Theme D
Theme E
Themes
Author’s ideas about themes
Figure 2.5 Structuring your critical literature review
Deciding on the structure
Your critical literature review is likely to form either one chapter or a series of chapters
in your project report. In being critical you will need to discuss different themes and
compare and contrast what different authors say about them. Your own opinions will
also need to be justified through your writing.
Many students’ early drafts of their literature reviews are simply listings of everything they have read, in which each author’s ideas are described one after another for
those themes they have discussed (vertical arrows in Figure 2.5). If you do this, your literature review will be boring and uncritical. You, and your readers, will find it much
more interesting if you take a thematic approach and write about one theme at a time.
As you compare and, where necessary, contrast those authors who have said something
about each theme, you will be writing critically (horizontal arrows in Figure 2.5).
2.8 Drafting your critical literature review
In our book Research Methods for Business Students (2016), we suggest that you should
think of your literature review as a thematic funnel. This means you should take the following steps:
1 Begin by providing a more general overview of your research topic before narrowing
down to your research question or research aim and objectives.
2 Start by providing a brief overview of the key themes and ideas in the literature
reviewed.

2.8 Drafting your critical literature review 53
3 For each theme, summarise, compare and contrast the research of recognised experts.
4 Within each theme, highlight previous research you have found that is most relevant to your own research, including work that may not be by recognised experts.
5 Provide a detailed account of the findings of this previous research and show how
they are related to the theme.
6 Use your review to highlight those aspects where your own research will provide
fresh insights, linking these explicitly to your research questions or objectives.
7 Tell your reader that these aspects will be explored in subsequent sections of your
project report.
8 Finish with a summary that links to your next chapter.
The key to you drafting an interesting and critical literature review is therefore to link the
different ideas you find within each theme in the literature, and then link the themes to
form a clear, well-justified argument. You need to ensure that this places your research in
the context of what is already known and justifies why your research is worth doing.
Getting on with writing
Although you will not be able to start writing until you have done some reading, it is
important you begin to draft your literature review before you have completed reading
all the literature. The process of drafting your literature review will help you get your
ideas about your research clear in your own mind. As you read more, you can update
and revise your draft as necessary.
You, like everyone we know, will find drafting your literature review difficult. As you
write, you have to get your ideas clear in your own mind and put them into words such
that anyone who reads them can understand. Phil refers to this writing process as the
‘rewriting process’, emphasising the need to improve early drafts of what you have written. Your literature review writing process will be no different, and you should expect to
have to rewrite your critical review more than once. Indeed, what you write at the start
of your literature review will almost certainly have to be updated and revised as you read
more over the course of your research project.
We often compare a poor literature review to a shopping catalogue. This describes
what your literature review should not be like. If you wrote your literature review as a
shopping catalogue, you would:
Include every item you have read.
Group each of the items you have read in some way.
Sort the items within each theme in some way, such as by date or author.
Describe each item using a similar number of words.
Obviously, your literature review will be far more than just a shopping catalogue, as you
need to be critical. Your writing, therefore, needs to show that you have:
Reflected on the value of the findings of each item you have read.
Reflected on the value of the ideas of each item you have read.
Made reasoned judgements about the value of each item you have read.
54 Chapter 2 Reviewing the literature critically
Included only those items that are relevant.
Organised the ideas and findings into a coherent logical argument by theme.
Look at the project report extracts in Research in practice 2.4. You will notice that in the first
extract, Ben simply lists the ideas from the literature one after the other like in a shopping
catalogue. The order in which he has written about each of the items does not appear to be
logical, the more general ideas of Rousseau et al. and of Schoorman et al. being described
after his summary of more specific and focused arguments by Möllering. In addition, Ben
has only listed the items and has not linked the authors’ ideas together to form an argument.
At the end of this extract, he quotes a Wikipedia definition rather than discussing the definitions used by academic researchers, something which is unlikely to please his supervisor.
Writing – a shopping catalogue or critical review?
Ben was feeling pleased. He had spent a great deal of time reading the literature and felt
he had described clearly everything he had read. He emailed the first draft of his literature review to his project supervisor in advance of his supervision meeting. At the meeting his supervisor was obviously not impressed by Ben’s writing. She commented that
his draft was ‘more like a shopping catalogue than a critical literature review’ and needed
to be ‘completely rewritten’. Ben listened carefully to his supervisor’s comments and
went back to his room to redraft his work as a critical literature review.
This extract is taken from Ben’s first draft:
This is the same passage after Ben had redrafted it as a critical review:
Research in practice 2.4
. . . Möllering (2001) argues that trust develops from favourable expectations based
upon interpretations of the reality to which trust relates, enabled by a suspension of disbelief and a corresponding leap of faith. Rousseau et al. (1998) argue that the development of trust theory has, to date, been disparate, focusing on a range of levels of analysis
from the interpersonal to the inter-organisational. Schoorman et al. (2007) recognise that
there is a need for a context specific understanding of trust. Wikipedia (2017) states that
trust is the willingness of one party to be vulnerable to the actions of another party, with
the reasonable expectation that they will behave in a beneficial way . . .
. . . Development of trust theory has, to date, been more disparate, focusing on a range
of levels of analysis from the interpersonal to the inter-organisational (e.g. Rousseau et al.,
1998) and recognising the importance of context (Schoorman et al., 2007). Although this
has resulted in a variety of definitions of trust, these exhibit a number of common elements, including notions of ‘favourable expectations’ and a ‘willingness to become vulnerable’. Möllering (2001) has sought to use and develop these common elements in his
research, arguing that trust develops from favourable expectations that are based upon
interpretations of the reality to which the trust relates, enabled by a suspension of disbelief and a corresponding leap of faith. This suggests that . . .

Summary
Now re-read the second extract in Research in practice 2.4. You will notice that,
55

unlike in the first extract, the order in which Ben writes about the ideas from the literature appears more logical, going from more general to more specific ideas. He starts
with a general summary statement, referring to two sources as examples (Rousseau et
al., 1998; Schoorman et al., 2007). In the next sentence, he provides a more detailed
evaluation, highlighting similarities between authors. Ben uses the phrase ‘although
this has . . .’ to link these two sentences. He then introduces Möllering’s more specific
and focused arguments, repeating the term ‘common elements’ from the previous sentence to help emphasise a clear link. This time Ben does not refer to Wikipedia. His final
phrase ‘This suggests that . . .’ provides a link between his summary of Möllering’s ideas
and what he is going to write next.
Finally, a quick reminder. Don’t forget to look at section 8.5, where we talk about the
process of writing in much more detail.
Your critical literature review should offer an overview of significant literature available in your chosen topic, including relevant peer-reviewed academic journal articles, textbooks and other sources. It should provide a discussion and evaluation
covering each of these, the level of detail reflecting the significance of each item. It
should develop a clear argument to contextualise and justify your research.
Critically reviewing the literature is important because it will provide the base on
which your research project is built, helping you to decide on your precise topic and
place it in the context of other research findings. It will also provide insights into
secondary data that are likely to be available, as well as possible ways of collecting
and analysing your own data.
The amount of literature available to you is expanding rapidly. In addition to the
library catalogue, your university library web pages will provide a comprehensive list
of the other sources you can access as a student and the types of literature they contain. The most useful sources will be the databases of academic journal articles.
Your literature review will consist predominantly of peer-reviewed academic journal
articles, although you are likely to find textbooks helpful, particularly early on in
your reviewing.
The process of searching for and obtaining the literature consists of five stages: (1)
decide on your literature search topic; (2) identify the search terms and phrases you
will use; (3) choose your databases; (4) undertake your search and (5) download the
relevant publications.
The usefulness of any item will depend on your research question and your aim and
objectives. When you review the literature critically, you will need to read as much of
the literature that is closely related and of value to your research question as time
permits.
Summary
56 Chapter 2 Reviewing the literature critically
As you read, you should make notes in your own words of ideas that are useful to
your research. Your notes should provide a summary of the item and remind you
why you thought the particular idea was important, how an idea fits with your own
research topic or even why you disagree with what you have just read. Making notes
will help keep your mind focused on what you are reading and to remember what
you have read.
As you write your project report, you will need to record the sources of all the information, research findings, theories and other ideas to which you refer in your text.
This process is known as referencing and allows you to acknowledge and give credit
to the work of others, helping ensure you are not accused of plagiarism.
The most common form of referencing in Business and Management is the author–
date systems (e.g. Harvard, APA). Your university will probably have a preferred referencing style and, if it does, it is essential that you obtain a copy of this and follow it
exactly.
Think of your literature review as a thematic funnel in which you link the different
ideas you find within each theme in the literature and then link these themes using
a clear, well-justified argument.
If you have not yet begun to think about critically reviewing the literature, start now.
Write your research question or aims and objectives on a piece of paper and refer to
them frequently to keep focused. (If you are still unsure what to do for your research
project, turn back to sections 1.4 and 1.5.)
Based on what you already know about your research topic, identify the terms and
phrases you will use to search. Be prepared to use a variety of different terms and
phrases.
Use the databases available through your university and your university library catalogue to search the literature for useful items. Although you will find textbooks listed
in the library catalogue useful in the early stages of your review, you should focus
mainly upon finding peer-reviewed academic journal articles, using databases of journal articles.
Obtain the articles and other items that seem likely to be useful and assess their relevance and value (usefulness) to your own research.
Use Google Scholar to find more recent items written by others that refer to those
items you have already found that you feel are most useful to your own research.
Make your own notes about these useful items and include a summary in your own
words. As you make notes, write about how the item compares with other items you
have read.
Thinking about your critical literature review
References 57
Note the full reference of each item, using your university’s preferred style of
referencing.
Begin to draft your critical literature review, using a thematic structure and comparing
and contrasting what each author has to say about a theme. Be mindful that you will
have to redraft your review more than once.
Harvard College Library (2011) Interrogating Texts: 6 Reading Habits to Develop in Your First
Year at Harvard
. Available at: http://bsc.harvard.edu/files/interrogating_texts_six_reading_
habits_to_develop_in_your_first_year_at_harvard.pdf [Accessed 12 October 2016].
Hofstede, G., Hofstede, G.J. and Minkov, M. (2010)
Culture and Organizations: Software of the
Mind
(3rd ed.). Columbus, OH: McGraw-Hill.
McSweeny, B. (2002) Hofstede’s model of national cultural differences and their consequences: a
triumph of faith – a failure of analysis.
Human Relations, 55(1): 89–118.
Saunders, M.N.K. (2012) Web versus mail: the influence of survey distribution mode on employees’ response.
Field Methods, 24(1): 56–73.
Saunders, M. N. K., Gray, D. and Goregaokor, H. (2014) SME innovation and learning: the role of
networks and crisis event.
European Journal of Training and Development, 38(1/2): 136–49.
Saunders, M., Lewis, P. and Thornhill, A. (2016)
Research Methods for Business Students (7th
ed.). Harlow: Pearson.
References
Chapter 3
Managing the research process
In Chapter 1 we noted that doing a research project gave you the chance to develop a
number of skills. Much of this chapter focuses in part upon the development of your personal organisational skills. Also in Chapter 1 we talked of the importance of managing your
own learning through choosing a research topic that will develop your skills and knowledge
as well as, possibly, your employment opportunities. We also noted the importance of
choosing a topic that matched the resources available to you, in particular time and data.
All of this suggests the importance of you taking control of the research process. It is a
process which you should lead, not one where you allow yourself to be led. In fact, we
would go as far as to say that self-organisation is a crucial factor in determining a successful
outcome to your research project.
So what are these components of the research process that need to be managed by
you? Well, assuming that you are doing primary research in a work organisation, the
organisation itself in the form of its key personnel, or its customers, are obvious starting
points. You will need to get access to the organisation in the first place. Then there are
the people from whom you wish to get the data. Obviously, paying attention to these
key resources is vital. Then there is the management of the components over which you
have more control. First, we concentrate upon the key component of managing oneself,
where attention needs to be paid to wider topics than just managing your time. Then
there is the relationship you have with your supervisor. Managing this does not mean
that you ‘take charge’ of your supervisor. It’s more subtle than that. Many students (and
lecturers, for that matter) talk of unsuccessful student/supervisor relationships. We help
you to ensure that your situation is successful. We then cover the relationship you have
with the university through, for example, the importance of paying attention to the
standards that are demanded of you. We represent the management of the research
process in Figure 3.1.
3.1 Why you should read this chapter
3.2 Getting access to your research organisation, respondents and participants 59
The final part of this chapter is about the ethics of doing research. This is a central consideration in the conduct of research. It’s essential that you consider the ethical implications
of your research in order to avoid falling foul of the expectations of your host organisation,
respondents, university and project supervisor.
Managing your research
organisation, respondents
or participants
Managing the
research process
Managing your
supervisor
Managing
your university
Managing
yourself
Figure 3.1 Managing the research process
So your research questions and objectives mean that it’s necessary to gain access to an
external organisation, or a number of organisations, to do some research. Much of this
section assumes that you will not be doing research using your employing or placement
organisation as the main context, so it’s one with which you are unfamiliar. (However,
we deal with issues that arise from being the ‘insider’ later in the section.) It’s possible, of
course, that you have had some introduction to the organisation, from a friend or a previous connection. But even if that is so, there are still key issues that you need to consider. These issues are divided into two main parts in this section. First, we consider some
of the main problems that you are likely to face when attempting to gain access to your
chosen organisation. Second, armed with an awareness of some of these problems, we
talk about how you may overcome them to achieve your goal of organisational access.
It would be good to think that the organisation in which you wish to do research will
be keen to help you. You may be lucky; it may. But the reality is that most will not be
keen. So what are some of the reasons for this possible lack of keenness to help?
What’s in it for us?
The reaction of the manager who receives the request is likely to be, ‘What’s in it for the
organisation?’ Managers are busy people who are paid by the organisation to help
achieve the organisation’s goals, not your goals! This seems a most obvious point, but
how often when we plan for a job interview, for example, do we concentrate upon what
we want and what we can offer and not think about the needs of the employer?
Getting access to your research organisation,
3.2 respondents and participants
60 Chapter 3 Managing the research process
Oh no! Not another request!
Be aware of the fact that your request may be one of many. You will be aware that there
has been huge growth in the number of degree students in recent years, including business and management students. This means that research access requests are much
more commonplace, particularly in a large national organisation.
Oh no! Not another student! Yet another demand on our time!
You must accept the fact that you may be a source of concern. This is probably the first
time you have done something like this, so the manager has to balance the wish to help
you (and many managers do genuinely want to help) and the risk that you may waste
valuable organisational time.
The manager receiving your request is going to be mindful of the resources that will
need to be devoted to you. That manager will need time to consider who may be responsible for dealing with your needs. In turn, the likelihood is that you will need to demand
the time of other people, for example, more junior employees. This will impact upon
the ability of people to do the jobs for which they are paid.
Who is going to get hold of this information?
There is also likely to be a worry about what you going to do with the information you
get from the organisation. We are not talking just about commercial secrecy here, but
organisational politics. Many managers are very sensitive about who knows what in
their own organisation, so they are going to be even more concerned about the public
availability of that knowledge outside the organisation. This may not necessarily be particularly rational – the information may not on the face of it seem sensitive. But we are
talking about human behaviour here, which isn’t always as rational as we may think!
Clearly you will need to give assurances about treating such information that you collect with strict confidentiality, assuming that this is what is required. You may also need to
promise that your participants or respondents will be anonymous so that their identity
cannot be traced.
Well maybe, but not right now . . .
Organisations will often cite reasons why they are not able to grant research access at the
time when you approach them. Of course, there may be an element of putting you off here,
in the hope that you will go somewhere else and leave them alone! But sometimes the reason
may be perfectly valid. It may coincide with a particularly busy period, such as new product
launch, or a sensitive time such as the announcement of a redundancy programme.
The discussion here refers to what may be called
physical access, or the ‘gatekeeper’
level of organisational access. But you will usually need to get through the ‘gate’ to see
Definition
physical access: gaining access to an organisation to conduct research.
3.3 What about access to information? 61
other people. Although ‘gatekeeper’ access is essential, it may also be necessary for
you to gain acceptance and consent from those within the organisation order to gain
cognitive access to the data that they are able to provide. These people may be other
managers, or their employees whom you may need to see as individuals or in groups.
So, let’s face it, the problems you may face when dealing with the
gatekeeper are equally
relevant at other levels.
Definitions
cognitive access: gaining agreement from individuals to providing research data, such as by the
answering of questions.
gatekeeper: the person, often in an organisation, who controls research access.
So far we have dealt with the question of access to the organisation in order to get into a
position where you can get the information you want. Now let’s deal with access to that
information itself. Here you need to be clear about precisely what data you wish to collect and the method or methods you intend to use to collect those data. This raises two
key questions:
1 Have you thought through fully the extent of the access that you will require in order
to be able to answer your research question(s) and meet your objectives?
2 Are you able to gain sufficient access to answer these research question(s) and meet
your objectives?
Both these questions point to the need for thorough preparation on your part – which
is a topic we will deal with in the next part of this chapter when we deal with strategies
to gain access.
Should organisational access be seen as a single event?
Your research question(s) and objectives may mean that you need to gain information
from people at different levels of the organisation, and from different parts of the
organisation, say different divisions or departments or geographical locations. Or it
may be that you need to gather data in fragments, separated by time. All of this suggests
that access is a continuing process and not just an initial or single event. So access negotiation becomes an even more important part of the management of the research process. Indeed, it may be continuous.
Getting access to the organisation as an ‘insider’
We mentioned at the start of this chapter that you may be an employee of the organisation in which you wish to do research, or you may be, or have recently been, a
3.3 What about access to information?
62 Chapter 3 Managing the research process
placement student in the organisation. This is great because people in the organisation
will know you and, hopefully, have positive views about you. But you are still likely to
face problems in getting the data you need. However, they may be rather different to
those faced by the ‘outsider’.
The likelihood is that, as the ‘insider’, you may still need to get formal approval to
conduct the research. The larger or more geographically diverse the organisation, the
more likely this is. All you may have is an introduction, so you may only be one step in
front of the ‘outsider’.
As the insider, you may come across the same issue of status. And it may not be just
because you are seen as quite a junior member of the organisation (indeed, you may be
quite senior). It will be an issue if, say, you belong to a part of the organisation which is
seen to be more powerful than the one in which you wish to do the research. Feelings of
suspicion may be even greater if you have been asked by your manager to take part in
the research project. (‘What are
they up to now?’) You may not feel much like a management pawn, but that’s the way you could be seen.
There is another disadvantage as the ‘insider’ researcher. You may feel reluctant to
ask very basic questions (e.g. ‘What job does Mrs X actually do?’ ‘Who is our biggest
client?’) because you feel you should know the answer. As the outsider, you would
not be expected to know the answers to these questions, so you can gather information
that you really need to know to set your research information in context.
It sounds as if we are being totally negative in this chapter so far! Well, getting access to
your research organisation certainly does pose challenges. But this chapter is all about
managing the research process, so you need to manage the challenges.
You will have to plan your approach carefully. How do you do this? We explain in
this section six strategies you can adopt to manage the process.
Strategy 1: Try to use your existing contacts
It’s undoubtedly easier to use any existing contacts you have to get access to an organisation to do your research. These contacts may come from friends, family, student colleagues, lecturer suggestions or contacts that you may make at meetings of the
professional association related to your specialism. Having an introduction at least avoids
that difficult stage where you approach the organisation as a complete unknown.
Although using a contact who has mentioned you may not put the manager who receives
your request under any obligation to give you the access, you are more likely to have the
chance to make your request than in the case of asking when you do not know anyone.
We have mentioned the possibility of using your work placement organisation as
your research organisation already in this chapter. If your placement has been a success,
you will have made useful contacts that may be able to help you. Potentially, this is
Six strategies for making sure that you get the
3.4 organisational access you want
3.4 Six strategies for making sure that you get the organisational access you want 63
much better than using contacts given to you by someone else. You are known and
trusted, and the organisation’s managers will know that you are familiar with the context of the organisation’s work. So your demand on resources may not be as great. It may
be even better to build on the work you did on the placement by doing a research project
that relates to this work. This may lead to the ideal situation, access to an organisation in
which you are welcomed to conduct some research that is valued by the organisation!
You may also consider using the organisation with which you are familiar as a
‘launch pad’ for your research, facilitating access to another organisation from which
you may gather your data. (Research in practice 3.1 is an example of this approach.)
Negotiating organisational access in an organisation unknown to you
Nazaneen was a part-time business studies undergraduate who worked full-time in one
of the major banks. She was very interested in the issue of cashless pay, both the technology that facilitated this development and the way in which it had the capacity to
change consumer behaviour.
She had followed the media coverage of this topic for some time and was intrigued
in the extent to which cashless pay had spread in Sweden where cash transactions made
up barely 2% of the value of all payments made in 2015 – a figure some see dropping
to 0.5% by 2020 (Henley, 2016). In Swedish shops, cash is now used for barely 20% of
transactions, half the number five years ago, and way below the global average of 75%.
In the United Kingdom, a major milestone on the path to a cashless society was passed
in 2015. This was the first year that consumers used cash for less than half of all payments, according to Payments UK, the organisation which represents the major banks,
building societies and payment providers. As an example, the UK ready-to-eat food chain
Pret a Manger reported that 65% of their customers use contactless payment cards and,
increasingly, mobile ‘phone systems’ (e.g. Apple Pay) to make their purchases.
Nazaneen’s reading of the relevant literature suggested to her that the main stumbling block to the widespread adoption of cashless pay in the United Kingdom was the
reluctance to embrace its use by small retail businesses. So she concluded that the best
way to understand their reluctance was to interview them. She was a bit wary of just
going around to small shops on a ‘cold call’ basis and speaking to the owners, and she
thought that something a bit more formal may be more appropriate.
Nazaneen talked over the matter of access with her research supervisor, who suggested that she contact the relevant trade organisation, which in this case is the National
Federation of Self Employed & Small Businesses Limited. This she did by establishing the
identity of the executive who seemed the most appropriate to her needs and then writing to that person with a full explanation of her project.
The letter which Nazaneen sent (she decided against using email – see strategy 4 in
this section), which included a copy of her formal research proposal, impressed the executive concerned. It resulted in a helpful interview at the organisation’s HQ. She was also
given a list of small businesses to approach which would be able to provide more data to
give a richer understanding of the nature of the perceived problems of the small retailer.
Research in practice 3.1
64 Chapter 3 Managing the research process
Strategy 2: Put yourself in the shoes of the manager receiving
your request
If the organisation is likely to be resistant to your approach, you have to think about
why that organisation is likely to be resistant. Put yourself in the shoes of the manager
receiving your request and think about how to overcome some of that resistance.
Strategies 3–6 offer specific guidance on practical steps you take to overcome resistance.
All these need you to think through how your request is likely to be seen by the manager receiving your request.
We noted above that the reaction of the manager who receives the request is likely to
be ‘What’s in it for the organisation?’ which seems a reasonable reaction. So it’s important to think about any possible benefits to the organisation. And there may be some.
Frequently, we have experienced participants who have confessed to us after an interview that they viewed the prospect of being interviewed as a bit of a chore. Yet we have
been delighted to hear from them that they found the interview helpful. Rarely do they
have the opportunity to talk through issues with an interested outsider. The more probing your questioning, based on thoughtful listening, the more value the participant is
likely to get from listening to the responses.
Participants(s) may also appreciate a summary report of the interview, although it is
not normally appropriate to send them a copy of your project report. The report is for a
different purpose. An interview summary helps in three ways. First, it enables the participant to check the accuracy of your summary, second, it allows the addition of material which may have been overlooked during the interview and last, it shows that you
have actually done something as an end product of the interview.
Let’s go back to the worry the manager may have that you will take too much valuable organisational time. Your request for access is more likely to be accepted if the
amount of time you ask for is kept to a minimum. And it’s important to be realistic
about the amount of time you ask for. Be honest. For example, falsely stating that each
interview will last for only 30 minutes, and then deliberately exceeding this, may very
likely upset the person you are interviewing and prevent your gaining further access.
Strategy 3: Make sure you give yourself sufficient time
to set up the arrangement
We hope that one of the messages that come across clearly in this book is that of giving
yourself plenty of time to complete your research project. To summarise this message:
start early!
Nowhere is this message more significant than the issue of allowing sufficient time
to set up access to your research organisation. It can take weeks of delay between your
original request and the final approval to go ahead (or, regrettably, the refusal). Even
where the go-ahead is granted there will be further delay before you gain access to the
people from whom you wish to get information.
Assuming that you are not using an existing contact (and even where you may be),
you are unlikely to receive a direct reply to your first approach. There will be delay while
you wait for this, and then for a response to your follow-up. This raises the question of

3.4 Six strategies for making sure that you get the organisational access you want 65
the amount of time you should wait, and the ‘tone’ you adopt when sending your
‘reminder’. You must ensure that you keep to the right side of the line between persistence and pestering. Whether you use email, telephone or letter (or a combination of
the three) depends on the situation. We have found that a telephone approach with a
written follow-up can be very effective.
The larger the organisation, the more time consuming the access negotiation process may be. Where you have no existing contact, it may be necessary to make several
telephone calls simply to establish the best person to ensure that your request for access
will be considered. If you use a contact given to you, you will still experience delay while
your request is considered and an interview meeting arranged at a convenient time.
This may take a number of weeks.
More time will be needed for access requests that are more complex. You may wish to
get data in a variety of ways from across the organisation which, of course, has the
potential for more delays. The official go-ahead from the organisational gatekeeper may
mean that you are ‘in’, but you still have to gain cognitive access: the consent, support
and trust of those participants you will depend upon for your research data.
One final note to emphasise is that the worst may happen and you receive a refusal.
Not only is it essential to have a plan B (or even C and D!), but make sure the necessity
to enact plan B does not seriously jeopardise your overall timescale. Following experimentation, psychologists recently have concluded that time passes quickest when we
are busy. Many of our students have told us that this is not true: it passes like lightning
during the time allocated to a research project!
Strategy 4: Make your written request professional
Putting your request for research access into writing is highly advisable. Unless you read
from script, it is unlikely that you could achieve the same level of precision in a spoken
telephone request.
There are several reasons why it is advantageous to put your request in writing in a
clearly written manner. Among these are:
1 It allows people to be aware precisely of what will be required from them. This is
important because asking them for their help without being clear about your needs
will probably lead them to be wary, since the amount of time you require from them
may be more than they may reasonably expect to give.
2 It gives you the opportunity to set your request in context. You can explain the overall purpose of the research, the reason you are doing it, the university and department in which you are studying and the course you are completing.
3 It gives you the opportunity to enhance your credibility by presenting a professional
image through a well-written and carefully considered request.
Your written request should outline briefly:
the purpose of your research;
how the person being contacted might be able to help;
66 Chapter 3 Managing the research process
the demands being made of those taking part in the research;
a guarantee of anonymity (where appropriate);
what you will do with the information you get from each person involved, including
any intention you may have of sharing the information during or after the data collection stage;
your contact details so that the person can reply to you.
Should your written request be an email or a traditional letter? The advantages of an
email are clear, but don’t discount the value of a letter. The impact of the letter may be
greater as it is more unusual these days; it is much less easy to delete or mislay; and it
gives you greater opportunity to personalise and, therefore, impress. Whichever method
you choose, it is essential that your communication achieves the highest standard of
which you are capable. This warning is especially appropriate in the case of emails. We
all know how easy it is for normal writing standards to drop when we send emails and
how tempting it is to hit ‘send’ without double- and triple-checking in the same way
that we may do with traditional written forms of communication. Don’t be tempted.
The person reading your email may (like us!) be a stickler for correct written English, and
may reject your request on these grounds alone (see Research in practice 3.2.
Writing your research request
Below is a copy of the letter Nazaneen sent to the National Federation of Self Employed
& Small Businesses Limited regarding her planned research which we explained in
Research in practice 3.1.
Research in practice 3.2
Dear Ms Armfield
Further to my telephone call to your assistant, Stephanie Dixon, last Friday April 11th,
I would like to introduce myself and explain the reason for me contacting you.
My name is Nazaneen Ghorbani. I am a part-time business studies undergraduate at the
University of ………………. I also work full-time in one of the major banks. As part of my
course I have to complete successfully a research project. I am very interested in the issue
of cashless pay, both the technology that facilitates this development and the way in
which it has the capacity to change consumer behaviour. This is the subject of my project.
In my review of the large amount of research material available on this topic it seems that
the widespread adoption of cashless pay has particularly large important implications for
smaller retailers. I would like to gain a closer understanding of the attitudes of small
retailers to cashless pay by speaking to you and a number of your members. I hope you
are able to help me in this regard.
The details of my research plan are in the attached document called ‘Nazaneen Ghorbani
Research Proposal’.

3.4 Six strategies for making sure that you get the organisational access you want 67
Strategy 5: Work hard to ensure there are no concerns about the
way in which you will use the information
You will have noticed that one thing that virtually all organisations have in common is
that they like to present a positive front to the world! Even if the news is bad, they work
hard to put a less negative aspect on that news than perhaps is justified. Why else would
large organisations spend millions on public relations specialists? There is a message
here for those of us who want to pursue organisational research. You are unlikely to
gain research access if the subject of your research can be perceived as negative. So an
inquiry into the reasons for an organisation’s market failure in a particular product area
is to be avoided. However, you may learn from the public relations experts and lend a
positive emphasis to this by, for example, using the organisation of your choice as a
context for an inquiry into the changing nature of the demand for a particular product
area and what may be learned. Your choice of language in the written request will have
to be very carefully considered to ensure that no hint of negativity is suggested.
A more predictable area of concern is that of anonymity. This relates to that information which is given to you and what you do with it once it is given. Some topics, some
organisations and some people will be extremely sensitive about anonymity. While this
occasionally may be hard to understand, the point remains that it is their information,
and you must respect their wishes.
The introductory email or letter offers you the opportunity to give a guarantee of
confidentiality in writing at the time of making the request for access. This is a good
time to make the initial guarantee, as confidentiality may be uppermost in the minds of
the managers who will consider your approach. But do think of this as only the initial
guarantee. As you meet more people, you will need to repeat any assurances about confidentiality. This can be done, for example, with an assurance that any information
given to you will not be attributed to any individual. It is essential that you honour this
assurance, not only for your credibility, but for those researchers who may follow you.
At the end of the document I suggest a set of topics/questions for the interviews I plan
to conduct, which I hope will be valuable for you in thinking about this prior to my visit.
I assure you that I will take no more than one hour of your time and that the information
you give me will not be credited to you in any report that I subsequently write without
your approval.
I very much hope that you are able to help. If so please email me at [email protected]
or call me on 07123 456789. I will then contact Stephanie Dixon to fix an interview at a
time and place suitable for you.
I hope we are able to meet in the not too distant future. Meanwhile, thank you for
reading this letter.
Yours sincerely
Nazaneen Ghorbani

68 Chapter 3 Managing the research process
It is quite common for the organisation to ask you to not place your written report in
the university library, although you may not have identified the organisation or any
individuals. You must check the course requirements here with your supervisor, but whatever the regulations, the confidentiality wishes of your research organisation come first.
Strategy 6: Underline your credibility!
Perhaps this sixth strategy is a result of the efforts you have put in making your request as
professional as possible. The more credible you seem, the more likely you are to receive a
favourable response to your request for research access. But there is one point we haven’t
mentioned so far which we think will really underline your credibility: find out as much
as possible about the organisation before requesting access. The more you know about
the organisation’s context, such as their products and services, markets, competitors,
trading position, current challenges and development plans, the better. This will really
impress the person you’re approaching, particularly if that knowledge blends thoughtfully with the information you are seeking and your research question(s) and objectives.
In the end, do what’s possible
Like many research considerations, getting access to your research organisation is a balance between what is ideal and what is possible. OK, so you want to make sure that you
get a representative sample, conduct your interviews in a uniform way and collect sufficient data to ensure the answers to your research question(s) and objectives are valid
and reliable. But life isn’t perfect. People will deny you access to information, limit your
interview time, lose your questionnaire, go on holiday, and leave the organisation in
the middle of your research. You are at the mercy of events. You can manage much of
the process, but you can’t control the uncontrollable!
Table 3.1 Checklist of points to follow to increase your chances of getting access
to your chosen research organisation
•  Be clear about the overall purpose of your research project.
•  Write your research question(s) and objectives.
•  Use existing contacts where possible.
•    Consider using your work placement organisation (if appropriate) as a setting for your
research project.
•    Approach relevant appropriate local and/or national employer, or employee, professional
or trade bodies to see if they can suggest contacts.
•    Make a direct approach to an organisation to identify the most appropriate manager.
•    Think about the possible benefits for the organisation, should access be granted to you.
•    Offer a report summarising your findings.
•    Allow yourself plenty of time for the entire process.
•    Allow sufficient time to contact intended participants and gain their consent, once access
has been granted.
•    If you make your initial request for access by telephone, follow this with an email or letter
to confirm your request.

3.5 Managing yourself 69
•    Make sure the construction, tone, language, spelling, grammar and presentation of an
introductory email or letter are all likely to persuade the person to help you.
•    Consider how you will address concerns about the amount of organisational time you would take up.
•    Ensure you have considered any sensitivities concerning your research topic.
•    Assure participants or respondents you have recognised any needs for confidentiality and/or
anonymity.
•    Think about a range of contact methods for potential participants to use to reply.
Table 3.1 Continued
We now move on to consider the management of those components of the research
process over which you have more control: managing yourself, your supervisor and
your university.
Managing your time
We have already mentioned in this chapter elements of time management, particularly
those that refer to the process of gaining access to your research organisation. Here we
consider time management more widely, both in terms of your research project schedule and the allocation of time to your research.
It is important that you complete a research project plan at the beginning of your
research. Indeed, this may be a university requirement, as it helps your supervisor assess the
viability of your research proposal. It’s useful if you divide your research plan into stages.
This will give you a clear idea as to what is possible in the given timescale. But don’t forget
that however well your time is organised, the whole process seems to take longer than you
planned. An example of the sort of schedule you may develop is shown in Table 3.2.
3.5 Managing yourself
Table 3.2 Research project schedule
Task To be completed by

Generate list of research ideas
Choose research topic
Undertake preliminary literature review
Define research questions and objectives and submit research proposal
Main literature reading
Literature review written
Methods chosen and draft method chapter written
Fieldwork commenced
All data collected and fieldwork notes completed
10.10.2016
1.11.2016
1.11.2016
10.11.2016
10.12.2016
31.12.2016
10.01.2017
20.01.2017
01.03.2017

All data analysed ready for draft findings and conclusions
chapters to be written
20.03.2017

Final draft submitted to supervisor
Final submission
15.04.2017
30.05.2017

70 Chapter 3 Managing the research process
Even if the inevitable happens and you find that you have some slippage in your
schedule, at least you know what you are slipping from! Not having a schedule is simply
unthinkable.
Just as important as the project schedule is your personal timetable which shows the
amount of time you are going to devote to your research. Here you will have to be realistic. You have other demands. Other modules require lecture preparation and assessment completion. But it is all too easy to fall into the trap of allocating time by ‘what’s
the next most important deadline’. Of course, there is always another deadline looming, but the trouble for your research project is that the big deadline of final project
submission is months away for most of the year and it’s forever being left until later. It’s
best to schedule some time regularly, say each week, to do something towards your
research. Do this at the start of the year and get into the habit of it. The beauty of personal timetables is that they become part of a ‘self-contract’. If you meet your contractual obligation, you can feel pleased with yourself; if you don’t meet it, make sure you
feel suitably guilty! There is no substitute for getting into a routine. You will probably
have heard this before when you started university, but it’s the same for doing research,
or for us in writing this text. Get into the habit of doing something towards your
research right from the start on, for example, Monday afternoons in the library. It’s
always easy to find other things to do when the deadline is not pressing. Just try not to
be like Oscar Wilde, who could resist everything except temptation!
One final point about time management. We have known students who have been
perfectionists and ‘just do enough to pass’ types. Our advice is to be neither. The perfectionist runs into time management problems because of the need to polish everything
until it dazzles. This is great if there’s time: but there never is. ‘Just doing enough’, on
the other hand, can often lead to not doing enough and the necessity to re-submit the
report.
Keeping up your motivation
You are likely to be doing your research project for an extended period, so expect your
motivation to vary over that period. You may start and end on a high, but there will be
times in between when you can’t see your way forward. There are ways in which you can
overcome this. Let’s look at three of these.
First, set for yourself short-term goals. Sports coaches are renowned for using this
technique to improve participant performance. You can distinguish between long-term
goals, for example, breaking the world record or successfully completing your degree;
intermediate goals, such as lowering the 100-metre sprint time by one second or getting
all your fieldwork done on time, and short-term goals. Short-term goals may concentrate on each training session, or, in research project terms, the completion of a set of
notes at the end of a two-hour session reading a particular article. Do remember that
it’s important that your short-term goals are SMART (specific, measurable, achievable,
realistic and timely) (see Chapter 1, section 1.7).
Second, while we are on the subject of goals, think about setting some fresh learning
goals. Learning how to conduct an interview effectively; use a new statistical analysis

3.6 Managing your supervisor 71
package or how to sift, label and categorise qualitative data in a systematic way will not
only add to your repertoire of business skills but give you a sense of achievement.
Third, do keep focused. You can do this by keeping in touch with your supervisor and
talking through your progress. This can be quite motivational, either because you are
further on than you thought, or in giving you a refreshed sense of direction if you have
lost your way. Another important way of keeping focus is to review regularly your
research question(s) and objectives. It’s so easy to wander away from the main point of
your research into all sorts of interesting diversions. A regular review of progress against
your research question(s) and objectives is a vital way to keep focused. OK, so this may
involve some revision of your question(s) and objectives, but at least this will mean that
you’re on track.
Keep in touch with individuals who can help
We have already mentioned the need to keep in touch with your supervisor. Don’t forget there are other people who can help. It is useful to talk to these people regularly.
Among these are those student colleagues who helped you think about your research
topic in the first place and family and friends. It may also be useful to talk to the specialist librarian in your research topic area. This expert may be able to point you in the
direction of new resources and to suggest different perspectives on your research.
When you look back on your time at university and remember your research project, it’s
certain you will remember your research supervisor. Usually, this is for the right reasons,
because the relationship has been a fruitful one where both you and your supervisor
have had your expectations met. But, of course, it can all go horribly wrong. Fortunately,
this happens rarely. But it can happen, so in this section we look at the expectations that
you should have of your supervisor, and those that your supervisor should have of you.
Being aware of these, and acting accordingly, should mean that your eventual memories
of your supervisor are as favourable as those that we hope our students have of us!
What should you expect of your supervisor?
Your supervisor is there to advise you at every stage of your research project, from formulation of the project through to completion of the final report. However, do remember that it is your research project, and your supervisor is not there to write it for you.
What expectations should you have of your supervisor? Here are some:
You receive assistance with the selection and planning of a suitable and
manageable research topic
You should not expect your supervisor to impose a topic upon you; it has to be your
choice. But you should be prepared to have your ideas both praised and criticised. You
3.6 Managing your supervisor
72 Chapter 3 Managing the research process
want encouragement, but if your supervisor thinks an idea is a non-starter, you should
be told this.
Your supervisor is sufficiently familiar with the field of research
to provide guidance
This can be quite a problematic area. With the increasing amount of students who require
supervision comes the need of the university to resource this requirement. This sometimes
means that there is not an ideal match between your research topic and your supervisor.
However, do bear in mind that your supervisor will concentrate on the research process
rather than simply on the topic. In our view, the more effective supervisors are those who
are familiar with guiding their students on how to conduct the project rather than those
who are subject-only specialists. However, talk to the supervisors who are specialists about
your research plans. Getting them interested may mean that you end up with both a subject specialist and someone who can guide you successfully through the project.
Your supervisor should be available for consultation and discussion
about the progress of your research
How often you see your supervisor will depend upon several factors, not least of which
will be the norm as it applies in your department. Usually, you will meet during the
early stages of the project as things get going. As you become more confident, and you
immerse yourself in reading or data collection, you would expect less frequent meetings. We usually find that our students value contact with their supervisor in the early
and late stages of their project, when they are setting up and writing up. But do bear in
mind that in most universities, the onus is upon you to initiate meetings, although
your supervisor may contact you to arrange the first meeting. You should check with
your supervisor what the procedure is here.
At the end of each meeting with your supervisor, it is good practice to arrange the
time of the next meeting and to agree what you should have done by that meeting. You
then have the basis of an agenda for the next meeting. There may also be a procedure
where a record is kept of the content of the meeting.
Your supervisor should respond to the work that you have completed
If you have sent your supervisor written work, as agreed between you, it is reasonable to
expect that your supervisor will respond to this. You should expect to have helpful suggestions about how the work may be improved.
Your supervisor should point you in the direction of facilities
or research materials
Again, the onus is upon you to make arrangements to negotiate access to organisations, or
libraries where you need secondary data. But it’s reasonable to expect that your supervisor
will have some ideas as the sort of data that may be of help to you and where this may be
available. As you make progress in your research, you will probably find that you know
more than your supervisor about your research topic. However, in the early stages you are
hungry for ideas, and it’s to be expected that your supervisor will provide some of these.

3.6 Managing your supervisor 73
What your supervisor should expect from you
You should, in conjunction with your supervisor, develop a plan and
timetable for completion of the stages of your research project,
and meet all deadlines
Meet all deadlines, you say! Well, yes. But for all sorts of reasons which we touched on
earlier in this chapter, some delays may not be in your control. But your supervisor will
expect you to pay attention to the set deadlines and make your best attempt to meet
them. Treating deadlines in a trivial way is a sure path to a hurried and ineffective conclusion to the project. The more you have to rush at the end, the less value you can
expect from your supervisor’s comments at that vital late stage when you are drafting
your report.
You should meet with your supervisor when arranged and report fully
and regularly on progress and results
It would be good to say that it’s rare for students to not turn up for an arranged meeting,
or to show up having not done the work that was agreed at the last meeting. But, unfortunately, it’s not that rare. You will want to get the best out of your supervisor, so make
sure that you demonstrate that you are businesslike and do what you have agreed to do.
That way, your supervisor will respond in a similar way and a fruitful relationship will
develop.
You should give due consideration to the advice and criticisms received
from your supervisor
It would be foolish to ignore criticism from your supervisor. But that’s not the same as
saying that you should follow slavishly every word. Indeed, when your confidence
grows as your work progresses, you may choose to listen to criticism but decide, after
careful thought, to stick to your guns. After all, it’s your project and you will have to live
or die by the final outcome. But bear in mind that your supervisor has probably seen
many similar projects in the past and is likely to be one of the people assessing your
project report, so do consider carefully any critical comments you receive.
You should keep yourself up to date on the subject of your research topic
The likelihood is that you will write your critical literature review early in your project’s
lifecycle. Try to avoid the temptation of then closing your mind to this aspect of your
work. There may be important developments in your topic area during the course of
your work, so don’t miss out on these. Ignoring recent news in some highly topical subject areas, such as new technological developments, will damage the credibility of your
work and threaten the success of your final submission.
You should keep in touch with your supervisor and make yourself available
for regular meetings at mutually convenient times
Most lecturers will tell you that the students who keep in touch with their supervisor
are the ones who usually complete the research project successfully, while those they

74 Chapter 3 Managing the research process
rarely see are on course for failure! This may be because the more diligent students
usually are successful. But we like to think that research supervisors can pass on a wealth
of experience and know-how which you would be foolish to ignore.
At the beginning of your course, or research project, you should receive from your programme leader a guide which will specify such items as aims, objectives and learning
outcomes of the research project. The guide should also list the relevant regulations for
the completion of the project. These regulations will cover such topics as any requirements to attend seminars, meet specific deadlines, submit the written report in a specified style and, of course, the academic standards you should meet in order to be awarded
a pass for your project. It is this last aspect of the university’s requirements that we concentrate upon here.
The learning outcomes of any research project summarise much of what we cover in
this book as they emphasise the process of doing research. Table 3.3 is a good example
of a statement of such learning outcomes.
3.7 Managing your university
Table 3.3 Example of a statement of learning outcomes for research project module
Upon completion of this module, students will be able to:
(a) show resourcefulness in the sourcing and selection of material;
(b) demonstrate the ability to exercise sound judgement in the selection of material;
(c) engage critically with the chosen material;
(d) organise complex information in such a way that data collected is integrated with
pre-existing material;
(e) structure clearly and logically a substantial piece of work;
(f) produce a clearly defined and usable outcome within a specified timescale;
(g) demonstrate a level of expert knowledge in a particular subject or issue;
(h) manage their own efforts effectively.
At the end of the successful completion of the module, the student will have completed a submission in accordance with the course and thus fulfilled the aims outlined
above.
You should also pay particular attention to the assessment criteria, which should be
contained in your research project guide. Table 3.4 is a good example of clearly defined
criteria which you can use as a checklist as you work through the research process and,
in particular, write your final report.
The criteria specified in Table 3.4 should come as no surprise to you, as they may
well be similar to criteria you have come across earlier in other modules. But do use

3.8 The ethics of doing research 75
Table 3.4 Example of marking scheme guide for research project final report
Criterion Strong Weak
Definition of
aims and
objectives
Clear and informative definition of the
aims and objectives of the research or
study topic/problem to be investigated.
Absent or weak and cursory
description of research topic/
problem to be investigated.
Method and
methodology
Correct selection of and justification for
chosen method. Full understanding of
values and limitations of method. Clear
rationale and understanding of
limitations.
No justification for selected
method. Inadequate data collection.
No evidence of understanding of
method and limitations.
Understanding
and coverage of
the literature
Excellent understanding and insightful
knowledge of the subject matter.
Comprehensive expert account of
topic. Fully referenced using Harvard
referencing.
No understanding of the subject.
Confused thinking and inadequate
knowledge. Limited and inadequate
referencing.
Critical analysis,
examination and
presentation of
findings
High-level analysis using appropriate
techniques. Thorough examination of
results to present clear findings linked
to evidence.
Weak and unacceptable analysis.
No critical evaluation of results.
Findings presented not linked to
evidence.
Conclusions
and/or recommendations
Logical and insightful conclusions
based on findings presented.
Recommendations for action based on
conclusions and findings and relevant
to the context of the study.
No attempt to present conclusions
and/or recommendations.
Structure and
presentation
Excellent layout. Conforms to all stated
requirements. Clear, logical writing
style with correct use of English. Clear
presentation of all tables, figures, etc.
Unacceptable layout in terms of
structure and logical argument. Key
requirements ignored. Inappropriate
use of English. Serious deficiencies
in presentation.
Source: Developed from the University of Plymouth (2016).
Your university will expect you to conduct your research ethically and has hopefully
made you aware of the ethical responsibilities you have while conducting your research.
3.8 The ethics of doing research
these for your research project. We even urge you to have a session with your supervisor
as you approach the writing-up stage to discuss the precise meaning of these criteria in
relation to your written project report.
Definition
research ethics: the appropriateness of the researcher’s behaviour in relation to the rights of those
who become the subject of a research project, or who are affected by it.

76 Chapter 3 Managing the research process
Let’s be clear about what we mean by ethics. Generally, ethics means ‘standards of
behaviour that guide the moral choices we make which govern our behaviour and our relationships with others’. Since virtually all research using primary data, that is data which we
collect specifically for the purposes of our research, involves our relationships with others,
this is clearly something we shouldn’t ignore. Or can we? Well, even if you wanted to act in
an amoral way towards your research participants, the rules of your university will stress the
need for you to behave in an ethical way in the conduct of your research (see Table 3.5).
Table 3.5 Example of code of practice for ethical standards involving human
participants
•    No research should cause harm, and preferably it should benefit participants.
•    Potential participants normally have the right to receive clearly communicated information
from the researcher in advance.
•    Participants should be free from coercion of any kind and should not be pressured to
participate in a study.
•    Participants in a research study have the right to give their informed consent before participating.
•    Where third parties are affected by the research, informal consent should be obtained.
•    The consent of vulnerable participants (e.g. children) or their representatives’ assent should be
actively sought by researchers.
•    Honesty should be central to the relationship between researcher, participant and institutional
representatives.
•    Participants’ confidentiality and anonymity should be maintained.
•    The collection and storage of research data by researchers must comply with the Data
Protection Act 1998.
•    Researchers have a duty to disseminate their research findings to all appropriate parties.
Source: Oxford Brookes University (2016). Full version available at: www.brookes.ac.uk/Documents/Research/
Policies-and-codes-of-practice/ethics_codeofpractice/
In this section, we explore some of the ethical issues you may encounter through the
different stages of your research. Ethical considerations impact upon how you decide
upon your research topic; design your research and gain organisational and individual
respondent or participant access; collect your data; process, store and analyse your data
and write up your research findings. At all these stages you will need to ensure that the
way you design your research is both methodologically sound and morally defensible to
all those who are involved.
Here we consider the ethical issues at three main stages of the research process:
research design, data collection and reporting. Of course, these issues don’t fit neatly
into each of the main stages. Nonetheless, we cover some of the main ethical points
which we then summarise at the end of the section in Table 3.6.
What are the ethical questions you should consider at the research
design stage?
Getting participants’ and respondents’ informed consent
The main issue to consider at the design and access stage of your research involve the
issue of respondent or participant consent. Look again at Table 3.5. Following the

3.8 The ethics of doing research 77
principles here means that your respondents and participants should understand what
the research is about and what is expected of them, should be free from coercion of any
kind and should not be pressured to participate in a study. Coercion is unlikely to be
the issue when you are approaching a senior manager as an organisational gatekeeper,
but it may be relevant when the senior manager arranges for you to interview participants more junior than that manager. You should be particularly sensitive to this situation. It is extremely unlikely that such people will refuse to cooperate, even if a
feeling of being coerced by a senior manager exists. However, you will need full support from your participants rather than grudging acceptance. Full support is more
likely if you give assurances of confidentiality before the interview in a bid to secure
informal consent.
The ethical situation relating to data collected by questionnaire is rather more
straightforward. The return of a completed questionnaire by a respondent is such that
completion by itself implies consent. However, this is not the same as giving you consent to use the data collected in any way you think fit. You should bear in mind that
when respondents agree to participate, this does not necessarily imply consent about
the way in which the data provided are subsequently used. You should still tell them in
the introduction to the questionnaire how you will use the data. Don’t ignore another
of the principles in Table 3.5: that honesty should be central to the relationship between
you and your respondents. Consider the example of the phone caller who calls you at
the most inconvenient time posing as a researcher but who, in reality, is a salesperson.
This is infuriating, and quite the opposite of the standards you should be aiming at!
Consent – not coercion!
Conducting research in your own employing organisation presents different concerns.
You may be tempted to apply pressure to others (colleagues or subordinates) to cooperate. They may cooperate from a feeling of social obligation to you. But the same effort to
obtain informal consent applies as discussed before. And like all respondents and participants, you should emphasise their right to withdraw from your study at any time.
If you are being asked by the organisation to conduct your research project, you have
a right not to be forced to choose a research project in which you are uninterested. This
prejudices a valuable learning opportunity for you. It also threatens one of the most
important ethical obligations you have to all concerned – to produce research of the
highest possible quality.
Your ethical obligation to use the time you have been granted
by the organisation is a way that is of benefit to the organisation
If you are using an organisation as your research context, you will have an ethical obligation to use the time you have been granted by the organisation in a way that is useful
to the organisation. That is not to say that you will necessarily tell the managers what
they want to hear. But even if your conclusions are unfavourable, then the organisation
has a right to expect that you frame your conclusions in a positive and constructive
way. At the same time, you have a right not to be coerced by the organisation’s managers to misrepresent the data which you have uncovered so as to portray the organisation
positively.

78 Chapter 3 Managing the research process
Respondent and participant vulnerability
Clearly, some topics carry a greater risk of a breach of ethical standards than others. A
brief look at Table 3.5 shows that the standard of seeking the consent of vulnerable participants or their representatives is one that is potentially sensitive. On the face of it,
this may relate more closely to schoolchildren, medical patients or social services clients. Yet, vulnerability may apply to employees whose jobs are changing or being made
redundant; or it may relate to sensitive areas such as poor employee performance. If
your research topic has obviously sensitive implications, you will need to pay attention
to the ethical dimension in your research proposal.
What are the ethical questions you should consider
at the data collection stage?
The importance of confidentiality
Often confidentiality is an important condition that your research organisation makes
when granting you research access. Organisational gatekeepers often insist that you do
not disclose the organisation’s identity in any way. We deal with how to cope with this
in your report in the sub-section on ‘confidentiality and anonymity at the reporting
stage’ later in this section.
It is often more important to guarantee confidentiality to individual respondents
and respondents. This is particularly important in relation to names, addresses and personal data that may allow them to be identified. Not only is the issue of confidentiality
important, but harm may be done to individuals by information that you have obtained
being attributed to them. You may, for example, have gained the trust of a participant
to the extent where you are told the individual’s career plans, something that may be
better kept from management. Again, as with the organisation, do make sure that your
participants’ identities are not revealed accidentally in anything you write. You can use
the same ‘participant A’ method, or adopt a style such as ‘one participant told me . . .’
The key point here is not to break your promise of confidentiality.
It’s not just in reporting your data that you may reveal the identity of your participants. This can happen in interviews. As you collect more data, you may develop
themes which have emerged during the process. Developing these themes may mean
that your participants can guess what previous participants have said from the questions you are asking. This can lead to participants indirectly identifying the person
responsible for raising the earlier point that you now wish to discuss with them.
The confidentiality points mentioned earlier relate mainly to the collection and
reporting of qualitative data. Such issues are equally important when it comes to the
collection and reporting of quantitative data. Here your main concern is to report
aggregated data, in order to establish the patterns that emerge. So making quantitative
data anonymous is less problematic. However, you still need to take care. Presenting a
table that reports agreement with the questionnaire statement ‘This organisation is a
great place to work’ is not anonymous (and could cause problems) if it showed that the
two senior managers in a named department both ‘strongly disagreed’.

3.8 The ethics of doing research 79
Causing harm through the collection of primary data through interviewing
In addition to confidentiality, conducting interviews raises other ethical considerations. The potential to harm your respondents can arise in many ways. We deal with
three here.
First, it is important to be sensitive to any reluctance that your participant may have
to answer your questions. As your confidence in interviewing builds, it may be easy to
start pressing participants for answers without realising that what they are experiencing may be quite stressful for them. It’s important to make clear to participants that
they have a right to refuse to answer any question, for any reason. It doesn’t happen
often, particularly when you have built rapport with the interviewee, but it can.
Second, do be careful that the questions you ask are not likely to be harmful to your
participant. These may relate to personal circumstances or be of such a nature that they
humiliate your interviewee. Such questions may be quite unintentional. They may not
form part of your planned interview schedule. But it’s easy for a follow-up question to
come out in a way that had not planned.
Third, do be mindful of the circumstances of your participant. Simple things such as
setting a time, location and interview duration that is convenient to your participant is
important. If the interview participants are quite junior in the organisation, it’s important that you stick to agreed arrangements as they are likely to have less time flexibility.
Be on guard for signs of discomfort because, for example, you’re running over the agreed
time. Not only is this discourteous, but, once again, you may be causing stress and
inconvenience to your participants.
Causing harm through the collection of primary data by observation
Should you wish to collect some or all of your data by observation, you will face particular ethical problems. Let’s say, for example, that you want to observe the behaviour of
airline cabin crew members as they deal with customers, in order to establish the extent
of the engagement that they demonstrate. Central to this is the choice you have over
whether you should disclose your purpose to those crew members you intend to
observe. Given everything that has been said so far in this section, this seems a clear-cut
decision: you should disclose your research purpose.
However, you then face the problem of ‘
reactivity’ (or ‘observer effect’), which is a
particular problem with observation. This happens when those you are observing adapt
their behaviour as a consequence of your observation. Obviously, this threatens the
reliability and validity of your data. So you have a choice: you disclose your purpose and
seek approval of the organisation and the crew members in the normal way, or you go
ahead and do your observation covertly.
Let’s say that you seek approval of the organisation and the crew members and then
you are refused access. You could carry on covertly, in your role as a customer, or
Definition
reactivity: reaction by research participants to any research intervention that affects data reliability.
80 Chapter 3 Managing the research process
abandon the plan. If you decide to carry on covertly, then those being observed may
discover your actions with the consequence that you must explain what they would see
as your deceit. It is our view that researchers should not practise deceit at any time as
this is a clear violation of the principles of ethical research.
Causing harm to through unauthorised use of secondary data
It’s not just through the collection of primary data that you can cause harm. The collection of secondary data also raises ethical issues. Say, for example, you want to establish a
sample of customers from whom you want to collect data on service satisfaction levels.
This may involve a study of customer databases to form the sample. From the databases
you may have personal details of many individuals who have not consented to you having this information. It’s important that you treat this information in the strictest confidence and don’t use it in any way that might cause harm to these individuals.
It’s also important that you are honest, stick to the declared purpose of your research
and not use secondary data in any way other than that which you originally intended.
Data collection and honesty
Having raised the topic of honesty in the previous sub-section, we touch here upon the
most apparent dishonest behaviour in data collection, that of making up the data. This
can be blatant, such as completing questionnaires yourself rather than delivering them
to intended respondents. On the other hand, it can be more subtle, where you decide to
include only responses which suit you for some reason and ignore those that do not suit
you. Not only do you have an ethical duty to report your findings honestly, but your data
will be unreliable and invalid if you have practised dishonesty in your data collection.
Ethical issues and data collection using the Internet
The emergence of the Internet has presented many new opportunities for conducting
research, including some in the business arena. The main ethical consideration is likely
to revolve around the issue of consent. The potential topics here are so varied that it is
well to check with your university’s research ethics guidelines if you think that your
research topic may breach any ethical guidelines. An example of the sort of topic which
may be involved is shown in Research in practice 3.3.
Ethical issues in research
Most organisations play an active part in helping the environment by making the decision to go green. One of the most visible ways of doing this is through an office recycling programme through which employees can help reduce environmental waste by
recycling office paper, cans, bottles and other materials.
Many organisations go beyond these basic recycling initiatives and encourage
employees to find other ways to save energy. Such initiatives may include such things as
Research in practice 3.3
3.8 The ethics of doing research 81
Compliance with the data protection legislation
If you are studying at a UK university, the principles of the Data Protection Act 1988
have implications for your research project if it involves the processing and storage of
personal data. If you are studying elsewhere, it is likely that similar data protection
legislation will exist that will have implications for your research project. It is therefore as well to be aware of these before you reach this stage of your research project. In
car-sharing for travel to work journeys, negotiation of deals with local motor suppliers to
supply cars with hybrid engines and the consumption of vegan food in the company
restaurant.
Ronnie was very enthusiastic about the green movement. It permeated many aspects
of her everyday life. So it was natural that she wanted to make this the subject of her
undergraduate research project. Ronnie had read that a leading maker of computers of
all sizes had told its employees that turning off their computers for one hour each day
could save the company $1 million per year in energy costs as well as help the environment. So she concluded that ensuring that all office computers were switched off at
night, when typically they would not be in use, would be similarly environmentally
friendly.
Ronnie was doing her student placement in a large pharmaceutical company. The
company had a sophisticated corporate responsibility policy which included a welldeveloped section on environmental action. The policy urged employees to switch off
their computers when they finished work at the end of the day. Consequently, Ronnie
devised a research design that involved discovering the extent to which employees were
committed to such an initiative and the degree to which the level of commitment was
translated in action.
The executive to whom Ronnie spoke about the matter was quite willing to offer
help, although she expressed some concern about the potential for ‘snooping’ which
such research may entail. She was in agreement with the testing of employee attitudes
but unhappy with the notion of ‘checking up’. Ronnie understood this concern but felt
that simply establishing employee attitudes to computers being switched off at night
was not enough: it was the sort of issue about which most people would indicate agreement that it was ‘a good thing to do’. But Ronnie was keen to discover whether such
agreement led to action; this, she thought, would be the true measure of commitment.
Ronnie explained this stumbling block to her research supervisor, who agreed with
her point about the potential difference between words and action, but he too sympathised with the company executive’s view about the potential for ‘snooping’ that such
research may involve. Checking whether computers were switched off at night may well
breach the university’s code of research ethics which stated that ‘participants in a
research study have the right to give their informed consent before participating.’
Ronnie thought hard and long about her supervisor’s strong recommendation to
think of another research design which would receive the company’s blessing and not
fall foul of the university research ethics committee. She was still committed to research
on an environmental topic, but reluctantly agreed to think of another research design
which would be consistent with ethical principles.

82 Chapter 3 Managing the research process
the United Kingdom, the principles mean that anyone collecting personal information must:
fairly and lawfully process it;
process it only for limited, specifically stated purposes;
use the information in a way that is adequate, relevant and not excessive;
use the information accurately;
keep the information on file no longer than absolutely necessary;
process the information in accordance with your legal rights;
keep the information secure;
never transfer the information outside the EEC without adequate protection.
Source: Gov.UK (2015).
What are the ethical questions you should consider
at the reporting stage?
Confidentiality and anonymity at the reporting stage
Earlier in this section we mentioned how organisational gatekeepers are often very
reluctant to allow the organisation to be named in anything you write as a result of your
research. You can overcome this by referring to the organisation as, for example,
‘Organisation A’, etc. But sometimes organisations are not so sensitive. If the research
topic is not controversial in any way, then confidentiality may not be an issue, and the
organisation can be named. If the organisation is particularly sensitive about being
identified, you must ensure there are no ‘clues’ in anything you write, so you will need
to conceal details such as the organisation’s location, products and market if these
details are likely to allow readers to guess the organisation’s identity.
If the organisation is named in your report, then it is almost certain that you will
need to let them read your work to understand the context within which they will be
named. Indeed, even if the organisation is made anonymous, managers still may want
to see the report to ensure that you have honoured any promise of confidentiality.
As well as the organisation, you have an ethical responsibility to protect individual
respondents’ right to anonymity in your project report, unless, of course, you have their
explicit permission to do otherwise.
Your ethical obligation not to report conclusions that would be harmful
to your participants
Earlier in this section we commented upon the ethical obligation you have to use the
time you have been granted by the organisation’s managers in a way that is of benefit to
the organisation.
Another ethical consideration is the use that may be made of the conclusions you
draw in your project report. Clearly you face a dilemma if information you have been
given freely by cooperative respondents is reported by you in such a way that it can be
used to harm them. This is particularly the case if you have their permission to name
them in your report. It may be less problematic if your respondents are made

Summary 83
anonymous. If you feel at the beginning of the data collection stage that your results
may point to conclusions that could result in actions which disadvantage your respondents, then it will be more honest to tell them this. It may mean that you are not given
access to all the data you wish for, but at least you will be behaving honestly.
Table 3.6 Guidelines for action to ensure you observe correct ethical standards
•    Use your university’s, or professional body’s, research ethics code of practice as a guide at all
stages of your research.
•    Anticipate potential ethical issues that will affect your proposed research and plan to
overcome problems.
•    Always get informed consent from your participants and respondents through the use of
openness and honesty at all times.
•    Don’t make dishonest promises about the likely benefits of your research.
•    Respect your participants’ and respondents’ rights to privacy at all stages of your research project.
•    Be honest and objective about your data, both in the way you collect them and their analysis.
•    Consider how you will use secondary data in order to protect the identities of those who
contributed to their collection or who are named within them.
•    Don’t forget that interviews mean that there is greater scope for ethical issues to arise.
•    Don’t refer to information gained from a particular participant when talking to other
participants, as this may allow identification of the individual with possibly harmful
consequences to that individual.
•    Be very careful about the conduct of covert research and consider its ethical implications.
•    Make sure that you honour the promises you give to all participants and respondents
regarding the confidentiality of the data obtained and their anonymity.
•    Ensure you comply with all of the data protection legal requirements.
•    Be aware of the complex ethical considerations in using the Internet and email to ensure
high ethical standards are maintained.
•    Make sure you preserve the anonymity of your participants and respondents in your project
report unless you have their permission to do otherwise.
Organisations are less likely to grant you research access if you are from outside the
organisation.
If you are from inside the organisation and wish to do research in that organisation,
you will face other problems such as those concerned with status.
There are strategies you can adopt to ease the process of gaining access to organisations to do research. These include using existing contacts; anticipating the
responses of the organisation’s gatekeeper; ensuring you have sufficient time to
accommodate the inevitable delays; presenting a professional written request and
boosting your credibility through reducing concerns of the organisation.
Strategies for managing yourself through the research process include managing
your time; maintaining your motivation and keeping in touch with individuals who
can help.
Summary
84 Chapter 3 Managing the research process
Strategies for managing your supervisor through the research process include being
clear about your expectations from your supervisor and what your supervisor should
expect from you.
Managing your relationship with your university involves being clear about the
standards expected from you, particularly the academic standards in the final written report.
You must be very conscious of your ethical responsibilities while conducting your
research.
The key ethical principles you should adopt are: not causing harm to research participants and respondents; ensuring people are not coerced into participating in
your research; seeking the consent of participants and respondents; maintaining
high standards of honesty and preserving confidentiality.
Awareness of your ethical responsibilities applies to all stages of the research process.
Think about the data you will need to answer your research question(s) and meet
your objectives. Assuming you need access to an organisation to get some or all of
these data, make a list of possible obstacles to gaining access. Then, using the checklist of points to follow to increase your chances of getting access in Table 3.1, plan
how these obstacles might be overcome.
Draft a request for research access to be sent to your research organisation’s gatekeeper. Your request should focus upon the purpose of your research, the demands
being made of the participant and what you will do with the information you obtain.
Draw up a ‘contract’ to discuss with your supervisor which should include the expectations you have of your supervisor and the expectations you think your supervisor
should have of you.
Get hold of a copy of your university’s code of research ethics. Note those aspects of
the code you that you feel are relevant to your research and list the implications for
action you need to take to ensure the code is followed.
Thinking about your research process
Gov.UK (2015) Data protection. Available at: https://www.gov.uk/data-protection/the-dataprotection-act [Accessed 15 June 2016].
Oxford Brookes University (2016)
Ethical Standards for Research Involving Human Participants:
Code of Practice.
Available at: https://www.brookes.ac.uk/Documents/Research/Policies-andcodes-of-practice/ethics_codeofpractice/ [Accessed 11 November 2016].
Henley, J. (2016) Sweden leads the race to become cashless society.
The Guardian, 4 June 2016.
University of Plymouth (2016) School of Marine Science and Engineering, Project Marking
Scheme. Available at: http://www.tech.plym.ac.uk/sme/mingproject/MScheme910.pdf
[Accessed 12 June 2016].
References
Chapter 4
Using secondary data
When you started to think about how you would obtain the data to answer your research question, you probably began by considering how to design your own questionnaire or the people
you were going to interview. While this is not unusual, it is a great shame, as it meant you
implicitly ignored the vast amount of data that have already been collected by other people for
other purposes, which could also be useful in answering your research question. In this chapter,
we look at using data that were originally collected for some other purpose, termed
secondary
data
, for your own research. Such secondary data can be contrasted with primary data, which
are data you collect specifically for your research project. We talk about this in Chapter 6.
The number of sources of potential secondary data and the ease of getting access to them
continues to expand rapidly alongside the growth of the Internet. Increasingly, university business schools have subscriptions to online market and financial databases providing access to a
wealth of data that have already been collected and collated from multiple sources. Many
national governments, non-governmental agencies and other organisations allow open access
to the data they have collected, making it available for anyone to use on the Internet. Such
data include the results from large-scale surveys such as national censuses as well as more
specific surveys and research reports. They include numeric data such as official statistics (often
in the form of downloadable tables) and non-numeric data such as government reports, interviews, documents, photographs and conversations. You can also find such data reproduced in
varying levels of detail in other published sources such as quality newspapers as well as on
associated websites. Quality newspapers, for example, contain a wealth of data on organisations and business and management issues in both numeric and non-numeric forms. You can
4.1 Why you should read this chapter
Definitions
secondary data: data used for a research project that were originally collected for some other
purpose.
primary data: data collected specifically for the research project being undertaken.
86 Chapter 4 Using secondary data
easily obtain up-to-date statistics such as share prices. They also sometimes contain transcripts
of interviews with business leaders and politicians (along with commentaries on the content of
the actual interview). When you re-analyse data reported in such media, you are using them
as secondary data, the media being the source where you found your data.
We believe that such secondary data can provide you with fantastic research opportunities, which would otherwise be outside your reach. For example, it is extremely unlikely
that you will have either the time or the financial resources to design and distribute a questionnaire to thousands of potential respondents to collect data for your research project.
Similarly, you’re unlikely to be granted access to interview the chief executive of a large
multinational company. However, through using secondary data such as that already collected through an existing large-scale survey or in the form of a published interview with a
chief executive, you will still be able to gain insights for your own research.
In this chapter, we begin by exploring the questions ‘What forms does secondary data take?’
and ‘Why should you use secondary data?’ We then discuss the pitfalls of using secondary data
and offer advice on how to assess the suitability of secondary data for your research project. We
finish with suggestions of where and how to find secondary data. However, before you decide
to use secondary data for a research project, we would urge you to check your research project
assessment regulations very carefully. For some research projects your university may expect you
to use only secondary data. For other projects you may be required to collect primary data or use
a combination of both primary and secondary data.
Secondary data comes in many forms. As we hinted in the introduction, these include
both
quantitative data, consisting of numbers such as tables of figures, and qualitative
(non-numerical) data
types. The latter include text materials such as organisations’ policies or minutes of meetings, and non-text materials such as video and voice recordings, and images such as photographs. Secondary data can include raw data that have
not been processed, such as actual responses to questionnaires available from a data
archive or the transcript of a television programme interview available on the television company’s website. These will have been originally collected and used for some
other purpose. Here the source of the data is the data archive or the television company’s website. Secondary data can also include as
compiled data, where the data
4.2 Forms secondary data can take
Definitions
quantitative data: data consisting of numbers or data that have been quantified, such as tables of
figures.
qualitative data: non-numerical data or data that have not been quantified, such as text materials,
and non-text materials such as videos, voice recordings and images.
raw data: data for which no processing has taken place.
compiled data: data that have been processed, such as through some form of summarising
or selection.

4.2 Forms secondary data can take 87
presented have either been selected or summarised from the raw data such as a table of
data in a journal article. Here the table contains the secondary data and its source is
the journal article. We find it useful to group the forms of secondary data into three
broad types: survey, documentary (in some form) and multiple source (Saunders et al.,
2016), which are summarised along with examples in Figure 4.1.
Secondary data
Survey Document Multiple source
Snapshots
Examples:
Content of:
• Industry reports.
• Governments’/
European Union
publications
and open access
databases.
• News reports.
• Market and
financial
databases.
• Books.
• Journal articles.
Time series
Examples:
Content of:
• Industry reports.
• Governments’/
European Union
publications
and open access
databases.
• News reports.
• Market and
financial
databases.
• Books.
• Journal articles.
Text
materials
Examples:
• Organisations’
communications such as
emails, letters, tweets.
• Organisations’ websites.
• Reports and minutes of
committees.
• Magazine articles.
• News reports.
• Diary entries.
• Interview transcripts.
Non-text
materials
Examples:
• Media accounts including
television and radio.
• Audio recordings.
• Video recordings
• Images including
photographs, web pages.
Continuous
and regular
surveys
Examples:
• Governments’
surveys.
• Organisations’
surveys.
Ad hoc
surveys
Examples

:

• Governments’
surveys.
• Organisations’
surveys.
• Academics’
surveys.
Censuses
Examples:
• Governments’
censuses.
Figure 4.1 Forms of secondary data
Source: Developed from Saunders et al. (2016).
88 Chapter 4 Using secondary data
Survey secondary data
Our first form of secondary data was originally obtained through a survey strategy, usually a questionnaire. Such data are often made freely available (particularly by public
authorities) either as compiled data tables or, more frequently, as data sets which can be
downloaded and opened in a spreadsheet. Not surprisingly, such data sets are predominantly quantitative, and so you will need to analyse them using statistics and graphs
(section 7.3). These data will have already been obtained using one of three types of
survey strategy:
Censuses are official counts usually conducted by a government or on behalf of a government to meet the needs of government departments, with the aim of collecting
data from every household in a country. Many governments have collected data
about their entire populations for more than a hundred years through a census: for
example, the United Kingdom’s decennial
Census of Population, the most recent of
which took place in 2011. Because they are official, participation in censuses is usually obligatory and so the level of response is usually high. Census data are usually
collected using some form of questionnaire; the precise method and questions used
being clearly documented. You will find such data are widely available and easy to
obtain online and in printed reports.
Continuous or regular surveys are those, other than censuses, that are repeated over
time but do not collect data from an entire population. They include surveys where
data have been collected both on behalf of governments and by non-governmental
bodies such as private organisations throughout the year or annually, often using a
questionnaire or structured interview. You will find that data collected by some of
these surveys, particularly those undertaken by or for government organisations, are
available online. Alternatively reports and associated presentations may be available
(Research in practice 4.1). However, where data have been collected for a commercial
purpose, or is of a sensitive nature, they are unlikely to be so easy to obtain. For example, although many organisations conduct ongoing surveys of their customers’ satisfaction as well as employee attitude surveys, it is unlikely you will find, or be allowed
access to, more than a brief summary of the associated findings to support your
research project.
Ad hoc or ‘one off ’ surveys, as their name suggests, are one-off surveys. Like continuous or regular surveys, these surveys do not collect data from an entire population. Ad hoc surveys are likely to be far more specific than the other forms of
survey in their subject matter. They include questionnaires and structured interviews undertaken by organisations and by independent researchers such as your
lecturers for a specific purpose. However, invariably because of their ad hoc nature,
it is much more difficult for you to locate such surveys, let alone gain access to the
associated data. At best, you’re likely to be reliant on an associated summary
report or journal article.

4.2 Forms secondary data can take 89
Documentary secondary data
Documentary secondary data consist of both text and non-text materials that were
originally collected for some other purpose. As with survey secondary data, it is relatively easy for you to access such data online. This can be done particularly through the
websites of public bodies such as government departments and media corporations
such as the BBC and Sky as well as from video-sharing websites like YouTube®. Indeed,
Private organisation’s regular survey
Jemma’s research objective was to explore the influence of trust and distrust on people’s
behaviours regarding companies. In particular, she was interested in the extent to which
Millennials’ (people born between 1980 and 1995 (Pricewaterhouse Coopers, 2013))
behaviours differed between companies they trusted and companies they distrusted.
When reviewing the literature, she had discovered that a number of authors had referred
to the
Edelman Trust Barometer so she had searched for it using Google. Her search had
revealed that the
Barometer was based on a regular survey conducted annually by the PR
company Edelman. Edelman’s website included links to an archive of summary reports
dating back to 2001 as well to a 70-slide presentation of the global results for the 2016
survey (Edelman, 2016). Conducted in October and November 2015, the research surveyed over 33,000 respondents aged 25–64, asking questions about trust and credibility.
The global results slides included data about the percentage of people engaging in specified behaviours with companies they trusted and companies they distrusted. These percentages formed part of the secondary data for Jemma’s research project:
Behaviour with
distrusted companies
Behaviour with
trusted companies
Refused/chose to buy products/services 48% 68%
Criticised/recommended companies to
a friend/colleague
42% 59%
Shared
negative/positive opinions 26% 41%
Disagreed with others/Defended the
company
35% 38%
Paid more than wanted to buy
products/services
/paid more
20% 37%
Sold shares/bought shares 12% 18%
Behaviour relating to distrusted companies is in italics; behaviour relating to trusted companies is in bold.
Source: Edelman (2016).
Jemma combined these regular survey data with reports from quality newspapers about
the changing levels of trust in companies. Using these data and research findings reported
in the academic literature, she was able to argue that there were significant differences in
people’s behaviours when they trusted companies and when they distrusted companies.
Research in practice 4.1
90 Chapter 4 Using secondary data
although individuals upload much of the video content on YouTube, many organisations also upload material on this site. Whether or not such materials are considered
secondary data is dependent upon how you use them.
When you analyse such ‘documents’ directly as part of your research, you’re using
them as secondary data. However, you’re more likely to be simply using them as a place
to find secondary data. Let’s take an example of a BBC news report uploaded on
YouTube. If you want to use this to explore the media’s attitude to the government and
in particular how they have reported government policies, then the BBC news report
on YouTube is your secondary data. This is because the video clip itself is the subject of
your analysis. However, if you’re using this YouTube video clip as a place to find secondary data, say, a record of government figures for the number of people who are unemployed and seeking work, the video clip is only the source of your secondary data. Your
secondary data are the number of people who are unemployed and seeking work. This
is because you are treating the data which have been reported in the video clip as the
subject of your analysis, rather that the video clip itself. In other words, the YouTube
news report is simply where you found these data. It is a source in which the secondary
data on government figures for the number of people who are unemployed and seeking
work you’re using have been recorded subsequent to their being collected. Once
obtained, you can analyse secondary data using either or both quantitative and qualitative techniques (Chapter 7).
Documentary secondary data will have originally been collected as text, materials,
non-text materials or a combination (Research in practice 4.2):
Text materials. This form of secondary data consists of a wide range of textual secondary data, including those documents held as administrative and public records such
as reports and minutes of meetings as well as diaries and transcripts of interviews
(including those originally broadcast on radio and television). Books and journal
articles are considered to be secondary data only when you are analysing them
directly as part of your research.
Non-text materials. Non-text documentary secondary data include video and voice
recordings, podcasts, films, radio and television programmes, images such as photographs and drawings as well as materials stored in organisations’ databases. As before,
it is the video or photograph rather than its source (e.g. YouTube) that are the secondary data.
Using social networking sites
Han’s research was concerned with the impact of social media on brand awareness. He
was aware from the academic literature that social media was important in marketing
and could influence consumers’ perceptions of different festivals as brands (Hudson
et al., 2015). Aspects such as a festival’s image and atmosphere were particularly important to festival brand loyalty and equity (Leenders, 2010). Based on the academic
Research in practice 4.2
4.2 Forms secondary data can take 91
Multiple source secondary data
Our final type of secondary data, multiple source, consists of different data sets that
have been combined to form a new data set prior to your accessing the data. Such secondary data can be based entirely on documentary or survey secondary data, or a combination of the two. You almost certainly come across such data on a daily basis as
tables and listings in quality newspapers such as the
Financial Times (and associated
websites). These are compiled from a variety of sources, an example that appears in the
Financial Times and on its associated website (http://www.ft.com) being the FT Global
500. This provides a snapshot of the world’s top 500 companies for that year, bringing
together data from different sources, to record their market capitalisation, sector, turnover, net income, total assets and number of employees. Such data can provide you
with extremely useful background material for your project when, for example,
researching a particular market sector.
Data are combined into a multiple source secondary data sets in one of two forms:
either as a snapshot or longitudinally as a time series.
Snapshot
Snapshot multiple source secondary data consist of data drawn from more than one
source which relates to a single time, much of it being accessible online. You will find
that these data are often combined geographically in country or region reports such as
the European Union’s
Eurostat Yearbook and Eurostat Regional Yearbook (Eurostat, no
date). These online-only publications are updated on a rolling basis, each article being
updated as new data, both quantitative and qualitative, become available. Data may
also be combined for a particular industry or sector to create a snapshot. Many university libraries have online subscriptions allowing online access to business intelligence
companies’ reports and data. Key Note, for example, provides downloadable market
insights and analysis reports covering over 30 different industry sectors for UK,
European and international markets. For those sources where your university does not
have a subscription, you may well find the price prohibitive, although very occasionally a reference copy may be held in your university library.
literature on branding and social media, Hans argued that, to use social media most
effectively, festivals needed to follow a three-stage process of providing material of
interest, engaging people and using them as advocates for their festival.
Along with many other festivalgoers, Hans had ‘liked’ the Facebook pages of the
festivals he attended. These pages often contained festival organisers’ posts about the
festival as well as comments and other posts from festivalgoers comprising both text
and non-text materials. Although the data in these posts were not originally intended
for Han’s research, he considered they would provide data about the effective use of
social networking sites for festival marketing. Because many fesitvals’ Facebook pages
were open to everyone, Hans considered that the information was in the public domain
so he could use it for his research without seeking consent.

92 Chapter 4 Using secondary data
Time series
Time-series multiple source secondary data are created by combining comparable variables (we defined these in section 1.7) collected through different surveys at different
times, or through the same survey that has been repeated over time. These include
series of data such as number of people out of work, manufacturing output and, as illustrated in Research in practice 4.3, change in retail sales. You can also combine different
Definition
time-series data: data recorded over time for the same variable or variables, usually at regular
intervals.
UK government longitudinal data
As part of his research project examining trends in the retail industry over the past
five years, Muhammad was interested in how UK retail sales had altered. Searching the
Research in practice 4.3
Source: © Office for National Statistics (2016). Contains public sector information licensed under the Open
Government Licence v3.0.

4.3 The potential of secondary data 93
secondary data sets to create time-series data, such as changing attitudes to gender
discrimination. For many research projects, this will be the only way in which you
will be able to obtain longitudinal data because of the time constraints set by the
research process.
UK Office for National Statistics website www.ons.gov.uk, he initially found a graph of
retail sales for the period January 1996 to September 2016 with hyperlinks to both
downloadable publications and data sets. Muhammad clicked on the hyperlink to the
bulletin
Retail Sales in Great Britain: September 2016 and read it carefully. The hyperlink ‘Data set’ linked to a downloadable spreadsheet comprising 27 tables of retail sales
data, for the period 2010 to September 2016. Muhammad subsequently reanalysed
these data as part of his research project.
We’ve already mentioned the potential of secondary data in terms of saving you time
and money by allowing you access to larger data sets than you could ever hope to collect
yourself. This huge variety of large, often high-quality data sets that are readily available
at no cost is one of the major opportunities offered by secondary data; as Mark says to
his students, ‘If someone else has already collected suitable high-quality data which
you’re allowed to use, why bother to collect the data yourself?’ The use of such data also
allows you to show that you can find and integrate valuable material into your arguments, giving your project an air of authority. However, in addition to access to larger
data sets than you could collect yourself, the authority such data sets can provide and
time and money savings, there are a number of other reasons for using secondary data
in your research project. These are listed in the following sub-sections.
Data are often already in the public domain
For many researchers, a key advantage of using secondary data is that much of these
data are already in the public domain. This means you do not need to negotiate access
to research participants (Chapter 3) or obtain permission to use the data. Indeed, many
public authorities include a statement explicitly granting all users permission in the
form of a licence to use the data, usually providing that you acknowledge the source
with an appropriate attribution statement such as ‘Contains public sector information
licensed under the Open Government Licence v3.0’ (Research in practice 4.3). However,
for data that are not in the public domain, such as that contained in organisations’
databases or administrative records, you will still need to obtain permission.
Data are often available in software compatible formats
While secondary data are available in many forms (section 4.2), they are increasingly
available using the Internet in formats that can be read directly into spreadsheets and
4.3 The potential of secondary data
94 Chapter 4 Using secondary data
other analysis software such as IBM SPSS Statistics (Research in practice 4.4). Some public authority data sets also allow you to select and download the precise data you require
rather than downloading a large data set, much of which you will not need. This means
you can focus on what data you actually need while also saving considerable time by
not having to type your data into, for example, spreadsheets.
It is an unobtrusive method
Not surprisingly, as secondary data have already been collected, you will not need to
ask your potential respondents if you can collect the data from them! This means you
will be respecting their privacy and rights as individuals to be left alone. You will also
not be taking up any of their time in either seeking permission (where data are in the
public domain) or actually collecting data. Issues relating to confidentiality of participants are also reduced or avoided as the data have already been anonymised as part of
the original collection process. This means you’re unlikely to risk breaching your university’s code of ethical practice (Chapter 3). However, if you intend to use data held by
a particular organisation, such as one for which you work, you will need to ensure the
anonymity of that organisation is preserved and to obtain written permission to use
their data for your research project.
Data sets can be readily combined
Data from a range of secondary sources, such as different surveys, can be combined to
create one new data set. You will find this particularly helpful if you wish to undertake
a longitudinal study. For such studies, you can combine data from a series of surveys
where the same questions have been asked into one data set. Indeed, for many public
authority data sets, aggregation of variables or the combining of data from different
surveys has already been undertaken before the data are made available (Research in
practice 4.3 and 4.4). However, as you can see in Research in practice 4.4, to fully
understand such data, you need to be clear about the precise definitions used for each
variable. You can also combine data sets from different geographical areas such as
countries or regions to make comparisons between countries or regions. This can be
useful if, for example, you’re comparing changes in manufacturing output between
two countries.
Eurostat (European Union) downloadable data
Jasvinder had decided to pursue a research project on tourism and, in particular, why
some European Countries were more popular with tourists than others. She noticed the
European Union’s Eurostat website included a link to the Main Tables for Tourism data
(Eurostat, 2016).
Research in practice 4.4
4.3 The potential of secondary data 95
Data are more open to public scrutiny
Choosing to use secondary data can provide you with access to data of very high quality. For example, the surveys conducted by professional researchers working in government departments or private survey organisations have to meet exacting scrutiny
This had a series of further links to downloadable tables, including four providing
data for each member state, which he felt could be most useful:
Nights spent at tourist accommodation establishments
Nights spent at hotels and similar accommodation
Nights spent at holiday and other short-stay accommodation
Nights spent at camping grounds, recreational vehicle parks and trailer parks
These tables presented data collected from 2006 to 2015, highlighting massive differences in the number of nights spent at tourist accommodation establishments between
member states, the highest number of nights in 2015 being recorded for Germany,
Spain, France and Italy; data for the UK not being available for that year. The data also
indicated that certain forms of accommodation were used far more in some member
states; for example, more nights were spent at camping grounds, recreational vehicle
parks and trailer parks in France than any other member states.
However, before using these data, Jasvinder needed to be sure she understood precisely what the data meant. She therefore clicked on the hyperlink to the ‘Manuals and
Guidelines’ web pages and then downloaded the
Methodological Manual for Tourism
Statistics
(Eurostat, 2014). This defined each of the types of tourism accommodation in
considerable detail, outlying what each type included and excluded. For example, camping grounds, recreational vehicle parks and trailer parks included ‘provision of accommodation in campgrounds, trailer parks, recreational camps and fishing and hunting
camps for short stay visitors, provision of space and facilities for recreational vehicles . . .
but excludes mountain refuge, cabins and hostels’ (Eurostat, 2014: 59).
Source: Eurostat (2016) Copyright European Communities, 2016. Reproduced with permission.
96 Chapter 4 Using secondary data
standards in terms of rigour of the research method used. These data sets are therefore
likely to be larger and of better quality than any survey you can design and deliver yourself. The majority of them, in addition to the data collected, provide you with information about precisely how the data were collected, including sample size and selection,
response rates, a copy of the data collection instrument used (such as a questionnaire)
and an assessment of how representative the data are. This information can be used
by you and other users to judge the quality of the data and can, as we outline in
section 4.4, also help you to assess the suitability of the secondary data for your research.
Data can provide contextual background
If you’re collecting your own data, it is often useful to set the findings from your own
data within a broader context. For example, if you have collected data for your research
project through interviews with a particular service organisation’s potential customers,
you can use secondary data such as a market research report for that service to locate
your findings in the broader context of that particular industry. Alternatively, you
might wish to use recent government population estimates to assess the generalisability
of your findings by assessing whether your participants were representative, being present in the same proportions for each age group as in that population.
As you will have gathered from reading this chapter so far, we are fans of using secondary data in research. However, our enthusiasm, and hopefully yours, needs to be tempered with realism. While secondary data offers considerable potential for many
research projects, its very nature means that it is open to a variety of often valid criticisms. As our definition of secondary data indicates, it was collected for a different purpose. This means the data may not meet your research needs fully, and it may have been
manipulated in some way. You will not have undertaken data collection. You’re therefore very unlikely to know precisely how the data were collected or the impact this will
have on the actual data you are now going to use. Finally, although public authorities
make their data available free of charge, do keep in mind that the cost of actually
obtaining such data is high. For this reason, other organisations (such as market
research companies) may ask you to pay for it. In addition, the commercial sensitivity
of some data means organisations may refuse to allow you to use their data even if they
employ you. These criticisms represent potential pitfalls about which you need to be
aware and which we will now discuss in more detail.
Only meets the research needs partially
Unlike data you collect yourself to answer your research question, secondary data are
unlikely to have been collected for that same purpose. This means the data may be
inappropriate for your research question or, as is more likely, only partially relevant.
4.4 Possible pitfalls of using secondary data
4.4 Possible pitfalls of using secondary data 97
Reasons for this vary but include the data not being current, problems with definitions
used for particular secondary data variables, the way in which data have been aggregated not matching your needs and because the data contain clear errors.
Most quantitative secondary data are still only available in aggregated tabular form
(Research in practice 4.1 and 4.4). Where the tables into which data have been aggregated
do not meet your needs, perhaps due to the categories used being too broad, this may
cause problems. Similarly, it is also likely to be problematic where definitions used for the
data variables (Research in practice 4.4) in these tables differ from those in other tables or
are unclear. Invariably, the solution here depends upon the extent to which you judge the
secondary data to be inappropriate for your own research. Where the data do not meet
your needs at all, your only solution is to find alternative data. However, where the data
meet your needs partially, you will need to adapt and compromise. The extent to which
you compromise will depend upon what other secondary data are available and whether
you’re able to collect appropriate data yourself. If you do decide to compromise, you will
need to explain in your project report your concerns and the compromises you have
made, and why you still feel able to use these secondary data. For example, if you’re comparing data on satellite TV subscribers between countries in Europe, the most recent data
you will be able to obtain is likely to be available only for some countries and at least a few
years out of date. In your project report, therefore, you need to make clear these omissions
and when the data were collected, explaining it is the most recent data available.
Data are not always value-neutral
The original purpose for which secondary data were collected is invariably reflected in
the way they are presented and the interpretations of those who produced them. Data
are often used to support an argument or make a particular point and so may be presented in the way that best supports that argument. This is often apparent if you compare the data presented about the same topic by two newspapers with different political
affiliations. When you do, you will probably see that, although some of the data used
are the same, different data from the same original data set will also have been quoted
selectively by each newspaper to support their different interpretations.
Definitions used can also change within a single data set over time, impacting on the
meanings attributed to the data. For example, between 1979 and 1994, there were nine
significant changes in the calculation method used for the United Kingdom’s monthly
count of unemployment claimants, resulting in the overall number of claimants being
reduced by 481,000 (Fenwick and Denman, 1995). Not surprisingly, the claimant count
became widely criticised as a data source because it ‘seriously underestimated’ the number of people out of work, the definitional changes making any serious estimation of
changes in the unemployment rate over time ‘virtually useless’ (Levitas, 1996: 46). It is
therefore important that you remember secondary data may have been manipulated by
people to serve their own purposes, and therefore assess the likely impact of this on
actual data values. In particular, you should not accept any secondary data without first
understanding the precise definitions used and, for longitudinal data, whether these
have been altered over time.

98 Chapter 4 Using secondary data
Unlikely to know precisely how the data were collected
When using secondary data, you have no control over the quality of the data. You
therefore have to infer the likely quality from other information provided. Researchers
should always provide clear information about the method of data collection and the
variables about which questions were asked. This should be in sufficient detail for you
to infer the quality of the data in both reports and articles as well as for actual data sets.
Despite this, it is unlikely as a user of secondary data that you will ever know the precise
context in which each of a series of interviews took place or the nuances of the relationship between an interviewer and the participant. Whatever the level of detail provided
about the method, you will obviously know less than if you had collected the data yourself. For some secondary data sets, no information will be provided about how they
were collected. It is best to treat these data with caution, as often this information is not
provided because it would indicate that there are problems with these data. However,
this is not always the case. A number of excellent market research companies provide
very limited information about their data collection methods as they consider such
details commercially sensitive.
Can be costly to obtain
Where data have been collected for commercial reasons, you’re likely to find access
both difficult and costly. Market research reports such as those produced by the Mintel
International Group, although easy to identify and locate online, are expensive to purchase. This means, if they are not available in your university library or at another
library you can visit, you will be unable to access the data such reports contain.
When deciding whether to use secondary data, there are a number of things you need
to think about. Like many students, you may think that using secondary data is easy
because you do not need to design a data collection method or collect the data yourself.
This is not the case. The time you have saved will be used in ensuring you have access to
the data and that the data will enable you to answer your research question. Fortunately,
because the data already exist, you can assess their suitability before you begin your
analysis. Inevitably, the suitability is dependent upon the data being relevant to your
own research. However, if the data are relevant, suitability will also depend on the purpose for which the data were originally collected and the method that was used. Let’s
now discuss these in more detail.
The relevance of the data
The most important criterion regarding the suitability to you of any data set is that it
provides you with the information you need to answer your research question. This
means that the data collected, and the definitions used for the variables in which you’re
4.5 Assessing the suitability of secondary data
4.5 Assessing the suitability of secondary data 99
interested, must be a close match to the data you ideally require. Often you will find
that the data available and the definitions that were used are not exactly as you would
wish for in an ideal world. For example, you may ideally require data on the purchasing
power of children, defined as the under-18 demographic group. If the secondary data
defines this group as less than 17 years of age, these data will not match your requirements exactly. However, the world of secondary data is not perfect. You may therefore
decide to use these data and state in your project report that they do not include 17-yearolds, offering an indication of the likely impact of this on your subsequent analysis. In
doing this, you’re offering both an assessment of the suitability of these data to answering your research question and ensuring that those who read your project report are
aware of possible limitations and their likely implications.
When using secondary data, you will need to exclude those data that are not relevant to your research question. Data on the United Kingdom will need to be excluded if
your research is only concerned with a post-Brexit European Union. Alternatively, some
secondary data sources, in particular those collected by government surveys, may not
include all the data variables you would ideally require for your analysis. You will need
to make an assessment as to whether these missing variables are essential to answering
your research question. If they are, then the data that are available are not suitable, and
you need to look elsewhere.
The original reasons for collecting data
Assuming you have found what appears to be suitable secondary data, you need to look in
more detail regarding the reasons why the data were originally collected. A good place to
start when trying to establish the original purpose is to look for the objectives or research
questions. For research reports and articles, these are often stated in the introductory section or, for data accessed online, in the accompanying documentation or notes.
Alternatively, you can look at who commissioned the research. For example: were data collected for a public authority such as a government department, or by a private organisation
that employed a market research company or perhaps an academic to collect the data? The
answers to such questions will give you an idea of the original motivations behind collecting these data and, as a consequence, your analysis and interpretation. If, for example, the
research was undertaken for a trades union, it is likely that the focus of the data collected
and the associated report will be different than if it was undertaken for an employer or
employers’ organisation. You may consider the original focus of the research too narrow in
relation to your own research question, highlighting the need either to find additional secondary data or to collect your own primary data. Alternatively, you may feel that the data,
although useful, needs to be treated with caution because of the potential for bias caused
by the original purpose of the research. Whatever you decide, it is important that you
explain the reasons for your decision clearly in your project report.
The method used
Assessment of the method used is concerned with how the data were collected, including the actual questions used to obtain the data. When assessing the method, you need
100 Chapter 4 Using secondary data
to establish precisely how data were collected, including any personal interaction
between participants and the person collecting the data. Data collected using an
Internet questionnaire will inevitably constrain those who participate (and therefore
the data collected) to people who have Internet access and, in most cases, an email
address. Similarly, data collected using a telephone interview will restrict those who
participate to people who have access to a telephone. In both examples, the people who
respond may not be representative of the population as a whole or those you wish to
research. You therefore need to treat the data with care. Data collected using structured
interviews conducted by a professional interviewer either by telephone or face to face
are likely to have followed a set and detailed script, meaning that all participants are
likely to have been asked exactly the same questions in the same way. In contrast, questions asked by amateur interviewers might unwittingly have been reworded for some
participants, prompting a different response indicating that perhaps the data quality is
more variable. You therefore need to pay particularly careful attention to who conducted the interview.
The questions used
For data collected originally using interviews or questionnaires, question phrasing is
crucial. For large-scale questionnaire surveys, organisations spend a vast amount of
resources in trying to ensure that the questions asked would not be misinterpreted and
are neither biased nor leading. Despite this, it is still worthwhile examining carefully
the questions asked to ensure that the data collected is relevant to your own research
and, of equal importance, that you do not misinterpret responses in these data.
The breadth of forms of secondary data we’ve discussed in this chapter serve also to
emphasise the variety of places where you may find secondary data. Despite the Internet
and availability of general search engines such as Google and Bing, finding suitable
data still depends in part on your awareness of
information gateways and potentially
useful secondary data sites. It is also important to recognise that not all secondary data
will be in electronic form or accessible online. Some will be paper-based and listed in
your own university library’s catalogue. Other paper-based secondary data will be in
reports that are held only in specialist libraries or organisations. The latter of these will
be the most difficult to locate and may involve you visiting a specialist library or, if permission is granted, an organisation to browse through publications or reports.
Remember, even if you intend to use data you know is held by an organisation for which
4.6 Where and how to find secondary data
Definition
information gateway: website that provides access to specific websites and pages, each site having
been evaluated prior to being added to the gateway.

Summary 101
you work, you will still need to seek written permission to reuse these data for your
research project.
For secondary data published by governments, finding the data online is relatively
easy compared to other sources. Precise references to particular government sources are
often given in journal articles and books, although often the associated web addresses
will have changed. Some governments and government departments provide guides or
lists of statistics they make available. However, like the majority of other secondary providers, most governments expect you to use either a combination of menus and web
links or the search engines within their own information gateway to find data that are
available. It is therefore important that you have a reasonable idea of the sort of data
you’re looking for and the search terms and phrases that are likely to have been used to
describe it. Providing you then know the web address of an appropriate information
gateway (Table 4.1), it is relatively straightforward to search for specific secondary data
published by governments.
You will find data held by non-governmental organisations more difficult to
locate. For data that are available online, there are a number of other information
gateways that are of use (Table 4.1). If you find that these do not eventually produce
useful data, general search engines such as Google, Bing and Yahoo can help you
locate potential data sources. In some cases, these data will be held in reports that
can be downloaded from sites hosted by professional and trade associations. In others, there will be only a reference to an internal report for a private company, and it
will be necessary for you to request the document and seek written permission before
the data can be used.
Table 4.1 Selected information gateways to secondary data sources
Name Internet address Content
UK Office for
National Statistics
http://www.ons.gov.uk/ UK national statistics
Direct.gov http://www.gov.uk/ UK government information service
Eurostat http://ec.europa.eu/eurostat European Union and member states
statistics
Europa http://europa.eu European Union information service
Morningstar http://www.morningstar.co.uk/ Financial information on companies,
trusts and markets
USA.gov http://www.usa.gov/ USA government information service
Secondary data are data that were originally collected for some other purpose. They
can be contrasted with primary data, which are collected for the specified research
purpose.
Summary
102 Chapter 4 Using secondary data
Secondary data comes in many forms including both quantitative (numerical) data
and qualitative (non-numerical) data. These can be grouped using the actual source
or sources of the data into survey, documentary and multiple-source secondary
data.
Secondary data can provide you with fantastic research opportunities that would
otherwise be outside your reach, by allowing you access to larger data sets than you
could ever hope to collect yourself and by saving you time and money.
Other reasons for you using secondary data include:
the data are already in the public domain, thereby often avoiding concerns about
access and permission to use the data;
the data are often available in software-compatible formats, allowing easy
analysis;
the data provide an unobtrusive method, respecting individuals’ rights to privacy
and being left alone;
the data can be readily combined with other data sets, allowing, for example, longitudinal studies;
the data can provide contextual background to a research project.
However, using secondary data has a number of possible pitfalls, including:
the data may only meet your research needs partially;
the data may have been manipulated in some way and so not be value-neutral;
the definitions used within the data may have changed over time;
the data can be costly to obtain.
The suitability of secondary data to your research project is dependent upon:
the relevance of the actual data to your research question;
the original reason for collecting the data;
the methods used to collect the data and the actual questions asked.
Increasing provision of online open-access data by governments has increased the
availability of secondary data considerably.
Not all secondary data will be available online. Some will be paper-based and available in your own university’s library. Other secondary data will be in reports that are
held only in specialist libraries or organisations.
Think about how you might use secondary data to answer your research question(s)
and meet your objectives. If you consider secondary data would be helpful, or its use
is a criterion for your assessment, make a list of the data variables you think you’re
likely to require, being as precise as possible in the terms you use.
Thinking about using secondary data
References 103
Initially, use the information gateways listed in Table 4.1 to search the Internet for
secondary data information sources that contain these variables. If these result in
no suitable data, try using general search engines such as Google, Bing or Yahoo, or
other sources.
Evaluate the relevance of the data variables in each of these sources against the data
variables you require (section 4.5). Remember, it is unlikely that you will find secondary data that matches your requirements exactly.
Evaluate the suitability of each secondary data source by considering the original purpose
of the research, the methods used and the questions used as outlined in section 4.5.
If the data are suitable, where appropriate seek and gain permission to use the data
before obtaining a copy.
Edelman (2016) 2016 Edelman Trust Barometer Global Report. Available at: http://www.edelman.
com/insights/intellectual-property/2016-edelman-trust-barometer/global-results/ [Accessed
28 October 2016].
Eurostat (no date)
Europe in Figures. Available at: http://ec.europa.eu/eurostat/statisticsexplained/index.php/Europe_in_figures_-_Eurostat_yearbook [Accessed 1 November 2016].
Eurostat (2014)
Methodological Manual for Tourism Statistics Version 3.1. Luxembourg:
Publications Office of the European Union.
Eurostat (2016)
Tourism: Main Tables. Available at: http://ec.europa.eu/eurostat/web/tourism/
data/main-tables [Accessed 1 November 2016].
Fenwick, D. and Denman, J. (1995) The monthly unemployment count: change and consistency,
Labour Market Trends, November, 397–400.
Hudson, S., Roth, M.S., Madden, J.T. and Hudson, R. (2015) The effect of social media on emotions, brand relationship quality and word of mouth: An empirical study of music festival
attendees.
Tourism Management, 47(1), 68–76.
Leenders, M.A.A.M, (2010) The relative importance of the brand of music festivals: a customer
equity perspective.
Journal of Strategic Marketing, 18(4), 291–301.
Levitas, R. (1996)
Interpreting Official Statistics. London: Routledge.
Office for National Statistics (2016)
Retail Industry. Available at: http://www.ons.gov.uk/
businessindustryandtrade/retailindustry#datasets4 [Accessed 31 October 2016].
Pricewaterhouse Coopers, University of Southern California and the London Business School
(2013)
PwC’s NextGen: A Global Generational Study. Available at: http://www.pwc.com/
us/en/people-management/publications/nextgen-global-generational-study.html [Accessed
27 October 2016].
Saunders, M., Lewis, P. and Thornhill, A. (2016)
Research Methods for Business Students (7th ed.).
Harlow: Pearson.
References
Chapter 5
Choosing your research design
Perhaps when you first started thinking about your research project, you thought about
collecting the data, maybe through a questionnaire. This is perfectly normal, as it is an
exciting part of any research project. But as on any journey, the research journey may take
a number of routes and be accomplished by a variety of methods. The most obvious option
is not always the best, whether deciding which route to use when travelling from one city
to another or on the method for researching the reasons why consumers prefer recording
television programmes rather than watching them live. So it will not surprise you to learn
that you have many options when designing your research project. What will be a major
factor in determining the quality of your research proposal is the extent to which you have
considered these various options, and the clarity of thought which you have displayed in
coming to a decision as to which design to adopt. The main part of this chapter is concerned with the different options you face when designing your research. We examine the
differing purposes of projects which we call exploratory, descriptive and explanatory
research. These project purposes may be thought of as the overall umbrella under which
more detailed research strategies shelter. We then go on to detail these research strategies:
the general ways in which to set about answering the research questions, the writing of
which we examined in Chapter 1. There is a choice to be made regarding the use of quantitative and/or qualitative methods in a research design. Consequently, we consider the
methodological choices of using one or more quantitative methods, one or more qualitative methods or combining them in a so-called mixed-method design. Our practical examination of research design and strategies ends with a consideration of the time frame
applying to the research. It may be that the research questions you have set dictate that
the research be carried out over an extended period of time. Alternatively, a snapshot may
be perfectly acceptable.
The chapter begins and ends with two other issues which impart quality to your research
design. The first of these is the subject of research philosophy. A consideration of your own
research philosophy helps you to examine your fundamental ideas about research: those
5.1 Why you should read this chapter
5.1 Why you should read this chapter 105
ideas that will underpin your research and which you may not have even thought about! In
other words, it starts you thinking about your thinking. The second research issue is that of
research credibility. Here we address the question of whether you have taken the steps in
your research design to ensure that your findings and conclusions are believable. Obviously,
this is crucially important. If the reader has doubts about the credibility of your research,
then you may have wasted your time and that of others. In addition, you may suffer practical consequences such as failing the research module!
The research onion
At this point, let us introduce our research onion (Figure 5.1). This serves as a route map to
chart our way through this chapter. In addition, the onion is a metaphor for describing the
layers of the research process. The outer two layers of the onion contain thinking about
research philosophies and approaches to developing theory. The next three (central) layers
reflect the need to consider methodological choices, research strategies and the time horizon. The centre of the onion, data collection and analysis, are the subjects of Chapters 6
and 7, respectively.
Data
collection and
data analysis
Cross-sectional
Action
research
Action
research
Grounded
theory
Multi-method
qualitative
Mixed-method
simple
Pragmatism
Positivism
Philosophy
Approach to
theory
development
Methodological
Time horizon Strategy(ies) choice
Techniques and
procedures
Critical realism
Postmodernism
Interpretivism
Deduction
Induction
Abduction
Multi-method
quantitative
Mono
method
qualitative
Longitudinal
Archival
Narrative research
inquiry
Survey
Experiment
Case study
Mono method
quantitative
Mixed-method
complex
Figure 5.1 The research onion here
Source: © 2015 Mark Saunders, Philip Lewis and Adrian Thornhill.
106 Chapter 5 Choosing your research design
What is meant by the term ‘research philosophy’?
The term research philosophy refers to a system of beliefs and assumptions about the
development and nature of knowledge. This sounds rather profound, but that’s just
what you will be doing when you conduct your research; developing knowledge in a
particular field. Your knowledge development may not be as fundamental as changing
the way we think about how organisations develop, but even establishing why customers prefer ice-cream cones to wafers from a particular ice-cream vendor is developing
new knowledge.
You almost certainly won’t think about your research philosophy at every stage in
your research, but you will make a number of types of assumptions which will influence
the way in which you set about your research. These assumptions fall into three main
categories, which are called ontological assumptions, epistemological assumptions and
axiological assumptions. Let’s have a look at what these mean and the implications
they have for your research.
Ontological assumptions are about the nature of reality. Your ontological assumptions shape the way in which you see and study your research objects. In business and
management these objects include organisations, management, individuals’ working
lives and particular organisational events. Your ontology, therefore, determines how
you see the world of business and management and, therefore, the direction your
research project will take.
For example, take the case of organisational change, in particular resistance to that
change. Traditionally, the ontological assumption made by managers and researchers
has been that resistance to change hinders the change process. So the focus of research
has usually been on how resistance to change could be eliminated. The emphasis has
been on the sources of that resistance, the reasons for it and how it may be overcome. But
recently, some practitioners’ and researchers’ ontological assumptions about change have
altered. They have begun to think about resistance to change as inevitable – something
that happens all the time whenever organisational change is initiated. Indeed, resistance
can be seen as beneficial as it can highlight problematic aspects of change programmes
and help to generate different ways of solving organisational problems.
Epistemological assumptions are about knowledge – what constitutes acceptable,
valid and legitimate knowledge, and how we can communicate knowledge to others. In
business and management a wide range of knowledge sources may be relevant. These
may include numerical data, written texts, visual data (e.g. photos and videos) and
personal diaries. These can all be considered legitimate. Consequently, your choice of
research methods may be heavily influenced by your epistemological assumptions. For
5.2 The importance of research philosophy
Definition
research philosophy: overall term that relates to the development of knowledge and the nature of
that knowledge in relation to research.

5.2 The importance of research philosophy 107
example, the positivist (see the next section) assumption that objective facts offer the
best scientific evidence is likely to result in the choice of quantitative research methods,
leading to research findings which are likely to be considered objective and generalisable. But, for a person with a different epistemological view, such as an interpretivist,
such methods are less likely to offer a rich and nuanced view of organisational life.
Axiological assumptions emphasise the importance of values and ethics within the
research process. The values that we are referring to here are our own as researchers, and
those of our research participants. There is no doubt our values play a tremendously
important part in what research project we choose and the way in which we go about
collecting our data. For example, choosing to collect data by interview suggests that
you place greater value in personal interaction with your respondents than were you to
decide to collect their views through an Internet questionnaire.
You may find it helpful to write your own statement of personal values in relation to
the topic you are studying. For example, if you are studying an organisation’s policy on
corporate social responsibility, your own values may have had a telling influence on the
choice of this topic and the way in which you design your research. In addition, reference
to your statement of personal values will help heighten your awareness of the value judgements you are making in drawing conclusions from your data. It would be perfectly normal had you not considered your own beliefs about the nature of the world around you,
what constitutes acceptable and desirable knowledge or the extent to you which your
values influence your choice of research topic and data collection techniques. But as a
researcher, it’s very helpful to develop the skill of reflecting upon such issues. Many
examiners will want to understand why you decided to go down a particular route.
We now turn our attention to the five research philosophies that form the outer
layer of our research onion.
Five research philosophies
Positivism
Positivism relates to the philosophical stance of the natural scientist. Here you study
observable social realities (e.g. organisations, managers) to produce law-like generalisations. Positivism promises unambiguous and accurate knowledge using methods
designed to yield pure data and facts uninfluenced by human interpretation or bias.
If you were to adopt an extreme positivist position, you would:
see the social realities you are studying as real in the same way as physical objects and
natural phenomena are real;
focus on discovering observable and measurable facts and regularities, and only phenomena that you can observe and measure would lead to the production of credible
and meaningful data;
Definition
positivism: a research philosophy similar to those used in the physical and natural sciences. Highly
structured methods are employed to facilitate replication, resulting in law-like generalisations.

108 Chapter 5 Choosing your research design
look for causal relationships in your data to create law-like generalisations similar to
those produced by scientists;
use these universal rules and laws to help you to explain and predict behaviour and
events in organisations.
In adopting a positivist philosophy, you may well use existing theory to develop
hypotheses. You would test these hypotheses in the expectation that they would be
confirmed, in whole or part, or refuted, leading to the further development of theory that could then be tested by further research. But it doesn’t necessarily mean
that, as a positivist, you have to start with existing theory. You could develop
hypotheses which would lead to the gathering of facts (rather than impressions),
which would provide the basis for subsequent hypothesis testing. As a positivist,
you would also try to remain neutral and detached from your research and data in
order to avoid influencing your findings. This means that you would undertake
research, as far as possible, in a value-free way. Positivists claim to be external to the
process of data collection as there is little that can be done to alter the substance of
the data collected. Positivist researchers are likely to use a highly structured methodology, such as questionnaires or structured observation, in order to facilitate replication. Furthermore, the emphasis will be on quantifiable data that lend
themselves to statistical analysis.
(Critical) realism
Like positivism, realism relates to scientific inquiry. It has two distinct strands. That
which concerns us most as business and management researchers is called critical realism. But first, let’s clarify what is a more extreme form of realism, termed direct realism.
Put simply, direct realism says that what you see is what you get: what we experience
through our senses portrays the world accurately. By contrast, the philosophy of
critical
realism
focuses on explaining what we see and experience, in terms of the underlying
structures of reality that shape the observable events. Critical realists argue that we
experience the world in two stages. First, there is the object itself and the sensations
conveyed by the object. Second, there is the subjective processing that is present in our
minds after that sensation meets our senses. Direct realists hold that the first step of
processing is sufficient.
Once again, we can hear you saying ‘What has all this to do with business and management research?’ The answer is that the distinction between direct and critical realism is important for much business and management research. We are often concerned
with first describing complex business situations in order to understand what is going
on. So we need to study not only what is immediately apparent but also what lies behind
what is immediately apparent. We need to understand the deeper structures and
Definition
critical realism: a philosophy which focuses on explaining what we see and experience with the
emphasis on understanding the underlying structures of reality that shape the observable events.

5.2 The importance of research philosophy 109
relations that are not directly observable but lie beneath the surface of social reality.
Let’s say you had to study a major organisational problem, such as that faced by
Samsung when the battery caught fire in its new Galaxy Note 7 mobile phone shortly
after its introduction in 2016. In order to arrive at a conclusion for the causes of such a
debacle from which others may learn, you would need to understand a good deal about
the context of the event. For example, you would need to know about the relationship
between product marketing managers and technical executives. Presumably the marketeers would wish the product to get to market with the minimum of delay, whereas
the technicians would be more concerned with getting the product right. Such a tension would not be immediately apparent to the researcher but would necessitate understanding the structures and relationships that were beneath the surface of what was
immediately evident.
Interpretivism
The concern of the critical realist with greater organisational complexity points the
way towards
interpretivism. Interpretivism relates to the study of social phenomena in
their natural environment. So a wish to understand what is going on in a work organisation, for example, would make it necessary to conduct research in that organisation
among its ‘social actors’. The term ‘social actors’ suggests the metaphor of a theatre.
As humans, we play a part on the stage of social life. Theatre actors have specific roles
which they each interpret in a particular way (it may be their own or that of the director), playing these roles in accordance with their interpretations. Similarly, as social
actors we interpret our everyday social roles in accordance with the meaning we give
to these roles. The way in which you quickly have to move from the role of student to
that of employee means a process of social adaptation on your part, which reflects
your interpretation of which behaviours are appropriate for this new role. If we interpret the roles that we play according to our definition of what is appropriate, then we
also interpret the social roles of others in accordance with our definition. This suggests that, as researchers, our values play a part in the research process. This is inevitable. It would be naive to think that our own personal values play no part in our
research. Even the choice of research topic, as well as the decision about the research
methods to adopt, reflects our values. Clearly, the skilled researcher needs to be wary
of this. Understanding the social world of our research subjects from their point of
view is the key here.
For business and management research, the interpretivist perspective is very relevant, particularly in such fields as organisational behaviour, marketing and human
resource management. Not only are business situations complex, but they are also
unique. They represent a particular set of circumstances and individuals coming
together at a specific time to create a unique social phenomenon.
Definition
interpretivism: a philosophy which advocates the necessity to understand differences between
humans in their role as social actors.

110 Chapter 5 Choosing your research design
Postmodernism
Postmodernism emphasises the role of language and power relations, seeking to question accepted ways of thinking and to give voice to alternative marginalised views.
Postmodernists go even further than interpretivists in their critique of positivism,
attributing even more importance to the role of language. Postmodernist business and
management researchers emphasise the importance of flux and change in organisational life. They believe that any sense of order is provisional and without foundation,
and can only be brought about through our language with its categories and classifications. At the same time, they recognise that language is always partial and inadequate.
In particular, it marginalises and excludes aspects of what it claims to describe, whilst
prioritising and emphasising other aspects.
Postmodernists argue that what is generally considered to be ‘right’ and ‘true’ is
decided collectively by powerful alliances in organisations. For example, in some organisations, what is right may be decided by accountants whose power derives from the
dominance of the organisation’s financial resources. This does not mean that, in this
example, the accountants’ way of thinking is necessarily the ‘best’ – only that it is seen
as such at a particular point in time by particular groups, such as shareholders. Other
perspectives that are suppressed are potentially just as valuable and have the power to
create alternative ‘truths’.
Postmodernist researchers seek to expose and question the power relations that
sustain such dominant realities. The goal of postmodern research is therefore to
radically challenge the established ways of thinking and to give voice and legitimacy to the suppressed and marginalised ways of perceiving issues that have been
previously excluded. Let us look at another example. As a postmodernist researcher,
you would, instead of accepting the concept of ‘management’ as a given, focus on
the ongoing processes of managing. You would challenge the accepted concepts
and theory of management and try to demonstrate what perspectives and realities
they exclude and leave silent, and whose interests they serve. This is likely to be the
position of the trade union researcher, who will see management, ideally, as a collective process involving employees as well as managers, both parties having a legitimate role to play in the process of management. As a postmodernist, you would be
using all forms of data – texts, images, conversations, voices and numbers – and
similar to interpretivists, you would undertake in-depth investigations of phenomena. A vital part of postmodernist research is the recognition that power relations
between the researcher and research subjects shape the knowledge created as part of
the research process.
Definition
postmodernism: a philosophy which emphasises the role of language and power relations that seeks
to challenge accepted ways of thinking and give voice to alternative views.

5.3 Differing approaches to theory development: deduction, induction and abduction 111
Pragmatism
Although our choice of research methods may reflect our values, it would be wrong to
think that the differing philosophies we have outlined above should be thought of as a
‘shopping list’ from which to choose a philosophy. In practice, you are far more likely
to be guided by what is possible. Indeed, if a research problem does not suggest unambiguously that one particular type of knowledge or method should be adopted, this
only confirms the pragmatist’s view that it is perfectly possible to work with different
types of knowledge and methods. This is the pragmatist’s position.
Pragmatists considers that the most important determinants of the research philosophy you adopt are
your research question(s) and objectives. Often, the pragmatist researcher starts with a
problem, their research aiming to contribute practical solutions that inform future
practice. The pragmatist would argue that it is quite possible to work with methods
that indicate, for example, a quantitative approach, say a questionnaire, alongside a
more qualitative stance, through interviewing the key participants at meetings about
their interactions. This echoes an argument that runs throughout this text: that mixing methods, both qualitative and quantitative, are possible, and may be highly appropriate, within one study.
Definition
pragmatism: a philosophy which argues that the most important determinant of the research design
adopted are the research question(s) and objectives, the aim often being to contribute practical solutions.
We now move to the next layer of the onion to examine three different approaches to
theory development: deduction, induction and abduction. We mentioned the topic
of theory in Chapter 1, and you will remember that theory is broadly defined as an
explanation of the relationship between two or more concepts or variables. The role
of theory will loom large in your study, as all research projects will need to link to
theory in some way. The most likely link will be to an existing theory explained in
the literature relevant to your research topic. For example, you may wish to explore
the theory in a different context, say, in your own employing organization or in a
different cultural context. The purpose will be to explore the extent to which the
theory applies or to suggest a modification to the theory based on your own findings
and conclusions.
The distinction between deduction and induction in particular takes us back a stage
to the development of the theory which you may be using from the literature. Although,
as we said, you may be using existing theory, it is useful to see how that theory may
have been developed.
Differing approaches to theory development:
5.3 deduction, induction and abduction
112 Chapter 5 Choosing your research design
Deduction: clarifying theory at the beginning of the study
Deduction is a research approach which involves the testing of a theoretical proposition
by using a research strategy designed to perform this test. There are five sequential
stages in deductive research. These are:
1 defining research questions from the general theory that exists;
2 operationalising these questions in a way that enables what is occurring to be
established (i.e. specifying the way in which the questions may be answered). This
may be in the form of a testable proposition (hypothesis) about the relationship
between two or more concepts or variables, or set of hypotheses, to form a
theory;
3 collecting data to answer the operationalized questions or test the hypotheses;
4 analysing the data collected to determine whether it supports the existing general
theory or suggests the need for its modification;
5 confirming the initial general theory or modifying it if the findings do not confirm
the existing general theory. (In the event of step 5 resulting in a modified theory, the
five sequential stages can be repeated to test the new theory.)
The key characteristics of deduction are, first, to explain causal relationships between
variables. For example, you may wish to establish the reasons for too many errors
being made by call centre operatives in giving information to customers. Having
studied the error patterns, it seems to you that there may be a relationship between
the number of errors and the length of the initial training programme received by
operatives. You define your research questions (stage 1) which explore the proposition that those operatives receiving the shorter version of the initial training programme are more error-prone than those who undergo the longer version. Secondly,
you need to operationalise the questions as testable propositions (stage 2). These
questions, therefore, need to be expressed in a way that enables what is occurring to
be established. In the call centre example mentioned earlier, the two key concepts in
the questions are error levels and training-programme length. The second concept
may be more straightforward than the first. For example, you would need to be clear
about what constitutes an error and distinguish between the seriousness of errors.
The third characteristic of deduction is the need to collect and analyse data to answer
the research questions (stages 3 and 4) and to see whether or not the existing theory
is confirmed (stage 5). The fourth characteristic is the use of a clearly structured
methodology to facilitate replication. This is important to achieve reliability, as we
see later in this chapter.
Definition
deduction: a research approach which involves the testing of a theoretical proposition by using a
research strategy specifically designed to collect data for the purpose of its testing.

5.3 Differing approaches to theory development: deduction, induction and abduction 113
Induction: conducting and developing theory from
the explanations that arise
If deduction has a ‘top-down’ flavour, then induction suggests a ‘bottom-up’ approach to
theory development. Inductive reasoning moves from specific observations to broader generalisations and theories. With inductive reasoning, we begin with specific observations and
measures, by observing patterns and repeated occurrences of phenomena and formulating
some speculative hypotheses or propositions from what has been observed which can be
investigated. All this is with a view to developing some general conclusions or theories.
When using the inductive approach, you are often trying to gain an understanding
of the meanings humans attach to events. Let’s consider again the example of the errorprone call centre operatives. An important impression you would want to gain is
whether the operatives actually like their jobs. The literature on call centres you have
reviewed suggests that it is possible to distinguish between those jobs that are intrinsically interesting and those that are simply dead boring! Some jobs may be more
demanding and stressful than others. Some operatives may simply not be cut out for
this type of work. Already you can see that alternative explanations are beginning to
form for the differential levels of operative errors. It may be that the explanation is
more complex than the amount of training that the operatives receive.
With induction, the emphasis is on a close understanding of the research context.
We raised the point earlier about the necessity to conduct research in more than one
call centre in the ability to generalise a theory across all call centres. From an inductive
stance, we are more likely to want to get a detailed picture of the experience of working
in one, or perhaps two call centres that have different environmental characteristics.
Induction possesses a more flexible structure to permit changes of research emphasis as
the research progresses. It may be that the stress in the call centre operatives you are studying
becomes a major issue in your data collection. This may present itself as an alternative explanation for high error levels to that suggested by insufficient training. In that case, it may be
sensible to build much more attention to stress experience into your research questions.
Abduction: combining deduction and induction
Instead of moving from theory to data (as in deduction) or from data to theory (as in
induction), an abductive approach moves back and forth, in effect combining deduction and induction. In essence, this matches what many business and management
researchers actually do.
Abduction begins with the observation of an unexpected
Definitions
induction: a research approach which involves the building of theory from analysing data already
collected.
abduction: approach to theory development involving the collection of data to explore a phenomenon, identify themes and explain patterns to generate a new – or modify an existing – theory which
is subsequently tested.

114 Chapter 5 Choosing your research design
occurrence and then works out a plausible theory of how this could have occurred.
Some plausible theories can account for what is observed better than others (Van
Maanen et al., 2007), and it is these theories that will help uncover more unexpected
observations. Surprising discoveries can occur at any stage in the research process,
including when writing your project report!
Using an abductive approach to our research on the reasons for too many errors
being made by call centre operatives would mean obtaining data that were sufficiently
detailed and rich to allow us to explore the phenomenon and identify and explain
themes and patterns regarding error incidence. We would then try to integrate these
explanations in an overall conceptual framework, thereby building up a theory of error
rate explanation. This we would test using evidence provided by existing data and new
data modifying our theory as necessary.
Combining research approaches
In the discussion of induction and deduction, you have probably gained the impression that there are rigid divisions between deduction and induction. This is not so. As
we have seen in our discussion of abduction, is it possible to combine deduction and
induction within the same piece of research. It is also, in our experience, often advantageous to do so although often one approach or another is dominant. It would be quite
usual, for example, to start with an exploratory study in order to arrive at a tentative
theory inductively before testing that theory in a deductive piece of quantitative work.
In our example of error-prone call centre operatives, a few focus groups may give some
clear indications as to the reason for some staff making more errors than others. These
indications may then form the basis for questionnaire design and administration. We
cover mixed-methods research later in the chapter.
Whether your reasoning will be predominantly deductive, inductive or abductive
depends on a number of factors – in particular, the emphasis of the research and the
nature of the research topic. If your topic enjoys the support of a lot of literature from
which you can define a theoretical framework, then it may be more suitable for a deductive approach. For a new topic about which there is little existing literature, it may be
more appropriate to adopt an inductive approach by collecting and analysing data and
reflecting upon what theoretical themes the data suggest. Alternatively, if your topic is
rich in information in one context but in the context in which you are researching
there is little information, this may persuade you to adopt an abductive approach, enabling you to modify an existing theory.
In addition, the time available to you may influence your decision about which
approach to adopt. The timescale for a piece of deductive research can be shorter,
although time will be needed to set up the study prior to data collection and analysis.
As data collection is often based on ‘one take’, it should be possible to predict the timescale. But inductive and abductive research can take much longer, and the timescale
can be more difficult to predict as the ideas have to emerge gradually.
Two other factors may determine which research approach you adopt. First is the
extent to which you are prepared to indulge in risk. Deduction can be a lower-risk

5.4 Differing purposes: exploratory, descriptive and explanatory studies 115
strategy, but with induction and abduction you have to live with the fear that no useful
data patterns and theory will emerge. Second, you have to consider your audience.
Most managers are familiar with deductive research methods and may doubt the reliability of data collected inductively.
Exploratory studies
Exploratory research is about discovering information about a topic that is not understood clearly by the researcher. This lends itself particularly well to new phenomena
where you may not be prepared to launch into a piece of full-scale research but want to
gain some insights that will inform your research design. The marketeer, for example,
may be aware that there is a change in the consumers’ preference away from organic
food after the initial enthusiasm for such food. But the marketeer may not fully appreciate what form that preference is taking and why the phenomenon is happening. An
exploratory study may well provide tentative answers to these initial questions, which
need to be followed up with more detailed research to provide more dependable answers.
The most usual ways of conducting exploratory research are:
searching the academic literature;
using unstructured observations;
using semi- and unstructured interviews.
The Internet is particularly useful for doing some basic literature search work in an
exploratory study. This may help in pointing you towards relevant academic journal
articles to be found on academic journal databases in your library such as Emerald
Insight and Business Source Premier. Of course, non-academic articles from the
Internet, such as those in newspapers or produced by commercial enterprises, should
not be thought of as an alternative to academic articles.
As well as literature searching, exploratory studies are well suited to qualitative methods such as semi- and unstructured interviewing or unstructured observation. For
example, informal discussions with consumers, organic food growers and supermarket
executives may provide useful insights into changing preferences for organic food.
Consumer focus groups are a well-known method of establishing consumer views
which may be valuable as a piece of exploratory research. Case studies of families’ food
consumption may be helpful in an explanatory study, as may small-scale pilot studies.
Differing purposes: exploratory, descriptive
5.4 and explanatory studies
Definition
exploratory study: research that aims to seek new insights, ask new questions and assess topics in a
new light.

116 Chapter 5 Choosing your research design
Although the methods used in exploratory research indicate a good deal of flexibility, they do not mean absence of direction to the inquiry. It means that the focus is initially broad and becomes progressively narrower as the research progresses.
While exploratory research provides insights into, and a fuller understanding of, an
issue or situation, definitive conclusions should be drawn only with extreme caution.
Yet it is very valuable in helping you to decide the best research design, data collection
method and selection of subjects. Of course, it may point to the fact that the issue which
you thought was of great importance is in fact a non-issue. In that case, you will have
saved yourself going up a blind alley and delivering a meaningless piece of research!
Descriptive studies
You may have received comments on your assessed assignments from lecturers, which
say something like, ‘It’s good as far as it goes, but it tends to be too descriptive’. This
means that they wanted you to analyse and explain what was going on with a given
topic. The question ‘Why did the situation occur?’ is the question to which they wanted
the answer – not the answer you supplied, which was likely to have been a description of
the situation. However good your description, the point remains that describing a phenomenon is much easier than explaining why it occurred. That said, descriptive research
certainly has an important role to play, often as the forerunner of explanatory research.
Descriptive study or research seeks to describe accurately persons, events or situations.
It is appropriate for asking such ‘what’, ‘when’, ‘who’, ‘where’ and ‘how’ questions as:
What is the employee absentee rate in particular departments?
When are employees most likely to be absent in those same departments?
Who are the employees who are most frequently absent?
Look again at the three questions above. You will note that they each require responses
than can be quantified. They each involve the collection of measurable, quantifiable
data. So, the data collection methods typically used in descriptive research are:
questionnaires;
structured interviews;
structured observation;
reanalysis of secondary data.
Descriptive research should be thought of as a means to an end rather than an end in
itself. This means that if your research project utilises description, it is likely to be a
forerunner to explanation. However, descriptive research can tell us a lot about the
world around us, which is very valuable in its own right, as the example in Research in
practice 5.1 indicates.
Definition
descriptive study: research designed to produce an accurate representation of persons, events or
situations.

5.4 Differing purposes: exploratory, descriptive and explanatory studies 117
The emergence of shopping ‘serial returners’ hinders growth
of UK businesses
Research in 2016 from payments company Barclaycard reveals that consumer demand
for free and easy returns when shopping online is placing increased pressure on retailers
and impacting their bottom line. The report notes the emergence of the ‘serial returner’ –
the online shopper who habitually over-orders and takes advantage of free returns.
Barclaycard found that:
Six in 10 retailers report that they are negatively impacted by consumers’ propensity
to return unwanted items.
Online-only businesses are hardest hit as three in 10 say managing returns is affecting profit margins.
One in five businesses have increased price of items to cover the cost of managing
and processing customer returns.
Four in 10 shoppers say standardising clothing and shoe sizes could help retailers
reduce their level of returns.
The findings indicate that in the previous 12 months, the increasing rate of returns has
presented a number of challenges for online retailers, with 31% claiming that managing the
returns process has an impact on their profit margin. This comes as online shopping continues to grow in popularity, with spending in digital channels rising 14.1% year-on-year in
2015, compared to just 1.1% in-store.
‘Serial returners’ regularly order more than they need with no intention of keeping
every item. Three in ten shoppers deliberately over-purchase and subsequently return
unwanted items, with 19% admitting to ordering multiple versions of the same item to
make up their mind at home – safe in the knowledge they can choose from the evergrowing number of ways to quickly and easily send items back, such as hourly courier
services and local drop-off points.
Six shoppers in ten say a retailer’s returns policy impacts their decision to make a
purchase online, and almost half (47%) of these would not order an item if they had to
fund the cost of sending it back from their own pocket. Consequently, web-based retailers are caught between trying to attract customers and remaining competitive while also
ensuring they protect their bottom line.
Fifty-seven per cent of retailers say that dealing with returns has a negative impact on
the day-to-day running of their business, leaving many with no choice but to find
another way to recover the cost incurred. A third of online retailers offer free returns but
offset the balance by charging for delivery, while one in five increase the price of items
to cover the cost of returns.
Concern about being able to afford the costs of managing the delivery and returns
process led to 22% of bricks-and-mortar retailers choosing not to sell online.
Thirty-eight per cent of returners said they would send back fewer purchases if businesses were to standardise clothing and shoe sizes, which can vary between and even
within retailers. Eighteen per cent said a better in-store experience, such as shortened
queues for clothing store fitting rooms so they can try on sizes without the wait, would
Research in practice 5.1

118 Chapter 5 Choosing your research design
The data in Research in practice 5.1 is valuable to online retailers in particular in giving them an idea of the scale of a problem. It also points to an explanation of why the
phenomenon may be occurring. But the strength of this piece of research is giving
retailers the basis upon which to formulate action to solve the problem of returns
while still retaining the goodwill of their customers and trading within the restrictions of consumer law.
Explanatory studies
Explanatory studies takes descriptive research a stage further by looking for an explanation behind a particular occurrence through the discovery of causal relationships
between key variables. As we saw in Research in practice 5.1, the question moves from
just ‘to what extent do customers return goods to retailers?’ to include ‘and why this is
happening?’ Answering such questions can use quantitative, qualitative or both types
of data. The methods you use to collect your data, therefore, depend to a large extent on
the focus of the research.
The methods typically used in explanatory research can typically include:
questionnaires;
interviews;
observations;
reanalysis of secondary data.
If the focus is on explaining the impact of different factors, such as the ease with which
returns can be made, quantitative values can be attached to the variables, and they can
be subjected to statistical tests such as correlation in order to get a clearer view of the
relationship (section 7.3). Alternatively, the focus may be on other possible explanatory factors, such as the reasons why product returns may vary by customer age and
location. Much of the explanation may centre upon differing attitudes or beliefs of
customers which may be difficult to quantify (section 7.4). Even if your feeling is that
qualitative research may be appropriate for your research questions and objectives,
remember, qualitative research may benefit at some stage from an element of quantitative processing of data.
also reduce the number of returns they make. In addition, 18% said they would like
retailers to introduce technology to help them better visualise an item when shopping
online, such as the ability to ‘try on’ clothing after uploading an image of themselves.
Source: Barclaycard (2016).
Definition
explanatory study: research that focuses on studying a situation or a problem in order to explain the
relationships between variables.

5.5 Differing strategies 119
Now we concentrate on the research strategies you may use. Let’s make it clear at the
beginning of this section that the label that is attached to a particular strategy is relatively
unimportant. What is important is that the strategy you choose will enable you to answer
your particular research question(s) and meet your research objectives. Nonetheless, the
labels are a useful way of categorising the different strategies available to you.
Each of the strategies can be used for exploratory, descriptive and explanatory
research. Some of these more clearly belong more closely to the deductive approach,
others to an inductive or abductive approach. But, as with the attachment of labels, it is
not helpful to allocate strategies to one approach or the other. We emphasise that your
research strategy will be guided by your research question(s) and objectives as well as
the extent of your existing knowledge, the amount of time and other resources you
have available, as well as your own philosophical leanings. Finally, we must point out
that these strategies are not mutually exclusive. So, for example, a case study may
include a survey and archival research.
The order in which we explain the strategies below is simply that which they appear
in the research onion (Figure 5.1). It does not imply that those explained earlier are any
more important than those below them in the list.
Experiment
The purpose of an experiment is to study causal links between variables; to establish
whether a change in one independent variable (e.g. the running of a sales promotion)
produces a change in another dependent variable (e.g. the level of sales). The essential
components of an experiment are the following:
1 Manipulating the independent variable. To assess the effect of a sales promotion on
the level of sales, the sales promotion may be manipulated by differentiating the
offer, altering the time period over which it runs or changing the level of advertising.
2 Controlling the experiment by holding all other independent variables constant.
Therefore, the sales experiment would be held at the same time of year, in the same
(or similar) location.
3 Observing the effect of the manipulation of the independent variable on the
dependent variable.
4 Predicting the events that will occur in the experimental setting.
5.5 Differing strategies
Definition
experiment: a research strategy that involves the definition of a theoretical hypothesis; the selection
of samples of individuals from known populations; the allocation of samples to different experimental
conditions; the introduction of planned change on one or more of the variables; and measurement on
a small number of variables and control of other variables.

120 Chapter 5 Choosing your research design
The steps in the experiment are the following:
1 Identify and define the issue that is to be studied (e.g. the effect upon sales of
promotions).
2 Formulate research hypotheses (e.g. the sales revenue of yogurt will double when a
‘Buy 2, get 1 free’ offer is run).
3 Design the experiment by:
(a) selecting the relevant product and sales promotion;
(b) identifying and controlling the factors which may affect the outcomes (e.g.
time, location);
(c) choosing the way in which the outcomes will be measured;
(d) conducting a pilot study to test the effectiveness of the experiment and predict
any problems which may be overcome by adjusting the experiment design.
4 Run the experiment and collect the data.
5 Assemble the data and apply relevant tests to ensure statistical significance of the
findings.
Although our sales promotion example is a situation where an experimental strategy
may apply in business, the strategy is not applicable to many business and management research questions. This may be for a number of practical reasons. For example, it
may be seen as unfair to apply certain disadvantageous working conditions to one
group of employees and not the other. Moreover, it would be unethical if the disadvantaged employees were not told of the experiment and the way it would involve them.
Seeking volunteers may be problematic, as some people are unwilling to participate in
experiments and so those who volunteer may not be representative of the population
you wish to study. This is why the experiment strategy is often used only on captive
populations such as university students! In addition, if the results are to be statistically
valid, then an experiment that involves a large group of subjects may be necessary with
all this implies in terms of cost and complexity.
Survey
When you mention the term ‘research’ to most people, it is often the survey strategy, usually using a questionnaire, that springs most readily to mind. Few days pass without the
news reporting the results of a new ‘survey’ that indicates a snapshot of social and economic life. Indeed, the notion of the researcher selecting a sample of respondents from a
population and administering a standardised questionnaire is the image with which you
may be most familiar. The survey strategy is popular in business and management research.
Definition
survey: a research strategy which involves the structured collection of data from a sizeable population. Data collection may take the form of questionnaires or structured interviews.
5.5 Differing strategies 121
Because it is so widespread, managers find it easy to understand and place a good deal of
faith in the results which flow from surveys. The survey strategy is particularly suitable for
asking questions such as: ‘Who?’ ‘What?’ ‘Where?’ ‘How much?’ and ‘How many?’ These
types of questions make them useful for exploratory and descriptive research. One of the
reasons the survey strategy is so popular is that it allows the collection of data about the
same things from a large number of people in a cost-effective manner. This is because sampling (section 6.2) makes it possible to generate findings that are representative statistically
of the whole population at a lower cost than collecting the data for the whole population.
The most common method for collecting data using a survey strategy is the questionnaire. The questionnaire comprises a set of standardised questions. It may be completed by the respondent or an interviewer in a face-to-face situation. Alternatively, it
may be completed online or by telephone. The standardised questions make it possible
to easily compare responses across different locations or time frames.
It would be wrong to think that administering a questionnaire is quick and easy.
Ensuring that your sample is representative, designing and piloting your data collection instrument and trying to ensure a good response rate are all very time consuming.
So is the analysis of the data, even though you will undoubtedly use a spreadsheet or
statistical analysis software. But from a practical viewpoint, the survey strategy does
offer you certain advantages. These come largely from the feeling of being in control of
the process. With more qualitative strategies, for example interviewing, you are often
dependent upon others for their time. The fact that you will be more in control of your
time schedule in the survey strategy makes it an attractive proposition for students who
usually have a set period of time, often an academic year, to complete a piece of research.
One of the drawbacks of the survey is that the data collected are unlikely to be as detailed
as those collected by other research strategies. Clearly, it is not advisable to ask a large number of questions of a large number of people, so comprehensiveness is bound to be limited.
We have all switched off when faced with the fourth page of a questionnaire, not a reaction
which will lead to credible research results! But perhaps the biggest disadvantage with the
questionnaire method is the attraction that it seems an easy option. Be warned: designing
a questionnaire is very easy; designing a good one is enormously difficult.
It would be misleading if we did not mention that the survey strategy is also associated with structured observation and structured interviews (see Chapter 6), both of
which are based on the principle of standardisation.
Case study
We noted in the previous sub-section that the survey method is suitable for asking questions such as ‘Who?’ ‘What?’ ‘Where?’ ‘How much?’ and ‘How many?’ In contrast, the
case study strategy is more appropriate for asking the question ‘Why?’ (although the
Definition
case study: a research strategy which involves the investigation of a particular contemporary topic
within its real-life context, using multiple sources of evidence (data).

122 Chapter 5 Choosing your research design
questions ‘What?’ and ‘How?’ are also relevant). Consequently, the case study strategy
is most often used in exploratory and explanatory research.
Case studies are particularly good at enabling the researcher to get a detailed understanding of the context of the research and the activity taking place within that context. So, if you are concerned with understanding why managers make decisions in
certain ways rather than an analysis of the decisions that are made, who makes them,
the frequency of decisions and their perceived importance, the case study may be the
best choice. The key word, of course, is ‘context’. Often the key to explaining social phenomena is context. For example, managerial decisions taken in an environment where
senior managers are highly supportive of their managers and reasonably tolerant of
their failure are likely to be much more adventurous than in an unsupportive climate.
Data collection techniques used in a case study may be varied and include a combination of interviews, observation and documentary analysis as well as questionnaires.
Case studies normally use a variety of methods along with secondary data, so you are
likely to need to triangulate multiple sources of data, a principle which we explain in
the last section (5.6) of this chapter.
The question arises as to whether a single case or a number of cases is most suitable
for your research. As with many other decisions related to strategy, this is likely to be
governed by what is ideal as a strategy for answering your research questions and what
is possible in terms of practical considerations such as access and resources. Like many
students, you may choose the organisation in which you work. This is often going to be
a practical option. Yet there are likely to be limitations on the extent to which you can
call your organisation typical of all similar organisations. For this reason, a case study
strategy can also incorporate more than one case. You may choose to use more than one
case if you need to generalise from your findings.
Some criticise the case study strategy because they feel that one case, or even a
small number of cases, is no basis for placing faith in the findings. They also argue
that close exposure to the study of the case biases the findings. For this reason, they
dismiss case study research as useful only as an exploratory tool. But a well-designed
and skilfully executed study of real-life issues will yield insights not possible in more
descriptive strategies.
Action research
In one application of the case study strategy, you may be working in one organisation
in order to gain insights into an aspect of that organisation’s life from which you may
draw conclusions. Similarly, with
action research, you may be working in one organisation. But with the case study, the likelihood is that you are ‘on the outside looking
in’ (although you may be a part of that organisation). With action research, the
Definition
action research: research strategy concerned with the management of a change and involving close
collaboration between practitioners and researchers.

5.5 Differing strategies 123
emphasis is on you taking an involved role as a participant in teasing out the issues,
understanding the organization and the project and acting upon what has been
found out (Figure 5.2); you are very much the insider. Here you are conducting experiments and acting upon them by making changes and observing the results. Your role
is decidedly active rather than passive.
There are four common themes within action research which make its practical
application, and value, much clearer:
1 As noted above, the purpose of the research is to be part of and study research in
action rather than to conduct research about action. A typical emphasis for the
research may be the change process in an organisation.
2 The joint working of organisational members, for example managers and other key
workers, and researchers in the action research project. The researchers may be pursuing academic research, or they may be consultants from within or outside the
organisation. So the researcher is part of the organisation and the change process.
This contrasts with more normal research or consultancy where participants are
objects of study.
3 The cyclical nature of the process of action research which consists of diagnosing,
planning action, taking action and evaluating action (see Figure 5.2). The diagnostic
stage will begin with the establishment of a clear project purpose and continue with
fact-finding and analysis on the situation under study. The core of the cycle involves
the planning, implementation and evaluation of the changes themselves.
Cycle 1:
Teasing out
the issues
Evaluating action Planning action
Context and
purpose
Diagnosing
Taking action
Cycle 2:
Understanding
the organisation
and project
Evaluating action Planning action
Diagnosing
Taking action
Cycle 3:
Acting on
knowledge
Evaluating action Planning action
Diagnosing
Taking action
Figure 5.2 The action research spiral
Source: © Saunders et al. (2016), reproduced with permission.
124 Chapter 5 Choosing your research design
4 The final stage in the cycle is the key one if you are using the action research strategy as
an academic project. This is the evaluative stage, where you will draw from the research
the lessons which may be learned. These lessons may have a practical dimension (for
example, the pitfalls that managers should avoid when introducing a new pay system)
or a more theoretical perspective (say, the development of a model of employee behaviours in the pay change process). It is clear here that action research should have implications beyond the immediate project. It must have a wider application.
You will see that action research differs from other research strategies because it is
concerned with action, in particular the promotion of organisational change. So it
is particularly useful for ‘How’ questions. A project using action research promises
excellent insights into organisational change processes as well as an absorbing project
report which, because few student projects use this strategy, will have a distinct novelty
value! You may ask why few students pursue action research. The answer is that it is
demanding both in terms of hours that need to be devoted to a project, and the length
of time needed to progress the action research cycle. It also demands a high level of
researcher skill and maturity.
Grounded theory
Grounded theory, usually associated with Corbin and Strauss (2008), belongs principally
to the inductive approach to research because you develop theory from data generated
by a series of observations or interviews. However, there is an element of deduction
(remember – theory to data) in this strategy.
Unlike the deductive approach, data collection starts without the formation of your
initial theory, as you would in a deductive approach. Theory is developed from data
generated by a series of observations, discussions and interviews. The element of deduction is introduced when these data lead to the generation of predictions which are then
tested in further observations. These further observations may confirm, or otherwise,
the predictions. A continual process of testing leads to the development of theory.
This all becomes a bit clearer with a practical explanation of how grounded theory
works, as you will see in the Research in practice 5.2 example.
Ethnography
Ethnography is concerned with understanding another way of life from the perspective
of those pursuing that way of life. Its traditional roots are in the anthropological study of
primitive societies. The key concern is learning from people rather than studying them.
Definitions
grounded theory: research strategy in which theory is developed from data generated by a series of
observations or interviews principally involving an inductive approach.
Ethnography: research strategy that focuses on describing and interpreting the social world through
first-hand field study.

5.5 Differing strategies 125
The use of grounded theory in a study of consumer behaviour
Jessica was a full-time marketing student who was doing a placement in a major retail
group where she performed a range of different duties. The group was conducting a
marketing trial with the objective of generating higher turnover of their more expensive
(and more profitable) food items. This was part of a wider strategy to achieve a more
exclusive market position for their brand.
One of the activities which formed part of the marketing trial was to offer customers
samples of the range of food items (e.g. luxury chocolates) while they were doing their
normal shopping. Having sampled the item in question, the customers were then invited
to purchase it at a ‘special opening offer’ price. The range of behaviours exhibited by the
customers who sampled the items ranged from indifference, resulting in a refusal to
buy, to enthusiasm and a purchase of the offer.
The managers in charge of the marketing trial thought that the difference in sales
success level of the trial food items was due to the items themselves, a clear pattern having emerged where some items were more successful than others. However, Jessica was
not so sure that this was the sole answer. At university she was particularly interested in
consumer behaviour, and she thought that the explanation of the different success levels of the trial items was more complicated than the intrinsic attractiveness of the product. She kept a diary throughout the trial and noted, for example, that some products
were received with mild enthusiasm by customers, but nonetheless a sale resulted. On
other occasions, an item was received with great keenness but no sale followed.
As the trial progressed, she talked with all the staff involved in sampling the customers and developed her notes to the point where a number of patterns began to emerge.
One of these related to consumer decision-making theory, in particular that of impulse
buying. As the trial neared its end, Jessica developed a tentative explanation of the differing success levels of the food items based upon the idea that some lent themselves to
impulse buying more than others, but just as importantly, some customers were more
inclined to make pure impulse purchasing decisions (‘Oh, why not?’) while others were
much more likely to make planned impulse decisions (‘I had in mind to buy some of
these . . .’).
In the final stages of the trial, Jessica took more detailed notes of the behaviour of
the customers with her developed theory in mind, and returned to university at the
end of the placement with a clear theoretical position to take into the final stages of
her project.
Research in practice 5.2
Such studies in business research are unusual. But one stands out as one of the most
famous of the late twentieth century. It was published in a book called
Working for Ford
by Huw Beynon (1973). Beynon spent much of 1967 in Ford’s Halewood plant on
Merseyside talking to workers, trade union officials and management. The picture
Beynon paints of life on the production line stems largely from the words of the workers themselves: an approach which really brings Beynon’s writing to life. Beynon concluded that the work was dull and boring and the whole experience pretty bleak. In his

126 Chapter 5 Choosing your research design
view, Ford was still then run on the principles established by its founder, Henry Ford –
principles which emphasised that man is sub-servient to the machine.
Ethnography is obviously very time consuming, since it takes place over an extended
time period, so it tends not to be used extensively in business research. Indeed, Beynon
started his research in the pursuit of a PhD, but had to abandon the plan simply because
he had so much data. He decided to write a book instead.
Although ethnography is time consuming, the grounded theory example in
Research in practice 5.2 has some similarities with ethnography. If you are conducting
research similar to that in this example, you may wish to look at the ethnographic strategy in more detail.
Archival research
Archival research uses administrative records and documents as the principal source of
data. These may, for example, be minutes of meetings, memos or emails containing
information or instructions, accounts, contracts or letters. For example, you may wish
to study the way in which key messages about company performance are communicated to staff.
Such research concentrates on past events and may allow changes over time to be
charted. But the extent to which you use the archival strategy will depend on the availability of key documents. Of course, even if relevant documents exist and you are
allowed access to them, they may not meet your precise research needs.
Archival research may not be appropriate for many business research projects as the
main research strategy. But it may play a useful supporting role at some stage in many
projects if you wish to supplement other data collection methods.
Narrative inquiry
A narrative is a story; a personal account which interprets an event or sequence of
events. Qualitative research interviews can involve a participant in ‘storytelling’. So
the term ‘narrative’ can be applied to describe the nature or outcome of a qualitative interview. But as a research strategy, narrative inquiry has a more specific meaning and purpose. It may be that your research context leads you to believe that the
experiences of your participants are best revealed by collecting and analysing their
contributions as complete stories, rather than discrete bits of data that are the result
of specific interview questions which you then treat as bits of data in your subsequent analysis.
Definitions
archival research: research strategy which analyses administrative records and documents as the
principal source of data.
narrative: account of an experience that is told in a sequenced way, indicating a flow of related
events that, taken together, are significant for the narrator and which convey meaning to the
researcher.

5.5 Differing strategies 127
The aim of narrative inquiry is to gain a deeper understanding of organisational realities, closely linked to their members’ experiences. As such, the role you, the researcher,
adopt in narrative inquiry is that of the listener facilitating the process of storytelling
by the participant (or group of participants – it is quite legitimate to have more than
one narrator) who is the narrator. The narrative which results may be a short story
about a specific event, a more extended story (for example, about a work project) or a
career history.
In-depth interviews are the main way of collecting narratives. But there may be other
ways: for example, participant observation, autobiographies, authored biographies or
diaries. This raises the issue of the narrative researcher adopting the role of narrator in
particular circumstances, which we will consider further later.
As with other research strategies, narrative inquiry may be used as the sole research
strategy, or it may be used in conjunction with other strategies to lend richness to your
data.
Combining research strategies
We made the point earlier that it is perfectly possible to combine research approaches
and strategies within the same study. You may wish to do some preparatory work in an
exploratory phase before you firm up your research questions and objectives. This may
involve informal discussion or interviews with key personnel. This could be supplemented with an element of scanning of organisational documentation. The exploratory strategy could then be continued with survey work using both questionnaire and
structured interviews. A final stage may utilise the explanatory strategy by attempting
to gain greater meaning to the exploratory work through, for example, a case study
approach. The important point to bear in mind is that different strategies may work at
different stages in the study, depending, of course, on their suitability to answer your
research questions and objectives.
Single, multiple or mixed research methods?
Within your overall strategy, as you can see from the research onion in Figure 5.1, you
will face a choice of whether to choose a single method approach to your work or use a
multiplicity of methods. For example, a quantitative research design may use a single
data collection technique, such as a questionnaire (mono method quantitative study)
or combine a questionnaire with structured observation (a multi-method quantitative
study). Alternatively, you may choose to base all your data collection on focus groups
(mono method qualitative) or combine focus groups with follow- up single interviews
(multi-method qualitative). You might also choose to start with semi-structured interviews and use the data collected from these interviews to help design a questionnaire
(mixed methods). Mixed methods may be simple (e.g. within a narrow time frame) or
complex (over a longer time frame) – see later in this chapter for a discussion of crosssectional and longitudinal research.
Business and management research lends itself to the use of multiple and mixed
methodological choices because these can overcome the drawbacks of using a single

128 Chapter 5 Choosing your research design
method. In addition, they promise deeper and richer data than may result from use of a
single method.
Combining approaches and strategies in the pursuit of an answer to your research
questions and objectives will usually involve the mixing of research methods. The preceding section on combining research strategies indicates that certain approaches and
strategies are better suited to particular stages of the research than others. In the same
way, using a variety of data collection methods may present advantages for the following four reasons.
Reason 1: Some data collection methods are more suited to particular tasks
than others
If you wish to establish the extent to which children use Snapchat, this can be done by a
questionnaire survey to a sample of children; if you want to understand why some teams
work better than others, it is better to use qualitative discussions and interviews. Those
two examples are, of course, separate projects. But the same argument applies when you
are thinking about one project. Your research objectives may be: ‘To what extent do
Internet-based suppliers use a multiplicity of carriers for shipping goods?’ and ‘What are
the reasons for the use of multiple carriers?’ The first question may collect data through
an Internet-based scan by you of leading supplier websites, followed by a questionnaire
of a sample of suppliers. You may follow this up with some interviews with the distribution managers of suppliers in order to get a grasp of the reasons for multiple-carrier
usage. The different methods are then integrated into a single research design.
Reason 2: Focusing on different aspects of the study
It may well be that different data collection methods may be better at different research
tasks within the study. So you may want to gain an overall view about the way in which
companies outsource aspects of their business from studying secondary sources, or
even from a quantitative survey. This will probably not give you a sufficient grasp of the
reasons why some companies are more likely to be more enthusiastic about outsourcing
than others. Nor will it tell you why some functions are greater candidates for outsourcing than others. This may need some interviews with key executives.
Reason 3: Corroborating your research findings within a study using two or
more independent sources of data or data collection methods
When you have finished this chapter, we hope you will have a very clear understanding
of the importance of establishing the credibility of your research findings. This topic is
dealt with in this chapter’s final section (5.6). One way you can do this is through the
process of triangulation. Triangulation is the use of two or more independent sources of
data or data collection methods within one study in order to help ensure that the data
are telling you what you think they are telling you. In an ideal world, a questionnaire,
say, will yield highly credible data, particularly if the guidelines to effective survey
design and implementation are followed. But if you have triangulated your questionnaire findings with some semi-structured interviews and find that these findings have
been confirmed, then you will feel more content in the knowledge that you have done

5.5 Differing strategies 129
all you can to provide credible results. Of course, it may be that both sets of data are
questionable, so even triangulation does not provide final confirmation.
As well as using two or more data collection methods to triangulate your findings, it
is possible to vary aspects of the data collection while using the same method. These
aspects may be the populations from which you are collecting the data, the time at
which the data are collected and the location of the data collection. For example, if you
are studying the behaviour of consumers in a supermarket, data collected through
observation may differ from town to town, or between mornings and afternoons. If
data are broadly the same, this suggests that the findings are robust. Differences may
point to the need for more research. If, say, consumers spend more time making buying
decisions in the afternoons, this may suggest less busy consumers. Alternatively, it may
be that there are fewer customers in the afternoon, which in itself prompts more time
to be taken by consumers. So triangulation in this case may not only cause concern
over initial findings; it may also indicate the need for a revision of initial conclusions.
It may also be fruitful to triangulate the research by using the same data collection
methods and sources of data but changing the data gatherer. So if you are the research
manager of a large market research company employing many interviewers, and you
notice that one set of data is seriously inconsistent with the others, this may tell you
more about the interviewer than sources of the data. It is not unheard of for some data
to be invented! But, more usually, two individual researchers conducting semi-structured interviews with similar subjects may generate slightly different data, at least in
emphasis, because of, say, the stress put on certain questions or the way in which
answers have been followed up.
Reason 4: Using qualitative methods to explain relationships between
quantitative variables
Think again about the sort of research findings you hear in the news every day. We
often hear that people in Northern Europe are heavier consumers of alcoholic drinks
than those in the South of Europe. This statistic may be of interest as far as it goes, but it
doesn’t go far enough. What is it about living in Northern Europe that leads to greater
alcohol consumption? Is it colder? Are alcoholic drinks cheaper? We don’t know from
the broad statistics. We need to ask some searching questions to find out what is leading
to this particular phenomenon.
Differing time horizons: cross-sectional or longitudinal?
If you are like most of our readers, you will be time-constrained in the completion of
your research project. For this reason, you will probably opt for a piece of cross-sectional
research. This is a ‘snapshot’ of a particular research setting at a particular time. On the
other hand, a research design which tracks events over time is called a ‘longitudinal’
design. The main difference between the two designs is the ability of the longitudinal
design to note change. Which you choose will probably depend on practical considerations. However, we will show that an element of longitudinal design is perfectly possible within the constraints of one academic year.

130 Chapter 5 Choosing your research design
Cross-sectional studies
In a cross-sectional research design you would collect data from participants at only
one period in time in what is often termed a ‘snapshot’. The data are typically collected
from multiple groups or types of people in cross-sectional research. For example, data
in a cross-sectional study might be collected from male and female consumers, from
those of different ages, from people in different socio-economic classes or from consumers with different levels of educational achievement. A classic example of the
cross-sectional research design is the opinion poll. It is a way of taking the ‘snapshot’
of current thinking.
In an opinion poll, the cross-sectional study would usually employ the survey strategy and use a questionnaire to collect quantitative data. As well as describing the incidence of a phenomenon, you can explain statistically the relationship between certain
variables: for example, the relationship between expenditure on sales promotions and
sales revenue. But a cross-sectional study may also use qualitative methods. For example, interviews conducted over a short time period, either as the sole data collection
method or as one of a mixture of methods, may be equally useful.
Longitudinal studies
As we mentioned earlier, the main advantage of a longitudinal study or research design
is the capacity that it has to study change and development over time. This may conjure up images of annual visits to managers to track their progress as part of a coaching programme. It could also be that the manager keeps a diary of relevant thoughts
and actions, much the same as the personal development journal that you may have
come across. These may be powerful ways of collecting data to chart change. But they
may be impractical for the student researcher, given time (and, possibly, access) constraints. However, what may be useful is to introduce a longitudinal element to your
research by using one of the secondary data sources produced by longitudinal surveys
that are available on the Internet, such as the Family Expenditure Survey (see Research
in practice 5.3).
Definitions
cross-sectional research: study of a particular topic at a particular time, i.e. a ‘snapshot’.
longitudinal study: study of a particular topic over an extended period of time.
The UK Labour Force Survey
The Labour Force Survey (LFS) is a survey of households living at private addresses in
the United Kingdom. Its purpose is to provide information on the UK labour market
which can then be used to develop, manage, evaluate and report on labour market
Research in practice 5.3
5.6 Making sure your research conclusions are believable 131
policies. The survey is managed by the Office for National Statistics in Great Britain and
by the Central Survey Unit of the Department of Finance and Personnel in Northern
Ireland on behalf of the Department of Enterprise, Trade and Investment (DETINI).
Since 1992, the LFS in Great Britain has run as a quarterly survey (1994–95 for
Northern Ireland). The quarterly surveys have until spring 2006 operated on a seasonal
quarterly basis. However, mostly due to a European Union requirement, in May 2006
the LFS moved to calendar quarters. The 2006–7 data is the first set of HSE data based
on the LFS to be affected by this change.
The LFS is intended to be representative of the whole population of the United
Kingdom, and the sample design currently consists of around 41,000 responding households in every quarter. The quarterly survey has a panel design whereby households stay
in the sample for five consecutive quarters (or waves), with a fifth of the sample replaced
each quarter. Thus there is an 80% overlap in the samples for each successive survey.
An extract of the summary of the data for the United Kingdom presented in the
October 2016 LFS report is given below:
Between March to May 2016 and June to August 2016 the number of people in
work and the number of unemployed people increased. The number of people not
working and not seeking or available to work (economically inactive) fell.
There were 31.81 million people in work, 106,000 more than for March to May 2016
and 560,000 more than for a year earlier.
There were 23.23 million people working full-time, 362,000 more than for a year
earlier. There were 8.58 million people working part-time, 198,000 more than for a
year earlier.
The employment rate (the proportion of people aged from 16 to 64 who were in
work) was 74.5%, the joint highest since comparable records began in 1971.
The unemployment rate was 4.9%, unchanged compared with March to May 2016
but down from 5.4% for a year earlier. The unemployment rate is the proportion of
the labour force (those in work plus those unemployed) that were unemployed.
The inactivity rate (the proportion of people aged from 16 to 64 who were economically inactive) was 21.5%, the joint lowest since comparable records began in 1971.
Average weekly earnings for employees in Great Britain in nominal terms (that is, not
adjusted for price inflation) increased by 2.3% both including and excluding bonuses
compared with a year earlier.
Source, and more details from: https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/uklabourmarket/october2016
When your supervisor or examiner sits back and fixes you with an inscrutable gaze and
asks you, ‘Why should I believe your theory and/or conclusions?’ you should know that
all the careful thought you have put into designing your research strategy is going to
5.6 Making sure your research conclusions are believable
132 Chapter 5 Choosing your research design
pay off! All the research design sophistication in the world is of no significance unless
you can say with calm authority, ‘I have done all I can to ensure that my conclusions are
valid’. Well, you cannot guarantee that your conclusions will be valid, but you can do all
you can to limit their invalidity.
One way we have used to help people to think about the validity of their research
conclusions is to perform the ‘reverse test’ in examining their validity. This starts with
questioning the validity of the theory developed and/or conclusions in relation to the
findings that gave rise to them and working backwards from there. So, the key questions are the following:
Does the theory developed, or do the conclusions, flow logically from the findings?
Are the summarised findings consistent with the data collected and presented?
Is it plausible to assume that the data are such that they would have been collected
by the methods stated?
Are the methods employed those that you would expect to find in the research strategy that has been articulated?
Is there coherence between the research strategy and the research questions and
objectives?
Let’s examine each of these questions in turn.
Does the theory developed, or do the conclusions, flow logically from the
findings?
You may not end your research with an overall theory. This may not be required.
Even if it is not, it may be implicit in what you have written. Let’s say that your overall conclusion is that the quality of service in fast food restaurants is the result of
effective staff supervision. This sounds reasonable enough. But one of your key conclusions may be that restaurants with high levels of consumer satisfaction reported
in consumer surveys are those that have the most experienced supervisory staff, as
measured by length of service. Does it necessarily follow that service quality is the
result of effective staff supervision? Here you are equating ‘length of supervisor service’ with ‘effective staff supervision’. It may be a valid assumption if those ineffective supervisors have been weeded out and only effective supervisors remain. But
you cannot assume it.
Are the summarised findings consistent with the data collected and
presented?
Your presentation of data may lean heavily towards lots of tables and graphs
depicting the quantitative data, which you have collected. If this is so, the findings about data representing consumer satisfaction and length of supervisor service must be correlated by the restaurant to support the findings that suggest a
link between the two. This implies the need for appropriate statistical techniques,
which are covered in section 7.3. In the case of qualitative data, the need for a
clear relationship between data and findings suggests the need for skilful and

5.6 Making sure your research conclusions are believable 133
convincing argument. This will ensure that the findings are supported by a narrative which persuades the reader that you have the data to substantiate the
findings.
Is it plausible to assume that the data is such that they would have been
collected by the methods stated?
Overstatement is a sin punishable by failure! This is particularly an issue with regard
to qualitative data collection. For example, it is easy to fall into the trap of making
extravagant claims about the data collected when it is based on a very small number
of interviews. Let’s go back to the example of effective supervision in fast food restaurants. If you have made claims for this as one of your findings, it does raise questions such as ‘What is effective supervision in this context?’ ‘What methods were
used to assess the extent to which it is practised?’ It is difficult to see how a small
number of interviews would enable you to collect data on effective supervision.
Some first-hand observation may be required. But even if interviewees are used, it
raises questions such as ‘Who was interviewed?’ ‘What were the interviewees asked?’
and ‘Did the interviewees have sufficient knowledge to be able to answer the questions credibly?’
Are the methods employed those that you would expect to find in the
research strategy that has been articulated?
So, you have said in the early part of your research methodology chapter that you are
going to pursue a piece of qualitative research using the inductive approach. When
your readers look at what you have done, or what you propose to do, they see quantitative questionnaire data or results based on tightly structured observation. This does not
mean that what you have done in employing these methods is wrong. But it does suggest presentational incoherence and, worse, muddled thinking.
Is there coherence between the research strategy and the research questions
and objectives?
This is the fundamental question and underpins all of this chapter’s content. It is a
question of ‘fit for purpose’. In this case, the purpose is ‘credible research findings and
conclusions’. If the consumers of your research cannot have faith in what you have
said, then what you have done will be of poor quality and questionable utility. So, the
question becomes ‘Which research strategy will enable me to answer my research
questions and meet my research objectives in a way which will yield results that can be
depended upon?’
Validity and reliability: what is the difference between the two?
Above, we use the term ‘validity’ in relation to the credibility of research findings and
conclusions. It is an important concept in designing a research strategy. So is reliability.
We now explain what we mean by these terms and what practical dangers exist in
research strategy design which may threaten the credibility of your research.

134 Chapter 5 Choosing your research design
Table 5.1 Principal factors which threaten the validity of research findings
and conclusions
Factor Refers to
Subject selection The biases that may result in selection of particular research
subjects which may be unrepresentative of the research
population.
History Specific events which occur in the history of the project (for
example, between first and second phases of the research)
which have an important effect on findings.
Testing Any effects that the data collection process itself may have on
the subjects (e.g. participants keen to impress the interviewer).
Mortality The loss of subjects during the research: this is a particularly
important issue for the conduct of longitudinal research.
Ambiguity about causal
direction
Confusion over the direction in which the flow of cause and
effect runs: for example, are poor call centre operator
performance ratings caused by a negative attitude towards the
way their performance was rated, or were the poor ratings
causing the negative attitude?
Definition
validity: extent to which (a) data collection method or methods accurately measure what they were
intended to measure and (b) the research findings are really about what they profess to be about.
Validity
Put at its simplest, validity is concerned with whether the findings are really about
what they appear to be about. Obviously, it is a crucial factor in research strategy
design. This is because any research can be affected by different kinds of factors which
can render your findings invalid. The practical message here is that you must eliminate all factors that threaten the validity of the research. The principal factors are
listed in Table 5.1.
Table 5.1 refers to ‘internal’ validity. Another type of validity is ‘external’ validity.
This refers to the extent to which your conclusions are generalisable to other
research settings. So, for example, if your research is set in one organisation, there
is obviously a question about the degree to which your conclusions are generalisable to other organisations. The simple response to this possible problem is to
increase the number of organisations in which you are conducting your research.
However, your concern may not be to produce a theory that is generalisable to all
populations but to try and explain what is going on in your particular research setting. This is fine, providing that you do not claim that your results, conclusions or
theory can be generalised.

Summary 135
Reliability
If your research is to be reliable, it must employ data collection methods and analysis
procedures which produce consistent findings. Such consistency refers to the degree to
which:
the measures you use will produce the same results if used on other occasions;
other researchers, when using the same methods and procedures in the same way,
will produce similar results;
those interpreting your research can see clearly how you came to your conclusions
from the data you collected.
As with validity, there are factors which threaten the
reliability of your research findings
and conclusions. These are listed in Table 5.2.
Table 5.2 Principal factors which threaten the reliability of research findings
and conclusions
Factor Refers to
Subject error Measurement which may take place at different times: for example, a
questionnaire administered to night-shift workers may produce
significantly different results to day-shift workers.
Subject bias Research subjects giving you unreliable information because they think
that telling the truth may, for example, show them in a bad light.
Observer error The way in which different researchers may, for example, ask the same
questions in different ways, thus biasing the results.
Observer bias The way in which different researchers may interpret the same data in
different ways, thus biasing the findings and conclusions.
Definition
reliability: extent to which data collection methods and analysis procedures will produce consistent
findings.
Considering your research philosophy helps you to ‘think about thinking’.
The main research philosophies are positivism, critical realism, interpretivism, postmodernism and pragmatism.
The principal approaches to theory development are deduction, induction and
abduction.
Three main purposes of research are exploratory, descriptive and explanatory.
The main types of research strategy are experiment, survey, case study, action
research, grounded theory, ethnography, archival and narrative inquiry.
Summary
136 Chapter 5 Choosing your research design
There can be significant advantages in combining methods in research design.
Research studies may be cross-sectional or longitudinal, or combine an element of
both.
Applying the ‘reverse test’ to the main theory and/or conclusions of the research
report is a way of thinking about the research’s credibility.
There are a variety of threats to research validity and reliability which need to be
controlled.
Look again at the material on deduction induction and abduction in this chapter.
Which of these approaches do you feel (a) appeals to you most, and (b) fits your
research question(s) and objectives?
Go back to your research question(s) and objectives and think about which of the
research strategies is most appropriate. Look again at any studies which are similar to
yours and see which strategies they have used and the reasons the authors have given
for their choice.
Think of your methodological choices. In what way may you combine different
research methods in your study? What advantages may there be in using multiple or
mixed methods?
How might longitudinal and/or cross-sectional time dimensions be used in your
design?
Make a note of all the threats to reliability and validity contained in your research
design.
Thinking about your research design
Barclaycard (2016). The emergence of ‘serial returners’ – online shoppers who habitually over
order and take advantage of free returns – hinders growth of UK businesses. Available at:
https://www.home.barclaycard/media-centre/press-releases/emergence-of-serial-returnershinders-growth-of-UK-businesses.html [Accessed 20 October 2016].
Beynon, H. (1973).
Working for Ford. London: Allen Lane.
Corbin, J. and Strauss, A. (2008).
Basics of Qualitative Research: Techniques and Procedures for
Developing Grounded Theory,
(3rd ed.). London: Sage.
Office for National Statistics (2016).
Labour Force Survey. Available at: https://www.ons.gov.
uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/
uklabourmarket/october2016 [Accessed 20 October 2016].
Saunders, M., Lewis, P. and Thornhill, A. (2016)
Research Methods for Business Students (7th ed.).
Harlow: Pearson Education.
Van Maanen, J., Sørensen, J.B. and Mitchell, T.R. (2007). The interplay between theory and
method.
Academy of Management Review, 32(4), 1145–54.
References
Chapter 6
Collecting data
You may be thinking, ‘I don’t need to read this chapter’ or ‘I can read this chapter later
when I need to collect my data’. It might be, as we mentioned in Chapter 4, that your
university expects you to use only secondary data for your project and so you feel there is
little point in reading this chapter. Alternatively, your university may only require you to
write an extended essay or literature review, and so you believe reading a chapter on collecting data will be of limited, if any, use. Or, you might be thinking that, as you have only
just started reviewing the literature for your project, it is too early to start learning about
collecting data. However, if you are thinking that you don’t need to read this chapter on
collecting data now, this is not the case.
To assess the suitability of secondary data for your own research project, you need to
understand the methods that were used to collect it (section 4.5). Even if your project is
an extended essay or literature review, you still need to read this chapter and learn about
different methods of collecting data as early as possible. Without this knowledge, you
will not be able to evaluate fully the journal articles, reports and book chapters you
review (section 2.6). This will reduce your ability to assess the quality of the data used. It
will also lessen your ability to evaluate the quality of the findings and conclusions. Even
if you are not collecting your own data, you need to know about methods for collecting
data. This will help you to infer the quality of potential secondary data for your project,
the quality of research reported in the literature you read and the value of both to your
own project.
The overall purpose of this chapter is to enable you to answer the question ‘Do these
data enable the research question to be answered?’ By being able to answer this question,
you will be able to do the following:
Assess the value of secondary data to meeting your own research question and
objectives.
6.1 Why you should read this chapter
138 Chapter 6 Collecting data
Assess the suitability of data used in research you read about in the literature and form
judgements about the quality of that research.
Ensure that you use appropriate data collection methods to collect data to meet your
own research question and objectives.
In this chapter, we start by talking about different ways of selecting samples from which
data will be collected. We then discuss how particular methods are suited to collecting different sorts of data, considering three frequently used methods of collecting primary data
in more detail. In this we look at how to design and distribute questionnaires, including the
use of Internet questionnaires; how to design and conduct face-to-face, telephone and
Internet-mediated interviews; and how to undertake structured and unstructured
observations.
Throughout your life, you have been selecting samples. Before you download a whole
album of songs from iTunes or purchase an MP3 download from Amazon, you will
probably listen to one or two tracks from that album to see if you like them. When you
were applying to your university for your course, you probably listened to opinions
from some current students and lecturers about that course before making your application. You were basing your decision to purchase on feelings after listening to a few
rather than all of the tracks on the album, and your decision to apply on the opinions
of some rather than all students and lecturers for that course. Each of your decisions
was based on a
sample (sub-group) of the population (complete set) that was likely to be
available to you. Yet it is unlikely that you spent much time considering your reasons
for using these samples rather than different samples, or even the whole population.
You also probably did not spend much time thinking about whether the sample you
selected was the most appropriate to help you make your decision. These aspects are
crucial for all research projects. You need to consider the appropriateness of the sample
used in each research article or report you read in relation to the research question
being answered. You also need to do this whether you are assessing the usefulness of
secondary data, or planning how to collect your own data to answer your own research
question. In this section, we look briefly at reasons for selecting a sample rather than
collecting data from the whole population, before talking about different ways of
selecting samples and their use.
6.2 Selecting samples
Definitions
sample: a sub-group of all group members or the whole population. The sub-group need not necessarily be a subset of people or employees: it can, for example, be a subset of organisations, places or
some of the tracks listed for a music CD.
population: the complete set of group members. The population need not necessarily be people or
employees: it can, for example, be organisations, places or the complete track listing for a music CD.

6.2 Selecting samples 139
Reasons for selecting samples
Researchers usually collect data from a sample rather than the whole population simply
because it is not practicable to collect data from the whole population. This may be
because you do not know what the whole population is, because it is difficult to make
contact with the whole population or because time or financial constraints prevent you
from collecting data from the whole population. However, you should not assume that
collecting data from the whole population is always better than just collecting data
from a sample.
You have a fixed amount of time to complete your research project because of
the submission deadline and your need to complete other work for your course.
Within the time you give to your research project, you may have to collect your
own data. This means that, like other researchers, if you spend a long time collecting data from a large population, you will have less time to work on the other parts
of your project report. Collecting data from a sample of this population will give
you more time for the other parts of your project. Some of this time can even be
used to test that the methods you use to collect the data will work, providing you
collect precisely the data you need to answer your research question and meet your
objectives.
Ways of selecting samples
The way a sample is selected depends, at least in part, on the research question being
answered and whether or not you know what the total population is and can get a full
list of all its members. This complete list of the population’s members is called the
sampling frame. Let’s say an organisation you work for has asked you to use data collected from a sample of all their employees to answer questions that relate to their
employees. Your organisation is able to give you a list of all these employees, using their
payroll system. Not surprisingly, this list is up to date and complete as employees are
keen to ensure they are paid. Using data you collect from the sample selected from this
list, you will, providing your sample is selected at random, be able to statistically estimate answers to questions for all employees. This is called using the sample to make
statistical inferences about the population. In contrast, it is unlikely that you will get a
full list of a supermarket’s customers. While the supermarket will record names and
contact details of those who have a loyalty card, this includes only some of their customers. In addition, even if you work for the supermarket, it is unlikely that you, as a
student, will be given access to the list of loyalty card customers for both commercial
and data protection reasons. Consequently, you will, as we discuss later, have to select
the sample in a different way.
Definition
sampling frame: complete list of all members of the population. You select the sample from this list
when using probability sampling.

140 Chapter 6 Collecting data
For populations where you are able to obtain a complete list, such as an organisation’s employees, you can select your sample using one or more sampling techniques
that use ‘
probability sampling’ (Table 6.1). Because you have a complete list, you can
select employees at random from your list. You can therefore state the statistical
chance or probability that each of these employees has of being selected for your
sample from this list. If your complete list contains 200 employees and you select a
sample of 100 employees using a probability sampling technique, then, as the sample is selected at random, each employee has a 50% chance of being selected. If you
select 50 employees from your complete list of 200 employees using a probability
sampling technique, then each employee has a 25% chance of being selected.
Providing you have selected your sample using a probability sampling technique,
your sample will represent your population statistically, and so you will be able to
make statistical inferences about the population. The level of certainty – also known
as confidence – with which you can say your sample represents your population and
the accuracy of any estimates you make are dependent upon the size of your sample
and of your population. Researchers normally work to a 95% level of certainty with a
margin of error of plus or minus 5%, or of plus or minus 2%. For a population of 200
employees, you would need to collect data from a sample of 132 employees to be 95%
certain that your sample represented that population statistically with a margin of
error of 5% (Table 6.2). As you can see in this table, the larger the sample, the more
certain you can be that it represents the population precisely. You therefore need to
collect data from as large a sample as possible. Like all researchers, you will need to
select a probability sample if you want to statistically estimate the characteristics of
the population. When you read section 7.3, you will discover that statistical tests also
require a sample of at least a certain size to minimise spurious results. In general, this
means your data needs to be collected from a probability sample of at least thirty
respondents.
Definition
probability sampling: variety of sampling techniques for selecting a sample at random from a complete list of the population. Because you have a complete list and select at random, you know the
chance or probability of each member of your population being selected.
Table 6.1 Probability and non-probability sampling techniques
Probability Non-probability
Simple random sampling Quota sampling
Systematic random sampling Purposive sampling
Stratified random sampling Volunteer sampling
Convenience sampling

6.2 Selecting samples 141
Often the data we and other researchers require cannot be collected using a probability sample. Let’s look at your supermarket customers again. As you cannot get a complete list of customers, you do not know how many there are and so will not be able to
work out the chance of each customer being selected for your sample. You will therefore
have to use one or more of a different group of sampling techniques that do not require
a complete list of the population. These are called ‘
non-probability sampling’ techniques
(Table 6.1) and, if you select your sample using these, you cannot say it represents your
population statistically. Consequently, you will not be able to make statistical inferences from your data. Rather, you will be normally be using your judgement to select a
sample that best enables your research question to be answered. Non-probability samples are often used in conjunction with qualitative data collection techniques such as
semi-structured interviews and unstructured observation.
As you read section 6.4, you will discover that, other than for quota samples (normally used with quantitative data collection techniques such as questionnaires), these
techniques usually require smaller samples. The actual sample size when collecting qualitative data will also depend in part on your research question. While for some research
questions such as those asking whether something exists or requiring a single illustrative
exemplar, a sample of one may be sufficient. However, for most research projects, a larger
sample will be needed. Following an analysis of the number of interviews used in
research published in top organisation and workplace journals, Mark and a colleague
noted that while between 15 and 60 interviews are likely to be considered sufficient, the
actual number of interviews is dependent on the research purpose and the nature of
data collected (Saunders and Townsend, 2016). However, this range refers to research
that has been published in top journals, not student projects, and it is unlikely you will
have time to undertake so many interviews. Building on advice of others, we therefore
Table 6.2 Probability sample sizes required for a 95% confidence level with
a 5% margin of error
Population size Sample size Population size Sample size
50 44 500 217
100 79 1,000 278
150 108 5,000 357
200 132 10,000 370
250 151 100,000 383
Source: Saunders et al. (2016).
Definition
non-probability sampling: a variety of sampling techniques for selecting a sample when you do not
have a complete list of the population. Because you do not have a complete list of the population,
you cannot select your sample from this population at random. This also means you do not know
the chance or probability of each member of your population being selected.

142 Chapter 6 Collecting data
suggest for questions requiring a sample selected from a population that is homogeneous (similar), such as male UK-born business studies students aged 21 at your university, a non-probability sample size of between 4 and 12 participants is likely to be
sufficient. In contrast, where questions need a sample selected from a heterogeneous
(more varied) populations, such as male and female students of all ages studying any
programme at your university, the sample size is likely to be larger, say between 12 and 30
(Saunders 2011).
Probability sampling techniques
Having given you a general overview of the difference between probability and nonprobability sampling techniques, let’s look first at the probability sampling techniques
listed in Table 6.1 in more detail.
Simple random sampling
If you have ever bought a lottery ticket, you will have experienced the results of simple
random sampling
. Simple random sampling uses a series of random numbers, such as
those generated by the =RAND function in Microsoft Excel, to select a sample of members from a population. Like the lottery balls, each member of the population in your
list (sampling frame) is given a unique number. Random numbers are then used to
select the numbers of those members of your population from the sampling frame who
will be in your sample.
Five sets of numbers drawn at random recently for the United Kingdom’s National
Lottery Lotto game (selected from the numbers 1 through 59) are:
14 16 23 24 25 57
4 15 19 44 45 49
20 21 22 24 27 50
4 10 18 41 42 47
11 30 38 45 46 49
Although each draw of six numbers has been selected at random, you can see that these
numbers are not spread evenly between 1 and 59. Rather, some numbers (such as 23, 24
and 25 in the first draw) are close together and then there are large gaps (for example, 25
and 57 in the first draw). This is a property of random numbers that often occurs when
you use them to select a sample of less than a few hundred members. It means that
researchers, including you, should not use simple random sampling to select a sample
of less than a few hundred if your sampling frame has been sorted in some way. For
example, if your sampling frame is sorted in order of employees’ seniority, although
Definition
simple random sampling: type of probability sampling in which each member of the population has
an equal chance of being selected at random and included in the sample. Each member is usually
selected using random numbers.

6.2 Selecting samples 143
your sample of employees will have been selected at random, the occurrence of random
patterns such as those illustrated by the five lottery draws may mean that certain levels
of seniority in your population are underrepresented and others are overrepresented.
Systematic sampling
For a systematic sample, members of the sample are selected from the sampling frame at
regular intervals. To ensure your sample is entirely random, your first sample member is
selected using a random number before subsequent members of your sample are selected
systematically. If we return to the example of an organisation’s employees and the sampling frame that you got from payroll, we can see how this sampling technique works.
The list you have been given by payroll is in descending order of salary, the organisation’s chief executive being listed first, followed by the most senior managers. The last
employee on your list is a newly appointed school leaver who earns the lowest salary. To
select a sample of one-fifth of employees (20%), you need to perform the following steps:
1 Select your first employee from the sampling frame at random using a random number, between one and five for a 20% sample, in this example employee number two.
2 As your sample is a fifth (20%) of employees, you then select the 7th, 12th, 17th,
22nd and 27th employee in your sampling frame, continuing to select every fifth
employee until you reach the end of the list.
By selecting the first sample employee using a random number, the rest of your sample
will also have been selected at random. This technique works well, providing there are
no regular patterns in the sampling frame. Let’s say your sampling frame consists of UK
Premier League football club first-team players who started the match for their team
last Saturday. These players are listed in your sampling frame by team starting with the
captain of each team, and ending with the goalkeeper. If you were to select a sample of
1 in 11 players (approximately 9%) and the first player you selected at random was the
11th player, then you would subsequently select the 22nd, 33rd, 44th and 55th player
until you reached the end of your list. However, your sample would consist entirely of
goalkeepers! Providing there are no regular patterns in your sampling frame, this technique usually works well and also has the advantage of being relatively easy to explain.
Stratified random sampling
Not surprisingly, given its name, stratified random sampling is a modified version of random sampling. When using this sampling technique, the sampling frame is divided
Definitions
systematic sampling: type of probability sampling in which the first sample member is selected from
the sampling frame at random, using a random number. Remaining sample members are selected
subsequently at regular intervals from the sampling frame.
stratified random sampling: type of probability sampling, in which the sampling frame is first divided
into relevant strata. Sample members are then selected at random from within each stratum, using
either simple random or systematic random sampling.

144 Chapter 6 Collecting data
into strata or layers that are relevant to your research question, your sample being
selected separately from each stratum or layer using either simple random sampling or
systematic sampling. Let’s say you are selecting a sample of web-based customers for a
small web-based retailer, using their complete list of customers as your sampling frame.
This alphabetical list includes each customer’s name and their postal region (United
Kingdom, mainland Europe, North America, rest of the world). You therefore perform
the following steps:
1 Split your sampling frame into relevant separate strata or groups (using the four separate postal regions as your strata).
2 Select a random sample from within each stratum or group using either:
(a) simple random; or
(b) systematic sampling.
By doing this, each postal region or stratum will be represented in the same proportion
in your sample as they are in your complete population.
Non-probability sampling techniques
As you read at the start of this section, researchers, including you, are often unable to
obtain a list of their population. This means that probability sampling cannot be used
as there is no sampling frame. Fortunately, there are a variety of non-probability sampling techniques that can be used. We will now look at those we listed in Table 6.1.
Quota sampling
You have probably been selected for a quota sample at some time in your life. Sometimes
when you are shopping, a person with a clipboard asks you if you would be willing to
answer a few questions. If you answer ‘Yes’, the next question you are asked is something
like, ‘Which one of the following age groups are you in?’ after which a number of different age groups are read out to you. This is often followed by a question about your occupation or, perhaps, the newspaper you read (which may be used to indicate your social
class). The person asking the questions (interviewer) has been given a number or quota
of people to interview, such as ‘3 working class males aged 25–55’ and ‘2 female students aged 24 or under’, and so on. If the interviewer has already filled the quota for the
group your answers place you in, you will not be asked any further questions. However,
if the interviewer continues to ask you questions from the questionnaire, you will
become part of the quota.
Quota samples are used as a substitute for a probability sample to select participants
when a sampling frame is not available. This means the sample size is similar to those
Definition
quota sampling: type of non-probability sampling that ensures the sample selected represents certain characteristics in the population that the researcher has chosen.
6.2 Selecting samples 145
used for probability samples (Table 6.1), and the data are likely to be analysed using statistical techniques (section 7.3). However, while data used to be collected from the sample by a team of interviewers, each of whom has a quota to fill, this is becoming less
common. Rather, as you will read in Research in practice 6.1, the quota samples used by
large commercial polling organisations are increasingly selected from an online panel
of volunteers, who then agree to take part in answering questions online.
Quota sampling
As a major polling organisation, YouGov conducts online polls about politics, brands and
consumer trends. The sample for each poll it conducts is selected from a diverse panel of
over five million volunteers from across 38 countries, including over 800,000 in the
United Kingdom. YouGov has selected their panel of registered volunteers to represent
all groups in society, although not in the same proportions as in the population as a
whole. For each poll, an active sample is selected, made up of a number of demographic
groups. These groups are specified in much the same way as a quota sample group
would be, using data such as volunteers’ age, gender, social class and region where they
live. This method of sampling is closer to quota sampling than random sampling, email
invitations to take part being sent to those who are selected for the active sample. Those
who actually take part form part of the responses for a particular demographic group,
answering the questions online.
Source: Developed from YouGov (2016).
Research in practice 6.1
Purposive sampling
Purposive sampling is the most frequently used form of non-probability sampling. It is
used particularly to select a small sample when collecting qualitative data. When a
researcher selects a purposive sample, the researcher is using their judgement to actively
choose those who will best be able to help answer the research question and meet the
objectives. Some of the population will have a chance of being chosen by the researcher
while others will not.
Let’s say you have read a journal article where the researcher has selected two major
banks to explore the performance of individual banks after the global financial crisis of
2007–8. These two banks have been selected purposively because they represent the
extremes of what happened to banks in one country. The bank at one extreme was
nationalised by the government; the bank at the other extreme received no support
from the government. The author of the article argues that, when contrasting the
Definition
purposive sampling: type of non-probability sampling in which the researcher’s judgement is used
to select the sample members based on a range of possible reasons and premises.

146 Chapter 6 Collecting data
findings from these extreme cases, differences will be more obvious and that her findings are relevant in understanding or explaining the performance of other banks. On
reading the article, you feel this is an appropriate reason for selecting the sample, which
allows her research question to be answered fully. You also feel the generalisations she
has made about other banks’ performance on which the impact has been less extreme
are logical.
You use purposive sampling when you need to understand what is happening so you
can make logical generalisations. This means that you, like other researchers who select
a purposive sample, need to explain clearly the criteria you have used to select your
sample, the reasons for this and the underlying premise on which these are based. Your
selection can involve a variety of different purposive sampling techniques, which we
have summarised in Table 6.3, along with reasons why you might use them and the
underlying premises on which your reasons are based.
Table 6.3 Varieties of purposive sampling
Purposive
sampling variety Reasons for use Underlying premise
Typical case Sample will be illustrative and considered
representative, albeit not statistically.
Sample is typical of the
population.
Critical case Sample will either make a point
dramatically or be crucial to addressing
the research aim and objectives.
Topic of interest is most likely to
occur in the sample selected, or
sample selected is essential to the
operation of the process.
Extreme case Sample consisting of unusual or special
participants will enable you to find out
the most.
Findings from extreme cases will
be relevant in understanding or
explaining more typical cases.
Heterogeneous Sample will have sufficiently diverse
characteristics to provide the maximum
variation possible in the data collected.
Any patterns that emerge are
likely to be of particular interest
and value, representing key
themes.
Homogeneous Sample consisting of one particular
sub-group will provide minimum variation
in possible data collected.
Will allow characteristics to be
explored in greater depth and
minor differences to be more
apparent.
Volunteer sampling
Volunteer sampling is when potential sample members are either volunteered by someone else or volunteer themselves to take part in the research. It is used particularly when
potential sample members are either difficult to identify of difficult to reach.
Definition
volunteer sampling: type of non-probability sampling in which the potential sample member either
volunteers or is volunteered to be a sample member.

6.2 Selecting samples 147
Snowball sampling is used when it is difficult to identify members of your population. Let’s say that you wish to research people who are working in the informal economy while claiming welfare and unemployment benefits. As these people are difficult
to reach, snowball sampling may provide the only possibility for finding potential
participants. Alternatively, your population may be like chief executives of Fortune
500 companies, easy to identify but difficult to access. Once again, by using snowball
sampling, you may be able to reach such people. To obtain your snowball sample of
chief executives of Fortune 500 companies, you would need to do the following:
1 Make contact and collect data from an initial chief executive of a Fortune 500
company.
2 Use the initial chief executive to identify and volunteer a number of other chief executives of Fortune 500 companies and to support your gaining access.
3 Use these chief executives to identify and volunteer further chief executives of
Fortune 500 companies and to support your gaining access, so the sample increases
in size like a snowball that is rolled.
Remember, those who are volunteered for a snowball sample are most likely to identify and volunteer others who are similar to themselves, resulting in a homogeneous
sample.
In contrast, if you ask your sample members to identify and volunteer themselves,
say, by putting an advertisement in appropriate media inviting them to take part in
your research, you are using
self-selection sampling. Increasingly, researchers are inviting potential participants using a variety of electronic media such as intranets, blogs
and bulletin boards alongside invitations through general letters or all user emails
(Saunders, 2011). However, when using self-selection sampling, you need to be careful.
Those who self-select and volunteer to be members of your sample often do so because
they have strong feelings or opinions about your research topic and consider it sufficiently important or interesting to give it some of their time. This means they are likely
to be different from people who do not offer to be involved in some way and so are not
representative of the population.
Definitions
snowball sampling: type of non-probability sampling in which, after the first sample member,
subsequent members are identified and volunteered by earlier sample members.
self-selection sampling: type of non-probability sampling in which possible sample members are
asked to identify themselves and volunteer to take part in the research.
convenience sampling: type of non-probability sampling in which the sample the researcher uses is
those who are easy to obtain rather than because of their appropriateness.
Convenience sampling
The last variety of non-probability sampling which we talk about, convenience
sampling
, is rarely used by researchers and is one, we hope, you will not use for your own
148 Chapter 6 Collecting data
research! It simply involves ‘using’ (‘selecting’ is too positive a word!) those who are
easiest to get hold of for your sample. Say you have to find out managers’ views about
something. We know you won’t do this, but let’s pretend that you decide to collect your
data from part-time students in your class. These people are only in your sample because
of the ease of getting hold of them and have little, if any, relevance to your research.
Like most convenience samples, they are likely to be of limited use in answering your
research question and meeting your objectives.
Questionnaires are used widely to collect data, and we have no doubt that you have
completed questionnaires and used data that were collected by questionnaires. Your
lecturers will have asked you to complete module evaluation questionnaires so they can
find out your, and your classmates’, opinions about their modules. You will also have
used the results of questionnaires in your assignments, discussing journal articles in
which research findings were based on data collected by questionnaires.
The term ‘questionnaire’ refers to all methods of data collection in which each
potential
respondent is asked to answer the same set of questions in the same order. It
therefore includes questionnaires that are used to collect data:
by the Internet, each respondent reading the questions and recording their own
answers;
by post, each respondent reading the questions and recording their own answers;
by hand, each respondent reading the questions and recording their own answers;
by telephone, where an interviewer also records each respondent’s answers;
face to face, where an interviewer also records each respondent’s answers.
When a questionnaire is used by an interviewer to collect data, it is often called a
structured interview.
Questionnaires are a good method for collecting data about the same things from
large numbers of respondents. Questionnaires allow you to ask the same set of standardised questions to a large number of respondents. Because the questions are standardised, the data collected by questionnaires are often used either for descriptive research,
such as students’ opinions of a module they have just taken, or for explanatory research
6.3 Collecting data using questionnaires
Definitions
questionnaire: general term that includes all methods of data collection in which each person is
asked to answer the same set of questions in the same order. Questionnaires can be distributed face
to face by an interviewer, by telephone, by hand, by post and using the Internet.
respondent: person who answers the questions in a questionnaire.
structured interview: method of data collection using a questionnaire in which each person is asked
the same set of questions in the same order by an interviewer who also records the responses.

6.3 Collecting data using questionnaires 149
to test a theory. These data are usually analysed statistically (section 7.3). Let’s say you
have a theory that the more time students spend working on their projects, the more
they will enjoy the work. In explanatory research, you use the data you collect to examine and explain such relationships statistically: in our example, to see whether or not
students’ enjoyment of their work is statistically related to the amount of time they
spend working on their projects.
In this section, we will look at how to:
1 design individual questions;
2 design the questionnaire;
3 write the covering letter or email;
4 conduct pilot testing;
5 distribute the questionnaire to a sample of potential respondents using each of the
distribution methods listed above.
How to design questions
A questionnaire is useful only if:
it collects data that are needed to answer the research question and meet the
objectives;
it collects data from a large enough number of respondents to answer the research
question and meet the objectives;
the questions asked are understood and interpreted by respondents in the way the
researcher wanted them to be understood and interpreted.
This means you need to be very clear about the data you need to collect and to design
your questions to collect these data, minimising the chances of questions being misinterpreted. In making sure that your questions (and so your questionnaire) will provide
you with enough data to answer your research question and meet all your objectives,
you are ensuring what researchers call ‘
content validity’. By designing your questions
carefully and ensuring that they actually collect data about what you intend them to
measure rather than something different, you are ensuring what researchers call
construct validity’. When you are reading about other researchers’ research that has used
questionnaires, you will need to look at the actual questions that were asked to see if
they are valid. This will allow you to better understand and assess the conclusions these
researchers reached, based on the answers their respondents gave to these questions.
Definitions
content validity: the extent to which a data collection tool, such as a questionnaire, provides enough
data to answer the research question and meet all the objectives.
construct validity: the extent to which the questions asked actually collect data about what they are
intended to measure.

150 Chapter 6 Collecting data
When you design your questionnaire, your first stage will be to design individual
questions to collect the data you need to answer your research question and meet your
objectives (Ekinci, 2015). When you do this, you can:
use questions from other researchers’ questionnaires (providing you reference the
source and obtain permission where necessary);
adapt questions from other researchers’ questionnaires (providing you explain how
you adapted them, reference the source and obtain permission where necessary);
design your own questions.
However, before you decide to use someone else’s questions, beware. There are a
lot of poor questions and badly designed questionnaires available! You also need to be
certain of the wording of each question will be suitable for your intended respondents
and the context in which they will complete the questionnaire. If your questions are
going to be answered only by accountants, you will be able to use far more specialist
language than if they are going to be answered by people in general.
Saunders et al. (2016) highlight seven types of question that can be used in questionnaires. These are listed in Table 6.4, along with a brief description of when to use them
and examples.
Table 6.4 Types of question for use in questionnaires

Type Use when . . . Examples
Open . . . unsure of the
response, require a
detailed answer, or
want to find out
what is uppermost in
respondent’s mind.
If there are any other areas or issues that concern you,
please feel free to comment below*
What do you like most about visiting this theme park?
List . . . you need to be
sure the respondent
has considered all
possible answers.
Which of the following fruit juices have you purchased in
the past week?
[please tick
all appropriate boxes]
Apple

Orange
Cranberry
Pineapple
Tomato
Other (please say) ………………………
Category . . . you need to
ensure the
respondent’s answer
will only fit into one
category.
Are you in receipt of a state pension?
(
Interviewer, listen to the respondent’s answer and tick the
correct box
).
Yes (receives state pension)
1
No (does not receive state pension) 2

6.3 Collecting data using questionnaires 151

Type Use when . . . Examples
Ranking . . . you want the
respondent to place a
list in rank order.
Number each of the holiday destinations listed below in
order of their attractiveness to you for your next holiday.
Number the most attractive holiday destination 1, the next 2
and so on. If a destination is not attractive, leave blank.
Rating . . . you want a
respondent’s
opinion or belief.
For the following statement please tick the box that
matches your opinion most closely.
How likely do you believe it is that you will pay off your
overdraft within a year of graduation?
∙ ∙ ∙ ∙ ∙
Quantity . . . you want the
respondent to tell
you a number or
amount.
How many dependent children do you have living with you?
[ ] [ ]
(
For example, for 2 write: [ ] [ 2 ] )
Matrix . . . the responses to
two or more
questions are selected
from the same set of
possible answers.
The following statements ask about your feelings regarding
the future of the Happy Toy Manufacturing Company.
of my career
at the company

Destinationn Attractiveness
England [ ]
Scotland [ ]
Wales [ ]
Ireland [ ]
France [ ]
Germany [ ]
Spain [ ]
Holland [ ]
Agree Tend to Tend to Disagree
agree disagree
I generally
believe what
∙ ∙ ∙ ∙
my manager
tells me
Very Quite Not Quite Very
likely likely sure unlikely unlikely
Strongly Agree Disagree Strongly
agree disagree
I feel the future
for the company
∙ ∙ ∙ ∙
is getting brighter
I would be
happy to
∙ ∙ ∙ ∙
spend the rest
*Some researchers would term this a ‘request’ rather than a ‘question’.
Table 6.4 Continued
152 Chapter 6 Collecting data
The example questions in Table 6.4 show how the answer you get depends on the
way your question is worded. Looking at the first open question, this could result in
some respondents writing hundreds of words commenting about the issues that concern them, and others writing only a few words or even nothing! In contrast, the second open question asks the respondent to only answer what they like most. For all your
questions, you will need to decide the data you need to answer your research question
and meet your objectives and, therefore, in how much detail you want your respondents to answer. However, beware of including too many open questions; they are
extremely time consuming to analyse.
Next, compare the list question and the ranking question in Table 6.4. You can see
that, although the list question asks about fruit juices and the ranking question asks
about holiday destinations, only the list question allows the respondent to include
another response of their choice. The list question gives the respondent the option
‘other (please say) . . .’ while the ranking question names eight holiday destinations,
giving no space for the respondent to add her or his own destination. Including the
‘other (please say)’ option alters the data collected, something you need to beware of
both when reading about others’ research and designing your own questions. This is
because it allows respondents to have a free choice in how they answer rather than be
constrained by the options you have given them.
Now look at the two examples of rating questions. The first of these questions has an
even number of possible answers (four) while the second has an odd number of possible
answers (five). By giving an even number of possible answers, you are forcing your
respondents to make a choice between, in the first rating question, agreeing and disagreeing. You are not allowing them to neither agree nor disagree. In contrast, in the
second example, the respondent is allowed to be ‘not sure’ being neither ‘quite likely’
nor ‘quite unlikely’. Once again, this emphasises that you need to be clear about the
precise questions asked, both when reading reports of others’ research and when
designing your own questionnaire.
Finally, look at the category question. A number in a smaller font size has been added
next to each possible response: ‘1’ next to ‘yes (receives state pension)’ and ‘2’ next to ‘no
(does not receive state pension)’. These numbers are the codes that will be used to represent each respondent’s answer on a paper questionnaire when the data are entered on a
spreadsheet, such as Excel or statistical software such as IBM SPSS Statistics, for statistical
analysis. Wherever possible, we suggest you add code numbers to possible question
responses on paper questionnaires at the design stage. This will save you having to code
the answers later! However, if you use an online survey tool, such as SurveyMonkey
or
Qualtrics
, to design your Internet questionnaire and collect your data, codes are added
automatically. Once you have collected your data via the Internet, it will only take a few
clicks of the mouse to download and save it as a spreadsheet file. This is far quicker than
entering the data yourself; it also means your data set is less likely to contain errors!
How to design the questionnaire
Your questionnaire should be pleasing to look at and, of equal importance, easy for
respondents to read and fill in their answers. This means you need to ensure your

6.3 Collecting data using questionnaires 153
questionnaire is easy to read and for Internet questionnaires optimised for viewing on
laptops, tablets and mobile phones. You also need to tell the respondents what the
questionnaire is about and why you want them to answer the questions. The order of
questions needs to be logical to those answering rather than to you. Your questionnaire
should be as short as possible, although you obviously need to ask enough questions to
collect all the data you need to answer your own research question and meet your objectives! These design points, along with others, are listed in Table 6.5.
Table 6.5 Points to remember when designing a questionnaire
Layout Question order
•    Title should be clear and a larger font size.
•    Introduction should be brief and explain why
the topic is important and what the
respondent should do.
•    Questions should be displayed clearly on the
page or screen.
•    The typeface (font) should be easy
to read.
•    Questions’ typeface (font), font size, spacing
and formatting should be consistent
throughout.
•    When printed, it should be on good-quality
paper/card.
•    Details of how to return the completed
questionnaire should be given at the end.
•    At the start, ask more straightforward
questions.
•    At the start, ask questions that are related
clearly to the stated topic of the
questionnaire.
•    Ask more complex questions in the middle
of the questionnaire.
•    Group questions into sections that will be
obvious to the respondent.
•    Use filter questions to stop the respondent
answering questions that are not relevant.
•    Make sure the wording of questions is
consistent throughout the questionnaire.
•    At the end of the questionnaire, ask
personal or sensitive questions.
If you look at Research in practice 6.2, you will see that sometimes you need to ask
only a few questions. This short questionnaire is designed only to collect a limited
amount of data about diners’ experiences at the restaurant. Yet, despite it being short,
the design includes many of the points made in Table 6.5. Starting with the layout, the
questionnaire has a clear title (‘How did we do?’) in a larger font, which also describes
the topic. The brief introduction is friendly in style and explains why (‘to know what
you think’), and what the respondent should do (‘tell us what you think’) in a little
more detail. The questions do not appear squashed and are printed using the same
typeface throughout. Despite this, you probably think that there is not enough space
for an answer to the open question at the end. We agree with you! We are told that the
questionnaire is printed on card, hopefully good quality, and that the return postal
address is printed on the other side. It would also have been helpful if respondents were
told there was a box for completed questionnaires to be put in, perhaps by the exit.
If you now look at the actual questions, you will see that these are both straightforward to answer and, other than the ‘Date of visit’, relate clearly to the stated topic. It
may be that the waiter or waitress fills this in before the questionnaire is given to the
customer. The first three questions are grouped as a matrix into one section and are
phrased consistently: ‘what was the . . . like?’ However, the last open question has an

154 Chapter 6 Collecting data
error: it should say ‘us’ rather than ‘use’. Unlike the designer of this questionnaire, you
need to proofread your questionnaire very carefully, remembering that a spell check
will not pick up mistypes that result in the wrong word being included.
Research in practice 6.2 does not include any sensitive questions or filter questions.
Whether or not a question is sensitive will depend on what is being asked and the person who is answering it. It will therefore be a matter of judgement. Fortunately, as you
Questionnaire design
Often when you visit a restaurant, you are given a ‘comment card’ with your bill similar
to the one below. This is usually printed on card rather than paper. After a short introduction asking you to respond, there are a few questions about your experiences in the
restaurant and some space for you to add your own comments and feedback. You are
then expected to either place the card in a box as you leave or, alternatively, post the
card to the restaurant. The postal address is printed on the other side of the card, and
usually the postage has been prepaid to encourage you to send it back.
Research in practice 6.2
How did we do?
At Mark ‘n Phil’s restaurant we’re always keen to know what you think . . . so if
you’ve got any comments about your visit we’d like you to tell us.
Date of visit? dd…………. mmm………. yy……
Excellent Good Average Poor Awful

What was the quality of your food like?
What was the service you received like?
What was your overall experience like?










If you have anything else you would like to tell us, please do so below . . .
will see later, this can be checked as part of pilot testing. Filter questions are used to
route your respondents through your questionnaire so they miss out questions that are
not relevant. Let’s say you have a questionnaire that includes some questions that are
only relevant to respondents who are members of a professional body. Your filter question is:
22. Are you a member of a professional body? Yes
(If no, go to question 30) No
6.3 Collecting data using questionnaires 155
The following (open) question is then only answered if the respondent has answered
‘yes’. It asks:
23. Please name your professional body: ………………………………………………….
Questions 24 to 29 are also about the respondent’s membership of a professional
body.
If you are using an online survey tool, answering the filter question will, as far as the
respondent is concerned, just take them to the next relevant question; in our example,
to question 23 if they answered ‘yes’ and to question 30 if they answered ‘no’. However,
if you are using a paper questionnaire in which the respondent fills in answers, the
respondent will have to find the next relevant question. Unfortunately, some respondents will not follow your filter question’s instructions and will try to answer the following questions whether they are relevant or not! You therefore need to use filter questions
sparingly in paper questionnaires.
What to write in the covering letter or email
In Research in practice 6.2, the purpose of the questionnaire was only explained briefly
to potential respondents. When you distribute a questionnaire by post or the web, you
will need a more detailed covering letter or email to explain the purpose of your questionnaire. The structure and contents of your covering letter are summarised in Table
6.6. You will often be expected to put a copy of your covering letter and a blank questionnaire as appendices in your project report.
Table 6.6 Structure of a covering letter or email
Section Contents
Letter/email head Official, with a logo, address, telephone number and email
address. On good-quality white paper if for a paper questionnaire.
Name and address of
potential respondent
Use name and full postal address for personal approach if for
paper, otherwise just name.
Date In full, e.g. 8th July 2017.
Greeting Use title and name if possible, e.g. Dear Mrs Penny.
Heading or email subject Brief descriptive title of questionnaire
1st paragraph What research is about, why it is useful, how respondent’s time is
needed and their answers are of value.
2nd paragraph Whether responses will be confidential and/or anonymous.
3rd paragraph How results will be used. Whom to contact if there are any
questions, with telephone number and/or email contact details.
Explanation of how to return the questionnaire, to whom and by
when.
Closing remarks Thank recipient in advance for help.
Signature and name Sign by hand if letter. Also put sender’s title, forename and
surname.

156 Chapter 6 Collecting data
Why pilot testing is important
Before distributing your questionnaire to potential respondents, you need to check that
it will work; that your respondents will have no problems in answering the questions
and that their responses will be recorded correctly. This is important for both Internet
and paper questionnaires. Even if you are really short of time, you still need to try out
your questionnaire and covering letter with a small number of people who are like
those who will answer it in your research. This is known as
pilot testing and will also
help to confirm that your actual respondents will understand the meaning of your
questions and will be able to follow the instructions on the questionnaire. Remember,
it is far easier to correct mistakes in your questionnaire at the pilot stage than to have to
recollect your data because mistakes had been overlooked.
Definition
pilot test: trying out of a questionnaire, interview schedule or other method of data collection with
a small group of respondents who are similar to those who will be used in the actual research to see
if it works. Any problems that arise in the pilot test can then be sorted out before the actual research
is undertaken.
How to distribute the questionnaire
At the start of this section (6.3), we listed five ways in which you might distribute your
questionnaire to potential respondents and so collect data from your sample. Whichever
way you are using, you are trying to get as many completed questionnaires returned to
you as possible. Although response rates from questionnaires vary considerably,
research by Baruch and Holtom (2008) based on looking at 490 academic studies using
questionnaires provides a clear indication of the rate you might expect. The average
response for studies where data were collected from individuals was 52.7%, while that
for questionnaires to organisations was lower, being an average of 35.7%. Studies using
Internet or telephone distribution reported higher response rates than traditional
paper-based questionnaire distribution.
Distributing Internet questionnaires
Internet questionnaires are normally distributed using an email with a hyperlink to the
actual questionnaire or using a link to a website that has a hyperlink to the questionnaire. For the first, this means you need to have a complete and up-to-date list of your
sample’s email addresses. For the second, you will need to invite people to contribute
and direct them to your website and your questionnaire. If you choose to do this, you
must follow the general guidelines for using the Internet. This is called netiquette and
means you should do the following:
Only send emails and posting to relevant user groups.
Remember that postings to more than 20 user groups at once are unacceptable to
many web users.

6.3 Collecting data using questionnaires 157
Avoid sending junk emails or spam.
Avoid emailing multiple mail lists as this will mean some people will get more than
one copy.
Avoid using email attachments as these may be thought to contain viruses.
To distribute an Internet questionnaire, you need to observe the following:
1 Ensure the Internet questionnaire is set up in the online survey tool and works as you
wish, the data set is generated automatically and that you have noted the direct
hyperlink to include in your accompanying email or on your website.
2 If you are emailing potential respondents, let them know in advance that you will be
emailing a hyperlink to a questionnaire or to a website that has a hyperlink to the
questionnaire.
3 Email potential respondents with your covering email and the hyperlink to the questionnaire or advertise your website widely, highlighting the closing date.
4 If you are emailing potential respondents with the hyperlink, email them again a
week after the questionnaire has been distributed, thanking those who have returned
the questionnaire and encouraging others to respond.
Distributing postal, or delivery and collection questionnaires
The distribution of postal questionnaires and delivery and collection questionnaires is
very similar. As you will have worked out from their names, the only real difference is
that the former is delivered and returned by post, while the latter is delivered and collected in person. To distribute a questionnaire using either of these ways, you need to
observe the following steps:
1 Ensure that the questionnaires and accompanying letters are printed and the envelopes addressed.
2 Contact potential respondents to let them know in advance that you will be posting/
delivering by hand a questionnaire to them.
3 Post or deliver by hand the questionnaire.
4 Contact potential respondents a week after the questionnaire is posted/delivered by
hand, collecting the questionnaire (delivery and collection) or thanking those who
have returned the questionnaire (postal) and encouraging others to respond.
Conducting telephone or face-to-face structured interviews
The quality of your structured interviews, whether conducted using a telephone or face
to face will depend on your interviewing ability. For telephone-structured interviews,
the clarity and tone of voice with which you read the questionnaire questions will be
crucial, while for face-to-face structured interviews, your appearance will also be important. To conduct a telephone or face-to-face structured interview using a questionnaire,
you need to pay attention to the following points:
1 Ensure that the questionnaires are printed or set up in an online survey tool so that
you can record responses easily.

158 Chapter 6 Collecting data
2 Where possible and resources allow, contact potential respondents to let them know
in advance that you will be telephoning/visiting them within the next week to conduct a structured interview.
3 Try to contact each potential respondent in person, noting down the date and time
of contact and whether or not the structured interview took place. If you arrange an
alternative appointment, note it down and visit then!
4 Try to contact each potential respondent at least twice more, each at a different time
and on a different day, noting down the same information.
Every day of your life you read about, listen to and watch interviews. Journalists interview politicians and business leaders, chat show hosts interview celebrities of all kinds
and business investors known as ‘dragons’ interview entrepreneurs with new business
ideas. All these interviews have one thing in common: they are purposeful discussions
between two or more people. Although you might argue that the purpose of some of
these interviews is to entertain rather than gather data for research, they still give you a
real insight into the process of asking questions and the preparation and skills you need
to interview people.
In the previous section (6.3), we talked about the structured interview: when each
respondent is asked the same standard questions from a questionnaire by an interviewer, either by telephone or face to face. In this section, we will look at two other
forms of interview: semi-structured and unstructured interviews. These are sometimes
also referred to as qualitative research interviews, the person answering your semistructured or unstructured interview questions being called the
participant. Like other
researchers, you will find these types of interview particularly useful where you:
wish to gather data about a particular topic;
generate data to enable theory development;
encourage participants to tell stories from their own perspectives;
gain insights into indivduals’ experiences or lifeworlds (Cassell, 2015).
In a
semi-structured interview, you (or the researcher) will have a list of topics to be
covered and questions to be asked, although the order in which you ask them will vary
Collecting data using semi-structured or unstructured
6.4 interviews
Definitions
participant: the person who answers the questions in a semi-structured or unstructured interview.
semi-structured interview: a method of data collection in which the interviewer asks about a set of
themes using some predetermined questions but varies the order in which the themes are covered
and questions asked. The interviewer may choose to omit some topics and questions and ask additional questions as appropriate.

6.4 Collecting data using semi-structured or unstructured interviews 159
from interview to interview depending on the responses from the participant. For some
semi-structured interviews, you may decide not to ask some of your questions or not to
cover one or two topics if they are not relevant to that participant. You may also decide
to ask additional questions to find out further details and explore your objectives in
more depth or, alternatively, to check that your understanding of what the participant
is telling you is correct.
Unstructured interviews are more informal and are used to
explore a general topic in which you (or the researcher) are interested in more depth.
Unlike semi-structured interviews, you do not have a list of questions to ask, although
you still need to be clear about the topics you wish to talk about. In unstructured interviews, you want your participants to talk openly and widely about the topic with as little direction from you as interviewer as possible. For this reason, they are sometimes
called non-directive interviews.
As you have read academic journal articles, you will have found out about research
that has used semi- and unstructured interviews. In many of these articles, such interviews will have been used to collect data that were analysed qualitatively (section 7.4).
These data will have been used to either explore what is happening and gain new
insights or describe what is happening and identify general patterns. Semi- and unstructured interviews may also have been used along with other data collection techniques.
For topics where little is known, the researcher may start by using unstructured interviews to explore what is happening. Alternatively, researchers start by collecting data
using a questionnaire, analyse these data statistically and then use semi-structured
interviews subsequently to understand the statistical relationships their analysis has
revealed.
Some years ago, we along with another colleague used a questionnaire to collect data
from a manufacturing company about employees’ use of and feelings about different
communication methods. Our statistical analysis showed that office employees looked
at the company’s noticeboards to find out what was happening, while production
employees rarely looked at these noticeboards. The reasons for this difference were
unclear from our questionnaire data. In our subsequent semi-structured interviews
with both office and production employees, we probed for reasons why this happened.
The office employees did not know why. However, the production employees we interviewed all provided a compelling reason. The noticeboards were in the main entrance
hall where office employees entered each day. In contrast, the production employees
used a different entrance where there were no noticeboards. Not surprisingly, we recommended the company put up noticeboards at the entrance used by the production
employees.
If you are going to use semi- or unstructured interviews, you need to prepare for the
actual interview. As part of your preparation, you will need to be clear about the topics
Definition
unstructured interview: a method of data collection in which the participant talks openly and widely
about the topic with as little direction from the interviewer as possible. Although there is no predetermined list of questions, the interviewer will have a clear idea of the topics to explore.

160 Chapter 6 Collecting data
you are going to cover to collect the data you need to answer your research question and
meet your objectives, as well as the location of the interview and your appearance.
Next, you will need to think carefully about how you might ask questions about these
topics. Following a pilot interview, you will have to conduct the interviews, making
sure that you test your understanding when necessary and show that you are listening
at the same time as taking notes or audio-recording the conversations. We will now
look at each of these.
How to prepare for semi- and unstructured interviews
Imagine you have a meeting with an important client of the public relations company
for whom you work. Before going to this meeting, you would prepare carefully. You
would make sure you knew about the client you were meeting and their organisation so
that you were credible. You would also think carefully about the likely conversation you
would have, the questions you would ask and the likely questions you would be asked
and how you would answer them. Preparing for a semi-structured or unstructured
interview involves thinking about similar things. In particular, you need to pay attention to the following:
Make sure you have found out as much as possible about the person you are going to
interview and, where appropriate, the organisation where they work.
Develop an interview guide, listing the topics you want to discuss and initial questions you will ask for each topic.
Choose a location that is convenient for your participant, where that person will feel
comfortable and you will not be disturbed.
Make sure your clothing and appearance is appropriate for the interview.
Work out how you will use your body language to show your interest and that you
are listening attentively.
These are illustrated in Research in practice 6.3. However, as well as these, to prepare for
an interview you will also need to:
develop a consent form;
think how you are going to record the data: if you are going to audio-record the interview, make sure that your audio recorder works, it has sufficient memory to store a
day’s interviews and you have spare batteries.
Preparing for the interview
Neve was preparing to interview small business owner-managers about the impact of
the new workplace pensions that had recently been introduced by the government. She
had obtained permission from an organisation that represented small businesses in her
Research in practice 6.3
6.4 Collecting data using semi-structured or unstructured interviews 161
In section 3.8 we talked about the importance of getting participants’ consent for
your research. Many universities expect you to ask each interview participant if they are
willing to take part in the research and, if they are, to sign a consent form like that in
Figure 6.1. This is important, as participants have the right to refuse to take part in your
research or withdraw from the interview whenever they wish.
university town to use of a small ‘interview room’ in their offices. The administrative
assistant had assured her that this room would be quiet and contained a low table and
two comfortable chairs. Neve prepared a notice for the room’s door which said ‘Interview
in progress. Please do not disturb.’
Neve was pleased the room had a low table as this meant there would not be a physical barrier between her and the owner managers she was interviewing. She was more
worried about the comfortable chairs as she knew that, if she sat back and relaxed, her
body language might suggest she was not interested. Neve therefore decided she would
lean forward attentively and ensure she maintained eye contact with each interviewee.
She also thought it was important she dressed formally for the interviews as many of her
participants would probably be wearing suits.
As her first interview was in a few days’ time, she began to develop her interview
guide:
Impact of workplace pensions research
Introduction
• Thank person for attending.
• Explain purpose of research and interview, emphasising that it is participant’s own
opinions that are important.
• Ask if willing to be interviewed and stress this is their decision.
• If willing, ask them to read and sign the consent form; if not, thank them for their time
and close the interview.
Interview
1 To what extent have the introduction of workplace pensions impacted on your
business?
(a)
Probe: In what ways? (ask for examples)
(b)
Probe: Can you give me an example (if possible) of how you think workplace
pensions will benefit your business?
(c)
Probe: Can you give me an example (if possible) of how you think workplace
pensions will cause problems for your business?
2 Do you think the government have provided sufficient guidance for small businesses
regarding workplace pensions?
(a)
Probe: How has this guidance been provided?
(b)
Probe (if insufficient guidance): What extra guidance is needed?
162 Chapter 6 Collecting data
As you can see from the consent form (Figure 6.1), the choice of whether or not to
audio-record the interview is the participant’s, not yours! Although audio recordings
provide an accurate and unbiased record and allow you to listen again to the interview,
if the participant does not wish you to record the interview, then you will have to rely
entirely on the notes you take. This is not easy to do at the same time as interviewing, so
we suggest you practise beforehand. As well as noting what the participant says, you
will need to note where and when the interview was held and relevant background
information about the participant. After each interview, we suggest you word-process
your notes immediately. Leaving your notes for a few days means you are less likely to be
able to read what you wrote!
Consent form
Title of the research project
Researcher’s name, Final year student at University of Anycity
1. I confirm that I understand what the research is about
and have had the opportunity to ask questions.
Please initial box
4. I agree to my interview being audio recorded.
Please initial box
Yes No
5. I agree to the use of anonymised quotations in
publications.
Name of participant:
…………………………….
2. I understand that my participation is voluntary and that
I can withdraw at any time without giving a reason.
3. I agree to take part in the research.
Researcher’s name:
…………………………….
Date:
…………………………….
Signature:
…………………………….
Signature:
…………………………….
Figure 6.1 Participant consent form here
Source: Developed from Saunders et al. (2016).
6.4 Collecting data using semi-structured or unstructured interviews 163
How to ask questions in semi- and unstructured interviews
In both semi-structured and unstructured interviews, you will have to ask questions
carefully and listen attentively to answers given. As in designing a questionnaire, you
will need to make sure the questions you ask are worded clearly using appropriate and
unbiased language. However, unlike when using a questionnaire, you will be able to test
your interpretation by summarising the response and, where necessary, asking further
questions. Brinkmann and Kvale (2014) identify a number of types of questions to use
in interviews. We have listed those our students have found most useful in Table 6.7,
along with a brief description of when to use them, and examples.
Table 6.7 Types of question for use in interviews
Type Use when . . . Examples
Introductory . . . you are starting a new topic. Could you tell me about . . .?
Probing . . . you want to find out more detail,
but without saying what.
Can you say a bit more about that?
Could you let me have a bit more
detail?
Specifying . . . you want to find out more detail
about a specific aspect already
discussed.
Can you say a bit more about why you
purchased a hybrid car rather than an
electric car?
How did you feel when your 91-yearold mother was sent home from the
hospital to an empty cold house?
Direct . . . you want answers about a topic
introduced by the interviewer to apply
to the participant.
Have you ever received a bonus for
good performance?
Have you used a consultant in the
past?
Indirect
questions
. . . you want answers about a topic
introduced by the interviewer to apply
to others.
How do other patients feel about their
treatment at this hospital?
Structuring . . . you want to show questions on a
theme have been completed.
I would now like to ask you about
another topic. Is that all right?
Interpreting . . . you want to check interpretation of
the participant’s response is correct.
You mean that . . .?
Is it correct that . . .?
So what you are saying is . . .?
If you look at Research in practice 6.4, you will see that the types of question you use
and the order in which you use them depends on the responses you receive from the participant during the interview. This means that, unlike with a structured interview, it is not
possible to work out all the questions you will ask or the order in which you will ask them.
This makes semi-structured and unstructured interviews difficult to conduct. In the short
extract in Research in practice 6.4, the interviewer (Deborah Meadon) opens the interview

164 Chapter 6 Collecting data
with the introductory question ‘I think you called yourself the biggest swing and dance
school in the world.’ Without giving participants time to answer, this is followed with the
specifying question, ‘How does that turn into cash?’ The participant’s response, although
detailed, does not provide sufficient detail about profit. A further specifying question is
therefore asked: ‘And net profit?’ Another ‘interviewer’ (dragon), Peter Linney, introduces
a new topic asking: ‘So how does it work? It’s like Sumba really isn’t it in terms of the
model?’ Although two questions have been asked, only the second is answered. This illustrates clearly how participants often fail to answer all questions if more than one is asked at
the same time. It is therefore better to only ask one question at a time.
Why a pilot test is important
Before conducting your interview with selected participants, you need to pilot-test your
interview and technique and check that your questions are likely to be understood, are
Interviewing
In the BBC reality business television show Dragons’ Den, entrepreneurs present their
business ideas to a group of investors (Dragons) in the hope of getting investment capital in return for a stake in their business. After their presentations, the Dragons interview
entrepreneurs about their business ideas.
In 2014 Scott Cupit asked the Dragons for an investment of £65,000 in his Swing
Dance Business ‘Swing Patrol’. After a swing dance-inspired presentation, he was interviewed by the five Dragons, which included the entrepreneur and investor Deborah
Meadon and businessman and investor Piers Linney. An extract from a transcript of the
televised interview shows some of the questions asked by Meadon and Linney and the
responses from Scott Cupit.
Meadon: Scott, let’s look at the business. I think you called yourself the biggest swing
and dance school in the world. How does that turn into cash?
Cupit: The revenue has had a steady growth over the last five years. It started at, perhaps, 36,000 and it gone up to 84, 120, 180, 210 and the last financial period is 280
thousand. Gross profit has reached 190 thousand.
Meadon: And net profit?
Cupit: This year was 67 thousand. It’s had a solid growth and all the projections are it
should continue to grow.
Linney: So how does it work? It’s like a Sumba really isn’t it in terms of the model?
Cupit: We wouldn’t pretend to be as big as Sumba because it’s so massive . . .
At the end of the televised interview one of the Dragons, Meadon, agreed to an investment of £65,000 in Swing Patrol London in return for a 20% stake in the company. The
entertainment brand now boasts a community of over 12,000 dance troop members and
the Guinness World Record for the world’s largest Charleston dance with 975 people.
Sources: Dragons’ Den (2014), Dunsby (2016).
Research in practice 6.4
6.4 Collecting data using semi-structured or unstructured interviews 165
not leading and will provide you with the data you need. This will give you an idea of
possible problems with questions as well as how long each interview is likely to take.
You also need to be sure that your audio recorder works properly, or you can take sufficient notes while interviewing. This is important for both semi-structured and unstructured interviews. Even if you are really short of time, you still need to pilot-test your
interview with a small number of people who are like those who will be participating in
your research. It is worth remembering that, unlike with questionnaires, it may be possible to partially overcome mistakes made in early interviews, such as missing out a
topic that later appears to be important, by amending later interviews.
How to conduct semi- and unstructured interviews
Semi-structured and unstructured interviews can be conducted face to face, by telephone or Internet-mediated using email, messaging software such as Messenger, web
conferencing or video chat apps such as Skype
and Facetime. For such interviews, the
crucial issue is the number of participants you will need to interview to answer your
research question. This depends on the nature of your research question and your population. Many texts recommend you establish the number of interviews you need inductively, simply continuing to conduct interviews until
data saturation is reached, that is,
until each additional interview provides no new insights. While this is good advice, it is
not particularly helpful when you are assessing research undertaken by others using
semi-structured and unstructured interviews, particularly as, for the former, this is not
always stated in journal articles. When planning your own interviews, use the suggestion in section 6.2 that for homogeneous populations the sample size is likely to need to
be between 4 and 12, while for heterogeneous populations the sample size will need
to be larger, say between 12 and 30. Although this does not remove the need for you to
ensure you have enough data to answer your research question or establish data saturation has been reached, it provides an idea of the likely number of interviews needed.
Definition
data saturation: where additional data collection provides few if any new insights into the research
question and objectives.
Conducting semi-structured and unstructured interviews face to face
Conducting semi-structured and unstructured interviews face to face makes full use of
your interviewing skills. The clarity and tone with which you ask questions will be crucial, as will your appearance and body language. To conduct these interviews face to
face, you need to take the following steps:
1 Ensure that your interview checklists and consent forms are printed.
2 Contact potential participants, explain the purpose of your research and invite them
to take part, providing a clear indication of the likely amount of time the interview
will take.

166 Chapter 6 Collecting data
3 If they agree to take part, arrange a mutually convenient appointment and place to
conduct the interview.
4 Arrive for your interview early and make sure the room is set as you wish.
5 Conduct the interview, making sure you also:
(a) thank the participant for attending;
(b) explain the purpose of the research, offering assurances of anonymity and confidentiality as appropriate and explain that they can withdraw at any time;
(c) if using an audio recorder, ask for permission to record the interview;
(d) ask participant to sign the consent form;
(e) remember to also take notes.
6 At the end, thank the participant for their time.
7 Word-process your notes as soon as possible.
At the end of the interview, some researchers suggest it is a good idea to offer participants
a summary of the findings or a copy of your notes of their interview. This is up to you. We
would, however, stress that, if you do make an offer, it is essential that you actually do
provide your participants with the summary of the findings or a copy of the interview
notes. If you do not, they will be less willing to participate in research in the future.
Conducting semi-structured and unstructured interviews by telephone
The process of conducting semi-structured and unstructured interviews by telephone
appears very similar to conducting them face to face, although without the visual contact! This lack of visual contact means conducting such interviews by telephone is not
easy. Remember, the purpose of semi- and unstructured interviews is to explore your
participants’ responses in detail. You will be unlikely to be able to do this unless you
have gained their trust, something that is often difficult to do if you have only talked on
the telephone. You will also not be able to see your participants’ body language, meaning you may interpret some of their answers incorrectly. We therefore recommend that
you only use a telephone for these forms of interview either where you already have met
the participant face to face and gained that person’s trust, or you have already established a trusting relationship.
To conduct a semi-structured or unstructured interview by telephone, you need to
pay attention to the following:
1 Ensure that your interview checklists are printed.
2 Contact potential participants by telephone, explain the purpose of your research
and invite them to take part, providing a clear indication of the likely amount of
time the interview will take.
3 If they agree to take part, arrange a mutually convenient time to telephone and conduct the interview.
4 Telephone and conduct the interview, making sure you also:
(a) thank the participant for their time;

6.4 Collecting data using semi-structured or unstructured interviews 167
(b) explain the purpose of the research, offering assurances of anonymity and confidentiality as appropriate and explain that they can withdraw at any time;
(c) ask for their consent and permission to record the interview (if using an audio
recorder);
(d) remember to take notes.
5 At the end, thank the participant for their time.
6 Word-process your notes as soon as possible.
Conducting semi-structured and unstructured interviews using the Internet
Internet-mediated interviews can be divided into three groups:
those that are conducted in real time (synchronous) using instant messaging software such as WhatsAppor Messenger;
those that are conducted in real time (synchronous) using Voice over Internet
Protocol (VoIP) or web conferencing services such as Skype
;
those that are, in effect, conducted offline (asynchronous), such as through emails,
forums or discussion groups.
Providing your participants are IT-literate and have good access to the Internet, all three
groups have a significant advantage, enabling you to interview participants who are
geographically dispersed. In addition, your participants’ responses can often be
recorded automatically for both typed and audiovisual conversations, although you
should obtain consent to save these recordings.
You will already be familiar with synchronous interviewing software through
your use of Messenger
and WhatsApp. These and other instant messaging software can be used to undertake real-time semi-structured interviews, providing netiquette (section 6.3) is observed. Similarly, you will also be familiar with asynchronous
interviewing software through your use of Internet forums, blogs and email. Internet
forums or discussion groups and blogs usually deal with one topic and need to
remain open for at least a few weeks if they are to generate sufficient posts (data).
Participants can read and comment on each other’s posts, but cannot edit them.
Posts are normally made by a variety of participants, and so they are really group
interviews. In contrast, email interviews consist of a series of email exchanges
between the interviewer and a participant, each consisting of a small number of
questions and the associated answers. Although it is possible to email one long list of
questions, this is really just a questionnaire and means you will not be able to adapt
later questions depending on participants’ responses to earlier ones. Because of the
delay between questions being asked and their being answered, such interviews can
often take more than a week.
To conduct semi-structured and unstructured Internet-mediated interviews, you
need to remember the following:
1 Ensure that your synchronous or asynchronous means of asking questions is set up
and works as you wish.

168 Chapter 6 Collecting data
2 If you are using:
(a) Email, instant messaging, or VoIP services to conduct your interview – contact
potential participants by email, explain the purpose of your research and invite
them to take part, providing a clear indication of the likely amount of time the
interview will take.
(b) Internet forums or blogs – invite people to join your discussion.
3 If you are using:
(a) Email to conduct your interview and those invited agree to take part – ask a few
questions at a time rather than all together.
(b) Instant messaging or VoIP services to conduct your interview and those invited
agree to take part – arrange a mutually convenient time to conduct the interview.
4 Conduct the interview, making sure you:
(a) thank the participant for their time;
(b) explain the purpose of the research, offering assurances of anonymity and confidentiality as appropriate and explain that they can withdraw at any time;
(c) ask for their consent and permission to record the interview or use the transcript
for research purposes;
(d) (where permission has been given) remember to save the interview recording or
transcript.
5 At the end, thank the participant for their time.
We spend our lives observing. When we sit in a coffee bar or café ‘people watching’, we
are observing. When we look at the length of the different check-out queues in our
local supermarket, we are collecting data by observing, although, compared with people watching, our observation has a more clearly defined focus and purpose – to establish which queue is the shortest so that we have to wait the least possible time.
In the earlier sections of this chapter, we have focused on collecting data by asking
people questions. Yet an obvious way of finding out what people do is, rather than ask
them questions, watch and listen to them do it; in other words, to systematically
observe and record their behaviours. We call a person whom we are observing an
informant. In this section, we will look at two forms of collecting data that already
exist: structured observation and unstructured observation.
Structured observation is
6.5 Collecting data using observation
Definitions
informant: a person who is being observed using structured or unstructured observation.
structured observation: a method of observing with a high degree of predetermined structure in the
seeing, hearing and subsequent recording of data to answer questions concerned with how much or
what happens.

6.5 Collecting data using observation 169
concerned with questions such as ‘What happened?’ ‘How much?’ ‘How many?’ ‘How
often?’ or ‘How long?’ Given its name, it is not surprising that what is going to be
observed is both predetermined and highly structured. What is seen and heard by the
researcher is recorded systematically to provide quantitative data on the frequency of
actions, or how long specified actions will take, such as the length of time it takes a
crew member in a burger restaurant to undertake each of the tasks needed to serve a
customer. In contrast,
unstructured observation is concerned with answering ‘why’ questions. It is far less structured, focusing on the physical setting, those being observed and
their activities, and the processes and emotions involved. It uses what the researcher
sees and hears as she or he observes the activities of the research subjects fully to provide more qualitative data about actions such as their meanings or explanations. This
might include how customers respond to different particular crew members, or the reasons why people consider the service received from a particular waiter or waitress in a
restaurant to be exceptional or awful.
In undertaking a structured observation, you make the assumption that what is
being observed, such as a meeting or the serving of a customer in a burger restaurant, can be broken down into a series of discrete aspects or elements. You define these,
using the literature you have reviewed, use a recording sheet to provide focus when collecting the data and passively observe rather than take part. For unstructured observation in its most extreme form, you immerse yourself fully within the situation you are
observing, taking part fully in what you are observing and developing a narrative
account of what is going on. You might, for example, be employed by the organisation
in which you are undertaking the observation. Alternatively, it might involve you taking part in a particular experience such as a summer working holiday in North America.
Because you participate fully, you come to understand the world you are observing,
allowing a deep and nuanced understanding of the interactions and the associated
meanings. Whilst both structured and unstructured observation can offer the advantages of collecting in-depth data in real time and within context, they are extremely
time consuming. In addition, particularly for overt observation, those you are observing (the informants) may act differently because they are being observed; this is known
as the Hawthorne effect.
Definition
unstructured observation: a method of observing in which the researcher observes the activities fully
and there is limited structure in the seeing, hearing and subsequent recording of data to answer
questions concerned with why.
How to prepare for observation
Like semi-structured and unstructured interviews, undertaking observation involves
careful preparation, including ensuring you have both ethical approval and consent to
undertake the research (Figure 6.1). Having decided whether structured or unstructured
observation will be most appropriate to answering your research question, you need to
make sure you know about the people you are observing and the context in which you

170 Chapter 6 Collecting data
will be observing them. You also need to think carefully about what you want to
observe, the degree of focus of your observations and how you will record them to
ensure you have the data you need to answer your research question. For structured
observation, it is likely you will use some form of observation sheet to record your
observations (Research in practice 6.5). In contrast, for unstructured observation, it is
more likely that you will record, usually in note form, as detailed a descriptive account
as possible of what you saw.
Research in practice 6.5
Preparing for the interview
Colin was preparing to observe how team members and the team leader interacted in
team meetings using structured observation. Drawing on the academic literature, he
devised six categories of interaction he was interested in observing. These were:
Providing facts or information
Seeking facts or information
Checking others’ understandings
MEETING OBSERVATION SHEET
Date of Meeting:
Location of meeting:
Purpose of meeting:

Providing facts or
information
Seeking facts or
information
Checking others’
understandings
Providing
clarifications
Expressing
opinions
Summarising what
has been said
Team
Leader
Team
member 1
Team
member 2
Team
member 3

6.5 Collecting data using observation 171
How you plan to record your data will also depend a great deal on the nature of
your observation. In particular, it relates to whether or not you intend to reveal
your identity as a researcher to those you are observing and whether or not you
will take part in the activity you are observing. This gives four possible scenarios
(Figure 6.2).
Researcher takes part,

observes and their
dentity is revealed
observes and the
Researcher only
observes and their
dentity is revealed
Researcher only
observes and the

Researcher takes part,
identity is concealed
identity is concealed
Researcher takes
part in activity
Researcher
observes activity
Researcher’s
identity is
concealed
Researcher’s
identity is
revealed
Figure 6.2 Different scenarios for observation
Colin obtained written permission from the company’s owner to observe the
meeting. He had stated that he would observe the meeting overtly and knew he
would have to explain to those he was observing how he would preserve their
anonymity. For this reason, he did not record names on his observation sheet.
Although he was concerned that his presence in the meeting and his obvious note
taking would have an impact on the interactions he was observing, he believed
the impact would lessen over time as they became used to his sitting in the corner of the
meeting room.
Providing clarifications
Expressing opinions
Summarising what has been said
He then devised an observation sheet so he could record each time a team member or
the team leader undertook one of these types of interaction using a tally mark. This provided a quick way of recording interactions in groups of five. He made one vertical line
for each of the first four interactions, the fifth interaction being recorded by a diagonal
line across these four.

172 Chapter 6 Collecting data
If those you will be observing will know you are observing them, then it is likely to be
possible for you to make notes as things happen, providing you are not taking part.
However, if you plan to undertake covert observation, then making notes immediately,
even if you are not taking part in the activity, may be difficult. You therefore need to
think carefully about where and when it will be possible to make notes. One of Mark’s
students who observed customer behaviour in a café frequented by tourists wrote most
of her notes on postcards, as she felt this would be a normal activity in a tourist café.
Obviously, if you plan to take part in the activity you are observing, it will be more difficult to record your observations, particularly if you are observing covertly. We have
both heard of researchers undertaking covert observations who wrote most of their
notes while taking a comfort break on the toilet! Whether your observations are overt
or covert, or whether you are taking part or just observing, we strongly recommend that
you plan to record your observations as soon as possible and always on the same day as
they occur. If you do not manage to do this you will, unfortunately, lose valuable data.
In preparing for observation in particular, you therefore need to:
Be clear about the purpose of the observation and how the data you intend to collect
will help you answer your research question.
Make sure you have found out as much as possible about the person or people you
are going to observe, the situation you intend to observe them in and, where appropriate, the organisation where they work.
Be clear about whether you will be taking part in the situation you intend to observe
or just observing.
Be clear about whether you will identify yourself to those you are observing or, alternatively, your observations will be covert and the likely implications for how those
you are observing will act.
For structured observation, develop an observation sheet (Research in practice 6.5)
listing those aspects you want to observe and the detail you wish to record for each.
This will be informed by the literature you have already reviewed.
For unstructured observation, think carefully about the aspects you want to observe
and how you will record your observations when you immerse yourself in the
research setting.
Make sure your clothing and appearance is appropriate for how you intend to conduct your observation, be it face to face, Internet-mediated, or using videography.
How to conduct observation
Observations can be conducted in-person (face to face), mediated by the Internet and
by using videography.
Internet-mediated observation involves collecting data by
Definition
Internet-mediated observation: an adaptation of in-person observation from oral and visual observation
to textual and digital observation in a virtual environment of an online community or communities.

6.5 Collecting data using observation 173
observing online communities replacing the oral aspects of observation with the
text, and the visual aspects with the digital images. As with in-person observation,
you can either take part in the activity you are observing or just observe. Depending
on the accessibility of an online community, it may also be possible for you to enter a
discussion group or online forum as a guest without revealing your identity and without participating other than reading or viewing the available material. This is known
as lurking and as your purpose as a researcher is not revealed, it may be considered
unethical by your university. This can lead to your being asked to leave if you reveal
yourself as a researcher and ask to undertake research overtly. Observation using
videography involves recording moving images electronically as observational data. It
can therefore be used to record research informants in a research setting such as
through a body-worn camera. You can also ask informants to record their own video
diaries, something that is very useful where it is difficult to gain personal access as an
observer.
Definition
videography: the recording of moving images and associated sound as observational data.
The way in which you conduct your observation will, as highlighted earlier, depend on
whether or not you reveal your identity, whether or not you are taking part in what you
are observing and whether your observation is structured or unstructured.
Conducting observation in person
Conducting observations in person share a number of similarities to conducting interviews. To conduct both structured and unstructured observation, you need to:
1 Ensure you are as clear as possible about the purpose of your observation.
2 Be clear about what it is you want to observe, ensure you have sufficient copies of the
means of recording your observations, such as an observation sheet (structured
observation) or notebook (unstructured observation) and sufficient consent forms
(if your identity as researcher will be revealed) printed.
3 Ensure you are clear how you will manage observing and recording data, particularly
where you are taking part in what you are observing.
4 Arrive for your observation early and, if you are not taking part, make yourself as
unobtrusive as possible.
5 Conduct your observation maintaining a positive but non-threatening self-image
and trying to ensure that those you are observing do not depart from their usual
ways of doing things. Where your identity is revealed:
(a) thank those being observed for allowing you to observe them;
(b) explain the purpose of the research, offering assurances of anonymity and confidentiality as appropriate and explain that they can withdraw at any time;

174 Chapter 6 Collecting data
(c) ask for permission to record electronically what is observed using the text or
video files;
(d) ask the participant to sign the consent form.
Remember, many universities will not give ethical approval where the researcher’s
identity is not revealed.
6 Record your observations as precisely as possible and as close to the time they occur:
(a) for structured observation, use your observation sheet;
(b) for unstructured observation, include descriptions of people, events and conversations as well as your own actions and feelings in relation to what you are
observing.
7 At the end, if you have revealed your identity, thank those you have been observing
for their time.
8 Word-process your observation notes as soon as possible.
Conducting Internet-mediated observation
The process of conducting Internet-mediated observations appears very similar to conducting them in person, although without the personal contact. However, you will also
need to determine whether what you are observing online represents all of the interactions, or whether those in the community also interact offline. You also need to be
aware that members can adopt different personas online to both protect the identities
and so they can offer views that they would never talk about face to face.
When conducting Internet-mediated observation you need to pay attention to the
following:
1 Ensure you are as clear as possible about the purpose of your observation.
2 Be clear about what it is you want to observe, that you are able to record your observations
either electronically or by paper and that your consent form is available electronically.
3 Conduct your observation maintaining a positive but non-threatening self-image
and trying to ensure that those you are observing do not depart from their usual
ways of doing things. Where you reveal your identity:
(a) thank those being observed for allowing you to observe them;
(b) explain the purpose of the research, offering assurances of anonymity and confidentiality as appropriate and explain that they can withdraw at any time;
(c) ask for permission to observe;
(d) ask participants to sign the consent form electronically.
Remember, many universities will not give ethical approval where the researcher’s
identity is not revealed.
4 Record what you are observing electronically as precisely as possible and as close as
possible to the time they occur. For unstructured observation, remember also to
record descriptions of people, events and conversations as well as your own actions
and feelings in relation to what you are observing.

6.5 Collecting data using observation 175
5 At the end, if you have revealed your identity, thank those you have been observing
for their time.
6 Word-process your observation notes as soon as possible.
Conducting observations using videography
You can collect your own observational data by, for example, videoing what you are
observing or asking your informants to provide you with their own videos. These are
relatively easy from a technical perspective, due to the relatively high-quality video
recording available on mobile phones. They can be undertaken both overtly and
covertly, the latter using a hidden camera. However, beware, recording an observational video is not as easy as it seems and we strongly recommend that you practice
with the equipment before observing for your research. Despite this, we have already
noted, it is easy to miss important data when undertaking observation, and videorecording can go some way towards overcoming this problem. Although the place
from which the video is shot gives a particular viewpoint, recording allows the video
to be replayed to gain a deeper understanding as you reflect on what is being
observed.
When conducting observation using videography, you need to pay attention to the
following:
1 Ensure you are as clear as possible about the purpose of your observation.
2 Be clear about what it is you want to observe and, even if you are using your mobile
phone to record the video, ensure that you are able to use the video equipment to
record your observations, alongside printing sufficient consent forms (if your identity as researcher will be revealed).
3 Conduct your observation, maintaining a positive but non-threatening self-image
and trying to ensure that those you are observing do not depart from their usual
ways of doing things. Where you reveal your identity:
(a) thank those being observed for allowing you to observe them;
(b) explain the purpose of the research, offering assurances of anonymity and confidentiality as appropriate and explain that they can withdraw at any time;
(c) ask for permission to observe;
(d) ask informants to sign the consent form.
Remember, many universities will not give ethical approval for covert videorecording.
4 Video-record what you are observing. For unstructured observation, remember to
also note descriptions of people, events and conversations as well as your own
actions and feelings in relation to what you are observing.
5 At the end, if you have revealed your identity, thank those you have been observing
for their time.
6 Word-process your observation notes as soon as possible.
176 Chapter 6 Collecting data
Your choice of sampling technique depends on whether or not you can obtain a
complete list of the population (sampling frame), your research question and your
objectives.
If you can obtain a sampling frame, you can use probability sampling techniques
such as simple random, systematic random and stratified random sampling. You use
probability samples if you want to estimate statistically the characteristics of the
population.
If you cannot obtain a sampling frame, you must use non-probability sampling techniques such as quota, purposive, snowball and self-selection sampling. We recommend you do not use convenience sampling. You use non-probability samples when
you want to make logical generalisations.
Questionnaires are used when you want to collect data by asking each person to
answer the same set of questions in the same order.
Before designing your questionnaire, you need to know what data you need to collect to answer your research question and meet your objectives. Individual questions
should then be designed before putting the questionnaire together. In designing
your questionnaire, use as few open questions as possible. Remember to design the
questionnaire so it is easy to read and respond to questions.
Questionnaires can be distributed face to face by an interviewer, by telephone, by
hand, by post and by using the web.
Semi-structured and unstructured interviews are used when you are unsure of the
answers respondents will give, your questions are complicated or you need to vary
the order of questions or the actual questions asked.
In semi-structured interviews, you ask about a set of themes, only some questions
being predetermined. The order in which the themes are covered and questions
asked can be varied. You can choose to omit some topics and questions and ask additional questions as appropriate.
In unstructured interviews, the participant talks openly and widely about the topic
with as little direction from you as possible. Although there is no predetermined list
of questions, you will have a clear idea of the topics to explore.
To prepare for semi-structured and unstructured interviews, you need to be clear
about the topics you are going to discuss, the questions you are going to ask, the location of the interview and your appearance. When you conduct the interviews, you
need to test your understanding as necessary, show that you are listening and make a
record of the conversation.
Semi-structured and unstructured interviews can be conducted face to face, by telephone and using the web.
Observation is used when you want to find out what people do by watching them
rather than asking questions.
Summary
References 177
Structured observation is concerned with systematically recording what has happened, often using a recording sheet, to provide data that is often quantified.
Unstructured observation is concerned with recording what is happening more
broadly. It is concerned with the physical setting, those being observed, the activities, the processes and your associated emotions and observations.
In preparing for both structured and unstructured observation, you need to be clear
about whether or not you will take part in the activity you will be observing, and
whether or not you will reveal your identity as a researcher to those you are observing. These decisions will impact on how you record your observations.
Observation can be conducted in person, mediated by the Internet or using
videography.
As you continue to read and note the literature, use your knowledge about samples,
questionnaires and interviews to help you assess the appropriateness of the methods
used by other researchers and the value of their research findings (section 2.7).
Think about the data you will need to answer your research question and meet your
objectives.
If you are going to use secondary data, use your knowledge about samples, questionnaires, interviews and observation to help you assess the suitability of the secondary
data for your own research (section 4.5). Make notes about this, as you will need this
level of detail to write the method section of your project report.
If you are going to collect your own (primary) data, use your knowledge about
samples, questionnaires, interviews and observation to help you choose the most
appropriate data collection method or methods (Chapter 6).
If you are going to collect your own (primary) data, make notes about the reasons for
your choices about selecting your sample and the method or methods you will use to
collect your data. Also, make notes explaining why you need to collect the data you
wish to collect, linking these data explicitly to your objectives. You will need this level
of detail to write the method section of your project report.
Thinking about collecting data
Baruch, Y. and Holtom, B.C. (2008). Survey response rate levels and trends in organizational
research.
Human Relations, 61(8), 1139–60.
Brinkmann, S. and Kvale, S. (2014).
InterViews: Learning the Craft of Qualitative Research
Interviewing
, (3rd ed.), Los Angeles, CA: Sage.
Cassell, C. (2015).
Conducting Research Interviews for Business and Management Students,
London: Sage.
References
178 Chapter 6 Collecting data
Dragons’ Den (2014). ‘Scott Cupit and Swing Patrol on BBC’s Dragon’s Den’ YouTube. Available
at: https://www.youtube.com/watch?v=lHYnxtR1u0I [Accessed 8 November 2016].
Dunsby, M (2016).
Dragons’ Den Success Stories: Swing Patrol. Available at: http://startups.
co.uk/dragons-den-success-stories-swing-patrol/ [Accessed 8 November 2016].
Ekinci, Y. (2015).
Designing Research Questionnaires for Business and Management Students,
London: Sage.
Saunders, M., Lewis, P. and Thornhill, A. (2016).
Research Methods for Business Students (7th ed.).
Harlow: Pearson
Saunders, M.N.K. (2011). Choosing research participants.
in C. Cassell and G. Symons (eds),
The Practice of Qualitative Organizational Research: Core Methods and Current Challenges.
London: Sage.
Saunders, M.N.K. and Townsend, K. (2016). Reporting and justifying the number of interviews
participants in organisation and workplace research.
British Journal of Management, 27(4),
836–52.
YouGov (2016).
Panel Methodology Available at: https://yougov.co.uk/about/panel-methodology/
[Accessed 7 November 2016].

Chapter 7
Analysing data
In Chapter 6, we noted how it was important to know about and understand different
methods of collecting data, even if you were not collecting data yourself for your research
project. We now make similar comments about analysing data. Obviously, if you collect
your own data, you will need to analyse these data to answer your research question and
meet your objectives. Similarly, if you’re using secondary data (Chapter 4) for your own
research project, you will still have to analyse these data. However, even if your project is
an extended essay or literature review, you will still need to know about different techniques for analysing data. Without this knowledge, you will not be able to understand or
evaluate fully the journal articles, reports and book chapters you review. If you only understand partially how the data were analysed and are unclear about the reasons why particular techniques were used, your ability to assess the quality of analysis, understand the
research findings or follow the discussion of these findings will be reduced. In addition, a
reasonable understanding of techniques for analysing data will help you to be clear about
the value of research reported in the literature to your own project.
So what do you need to know about analysing data to help you do your project and
write the project report? Well, whether you’re collecting your own data, using secondary
data or basing your project entirely on the literature, you still need to know the same
things. You need to know how to get data ready for analysis, when to use different analysis
techniques and how to interpret the results of these different analysis techniques. In addition, you need to know a bit about different types of data and the analysis techniques you
should use depending on the type of data you have.
The first section of this chapter is about the different types of data you might be analysing or reading about. As you would expect from reading about collecting data (Chapter 6),
this section looks at both quantitative and qualitative data. Following a similar structure to
Chapter 6, the next section looks at how to analyse data quantitatively. Within this, we
look at how to prepare your data for quantitative analysis using a spreadsheet such as
7.1 Why you should read this chapter
180 Chapter 7 Analysing data
Microsoft Excel or statistical analysis software such as IBM SPSS Statistics. We also look at
when you should use different tables, graphs and some of the more straightforward statistics as well as what the results of your analysis mean. Our final section looks at how to
analyse data qualitatively. Once again we will look at how to prepare your data for analysis, this time looking mainly at manual analysis. Within this, we talk about ways of coding
and analysing your data as well as what the results of your analysis mean.
Data
Quantitative Qualitative
Categorical Numerical
Nominal
(descriptive)
Ordinal
(ranked)
Interval Ratio
Text Non-text
Audio Still
Image
Video
Figure 7.1 Types of data
Data can be split into two main types: quantitative data that are numerical or whose
values have been measured in some way, and qualitative data that are not numerical
and have not been measured in some way (Figure 7.1). As we saw in section 6.3, quantitative data are collected in a standardised way, such as by using questionnaires. These
data are collected about different variables (we defined this term in section 1.7), which
are the building blocks of quantitative analysis. Your variables are usually described
using numbers and analysed using diagrams and statistics, including testing
hypotheses.
In contrast, qualitative data are collected usually in a non-standardised way, such as
interviews (section 6.4). These data are analysed by developing and testing propositions, using justified argument. To be able to analyse both quantitative and qualitative
data, you need to know more about each of these types, as they will dictate the analyses
you can do.
Quantitative data
As you can see in Figure 7.1, quantitative data are split into two main types: categorical
and numerical data. Categorical data, as the name suggests, are data that have been
grouped into a descriptive set or put in rank order. Let’s say you have collected data
7.2 Different types of data
7.2 Different types of data 181
from a sample of students at your university using, among others, the category
question:
In which faculty are you studying? Arts and Humanities
1
Business and Law
2
Physical Sciences
3
Social Sciences
4
The data you get from this category question will be descriptive, your variable ‘faculty
in which studying’ being made up of a descriptive set of categories that are the four faculties in your university. Although these data are coded using numbers, the numbers do
not mean that the Arts and Humanities faculty is first, Business and Law second and so
on; they are just codes, and there is no relevance to the number order. In contrast,
ordinal data are categorical data that have been put into rank or definite order. These are
often called
ranked data. In your questionnaire, the next question is one that also
appeared in Table 6.3:
How likely do you believe it is that you will pay off your
overdraft within a year of graduation?
Very likely

1
Quite likely
2
Unsure
3
Quite unlikely
4
Very unlikely
5
This rating question collects data on the variable ‘likelihood of paying off overdraft
within a year of graduation’. The responses are in ranked order, ‘very likely’ being
ranked first, ‘quite likely’ second and so on.
Numerical data are where the data are measured using numbers and can be measured using either intervals or ratios. For
interval data, you can state the difference
between any two data values: such as, the difference between 3° Celsius and 9° Celsius is
6° Celsius. However, you can’t say that 9° Celsius is three times as hot as 3° Celsius! For
ratio data, you can say what the actual difference and what the relative difference are
between two values. For example, if one student’s overdraft is £15,000 and another student’s overdraft is £30,000, you can say that the difference between them is £15,000,
the second student having twice as large an overdraft.
Definitions
ordinal or ranked data: categorical data that are put into a definite order.
interval data: numerical data whose values are measured numerically so that the numerical difference between two values can be stated, but not the relative difference.
ratio data: numerical data whose values are measured numerically so that both the numerical and
the relative difference between two values can be stated.

182 Chapter 7 Analysing data
Numerical data can take either discrete values (discrete data) or any value (continuous
data
). If we return to your student questionnaire, we can again use questions to illustrate
these two types of numerical data. In your questionnaire, the next question is:

On how many days did you visit the library last week?
(For example, for 2 write: [ 2 ] )
[ ]

This quantity question collects discrete numerical data on the variable ‘number of days
visited the library last week’. The values for this variable can only be 0, 1, 2, 3, 4, 5, 6
or 7, depending on the number of days a respondent visited the library last week. In
contrast, your next quantity question collects continuous data for the variable ‘time
spent studying last week’. Although this variable is measured to the nearest hour, these
data could theoretically take any value, depending on the level of accuracy used:

How much time did you spend studying last week to the nearest hour?
(
For example, for 2 hours, 15 minutes write: [ ] [ 2 ] hours)
[ ] [ ] hours

You will notice from the definitions above that nominal data are sometimes called
descriptive data, and ordinal data are sometimes called ranked data. Where these terms
are used, you will often find that numerical data has been divided into continuous and
discrete data rather than interval and ratio data. Don’t worry; it’s just a different way of
grouping data variables!
Definitions
discrete data: numerical data whose values are measured numerically as quantities in discrete units
and can therefore only take a finite number of values.
continuous data: numerical data whose values are measured numerically as quantities and can
theoretically take any value, depending on the level of accuracy with which they are measured.
nominal or descriptive data: categorical data that are grouped into sets (categories) that have no
obvious rank order.
Qualitative data
Looking again at Figure 7.1, you can see that qualitative data are also split into two
main types: text and non-text. Text data, as the name suggests, are data in the form
of words that have been recorded as text and are usually word-processed. They therefore include written answers to requests such as, ‘If there are any other areas or issues
that concern you please feel free to comment below’, in Table 6.3. The majority of
audio recordings, and the audio parts of video recordings, are transcribed and wordprocessed and then analysed as text data. As we discovered in Research in practice 6.4,
these are known as interview transcripts and contain both the questions asked by the
interviewer and the participant’s answers. These transcripts often also record details
about the research setting as well as the interviewer’s and the participant’s non-verbal

7.3 Analysing data quantitatively 183
behaviour. In contrast, although methods of analysing visual data such as video and
other images are developing, they are not yet in widespread use and consequently are
outside the scope of this text.
How to prepare your data
If you’re analysing your sample data quantitatively, you will almost certainly be using
either a spreadsheet or statistical analysis software. These software require your data to
be in the format of a
data matrix. Look at Figure 7.2. This extract from a spreadsheet data
matrix has one column for each variable, representing questions in your student questionnaire (section 7.2). The first row in the spreadsheet contains a short description of
each data variable. Each of the remaining rows contains the data for all these variables
for one student, each cell containing the code representing a student’s response for a
particular variable.
The codes for each variable mean different things. If you look at the first row of data
(row 2 in the spreadsheet), you can translate the codes. The first cell in the row contains
the student identifier. This is not to identify the student by name! Rather, by writing
this number on the questionnaire containing these data, you can link the questionnaire to your data matrix. This makes it much easier for you to check your data for possible typing errors. Student 1 is studying in the ‘Faculty of Business and Law’ (code 2).
7.3 Analysing data quantitatively
Figure 7.2 Data matrix in Excel
Definition
data matrix: the table format in which data are typed into spreadsheets or statistical analysis software. Each column represents a separate data variable, and in each row a separate member about
whom data have been obtained.

184 Chapter 7 Analysing data
The likelihood of this student’s overdraft being paid off within one year of graduation is
‘quite likely’ (code 2). This student visited the library three times last week and spent 28
hours studying. Numeric codes have been used because they make subsequent analysis
more straightforward. In addition, numbers are quicker to type in and you’re less likely
to make errors.
If you have collected your own data, but not used an Internet questionnaire, you will
need to code it and type it into your analysis software as a data matrix. To do this, you
will need to do the following:
1 Work out the number of variables and give them clear names.
2 For each variable, work out a coding scheme (if you collected your data using a questionnaire, you should have already done this for most variables; see section 6.4).
3 Code each variable, leaving variables with no data blank.
4 Set up your data matrix as in Figure 7.2.
5 Type in your data, saving your file regularly and also making a backup copy.
6 Check your data for typing errors.
No matter how carefully you type in your data, you’re almost certain to make a few
errors. You should always check your data for errors before starting your analysis. This
will save time as it prevents you having to redo analyses because your data were wrong.
Two common errors you should look for are:
Illegitimate codes – these are code numbers which appear in a data variable that you
have not used for that data variable.
Illogical relationships – these are relationships that are very unlikely to occur in your
data, such as a person aged 91 who is still in full-time work.
As we pointed out in section 4.2, online secondary data are in formats that can be read
directly into your data analysis software as a data matrix. These data have already been
checked for errors and, as you will not have to type the data in yourself, you will save
time. However, don’t forget to download the codebook and list of definitions so you
know exactly what everything means and can interpret the data correctly.
How to present data
Your first stage of any quantitative data analysis is to explore and understand your data.
The easiest way to do this, and we believe a most helpful way, is by presenting your data
as tables and graphs. Those that are most useful in helping you answer your research
question and meet your objectives will also appear in your project report. However,
before you start designing tables and graphs, you need to be aware of the following:
Particular tables or graphs are better than others for highlighting certain aspects of
your data (Saunders et al., 2016).
Categorical and numerical data often require different tables and graphs.
All tables and graphs must be labelled clearly and designed so as not to distort the data.
7.3 Analysing data quantitatively 185
In this section, we look at how to highlight different aspects of your quantitative data
using tables and graphs. Within this, we talk about those tables and graphs which are
used to present categorical data and those which are used to present numerical data,
also looking at the importance of clear design and labelling. These are summarised in
Table 7.1.
Table 7.1 Presenting data using tables and graphs

For . . . use a . . . to present . . . which allows the . . .
categorical data table a summary individual values to be read
categorical data bar graph values for each
category
highest and lowest values for
a variable to be seen
2
numerical
continuous data
1
histogram values for data
grouped in categories
highest and lowest values for
a variable to be seen
categorical data1 pie chart proportions in each
category
relative proportions in each of
a variable’s categories to be
seen
numerical data line graph values for a variable
over time
trend over time to be seen2
numerical and
categorical ranked
data
scatter graph a relationship relationship between two
variables to be seen

1 Numerical data will need to be grouped into categories.
2 To also compare variables, use a multiple bar chart or a multiple line graph.
How to summarise data so specific values can be read
Tables are used when you want to summarise data so that specific values can be read
easily. When your table summarises the data for one variable, such as the number in
each category for a question such as ‘In which faculty are you studying?’ it is sometimes
called a frequency table or frequency distribution.
Now read Research in practice 7.1. The table Harry pasted into his project report
summarises the number of hectares used for organic farming in each member of the
European Union’s 27 member states. Each EU member state is represented by a separate row. The table contains data for more than one variable. The total number of
hectares used for organic farming in 2010 and 2015 for each member state is represented by the second and third columns. A further variable, represented by the fourth
column, has been included. This has been calculated using the second and third columns of data.
While the individual values in this table are easy to read, their meaning is less easy to
understand. This is because Harry has simply cut and pasted this table into his project
report rather than thinking! He needs, as pointed out by his supervisor, to make sure
the table has a clear title, states the source of the data and that the various abbreviations and notes are explained more fully.

186 Chapter 7 Analysing data
How to present data so the highest and lowest values can be seen
As you can see from Research in practice 7.1, tables do not emphasise any particular
values. If you wish to do this, you need to use a bar graph (sometimes called a bar chart)
for categorical data or a histogram for numerical data. The bar graph in Figure 7.3 shows
visually the organic farming area for each European Union member state: Spain has the
largest area of organic farming, while Cyprus, Luxembourg and Malta have the smallest
areas of organic farming. Because of the scale at which the graph is drawn, it is not possible to be sure which of these three countries has the smallest area. However, if you
look at the specific values in the table in Research in practice 7.1, you will find out that
the lowest member state is Malta, with only 30 hectares of organic farming.
Designing and labelling tables
As part of his research project on the market for organic food, Harry decided to present
some of the secondary data on the growth of organic farming, using a table he had
found on the European Commission’s website. This showed the organic crop area in EU
member states. He cut the table from the web page and pasted it into his chapter, which
he later emailed to his project supervisor for comments. The extract below includes his
supervisor’s comments.
Source of table: © Eurostat Press Office (2016).
Research in practice 7.1
7.3 Analysing data quantitatively 187
More complicated forms of bar graphs can be used to compare the highest and lowest
values for two or more variables. These are known as multiple bar graphs. When you
draw these, you need to make sure the data you want to compare are represented by bars
that are next to each other. So, if Harry wished to compare the total organic farming
area in each EU member state between 2010 and 2015, he would draw the two bars for
each member state next to each other. For each member state, the first bar would represent the total organic farming area for 2010 and the second the total area for 2015.
2000000
1800000
Hectares
European Union member state
Total organic farming area by European Union member state (2015)
Source: Eurostat (2016)
1400000
1600000
1200000
1000000
600000
800000
200000
400000
Belguim
0
Bulgaria
Czech Republic
Denmark
Germany
Estonia
Ireland
Greece
Spain
France
Croatia
Italy
Cyprus
Latvia
Lithuania
Luxembourg
Hungary
Malta
Netherlands
Austria
Poland
Portugal
Romania
Slovenia
Slovakia
Finland
Sweden
United Kingdom
Figure 7.3 Bar graph
Source: Eurostat (2016).
Histograms are used if you have numerical continuous data, the continuous nature
of the data on the horizontal access being shown by there being no gaps between the
bars. Let’s look at numerical data you collected from students using the question:

How much time did you spend studying last week to the nearest hour?
(
For example, for 2 hours, 15 minutes write: [ ] [ 2 ] hours)
[ ] [ ] hours

These data would consist of a series of numbers, such as 23, 34, 8, 14, 46 . . ., each
number being the number of hours a student had spent studying last week. To draw

188 Chapter 7 Analysing data
a histogram, you could place these data into a series of equal width categories
such as:

less than 5 hours
5 to less than 10 hours
25 to less than 30 hours
30 to less than 35 hours

10 to less than 15 hours 35 to less than 40 hours
15 to less than 20 hours 40 to less than 45 hours
20 to less than 25 hours 45 hours or more
As one of these categories ends, the next starts, the lack of a gap between the categories
reflecting that your data are continuous. Your histogram would consist of 10 bars, the
area of each bar representing the total number of students whose time was spent studying last week for a five-hour time period (Figure 7.4). Because there is no gap between
each of your categories, there is no gap between each of the bars in your histogram
(Figure 7.4). Once again, the highest and lowest values are easy to see: the most frequent
amount of time spent studying last week by students was 20 to less than 25 hours, the
least frequent being 45 hours or more.
30
Number of students
Hours spent studying
Hours spent studying by students last week
Source: Surveys of Students, 20##
25
20
15
10
5 0
0 5 10 15 20 25 30 35 40 45 50 plus
Figure 7.4 Histogram
How to present data so that proportions can be seen
As we are sure you already know, pie charts are useful if you want to show the proportions in different categories for a variable. As with histograms, if you’re using numerical
data, you need to put these data into categories. The pie chart in Figure 7.5 has been
drawn using categorical data collected from your student questionnaire which used
responses to the question, ‘In which faculty are you studying?’ The largest segment,
Business and Law, represents the faculty that had the greatest proportion of students
responding. We’ve added the annotations to remind you how to design and label

7.3 Analysing data quantitatively 189
graphs, whatever type you’re using. You will notice that one annotation states ‘(Twodimensional graph used so as not to distort or misrepresent the data)’. While it is fun to
draw three-dimensional graphs using multiple colours viewed from weird angles, a
quick look at some of graphs in newspapers shows how easy it is to use them to misrepresent data. You should not do this in your project report.
9%
Physical Sciences
Clear title Denser shading
used for smaller
areas
Legend included
(as needed)
Source of data
stated
Size of sample
stated
Units of
measurement
stated
(Segments in
a logical
sequence –
smallest to
largest)
(Two-dimensional
graph used so as
not to distort or
misrepresent
the data)
Proportion of students by faculty
Source: Survey of students, 20##
Sample size = 134
21%
28%
42%
Arts and Humanities
Social Sciences
Business and Law
Figure 7.5 Annotated pie chart
How to present data so that a trend can be seen
You should use line graphs to highlight changes in numerical data over time (Table 7.1).
For most research projects, your line graphs will be drawn using secondary data such as
that available from government websites (Research in practice 7.2) or online market
and financial databases. If you want to compare trends over time for two or more variables, you simply draw an additional line for each variable, adding a legend so you know
what each line represents.
How to present data so interdependence between two variables can be seen
Interdependence or the relationship between two numerical variables is best shown
using a scatter graph, sometimes called a scatter plot (Table 7.1). Let’s say an organisation has asked you to find out if there is any relationship between how well its 30 sales
employees performed in an aptitude test and their sales performance. The first of your
variables is each employee’s aptitude test score out of 100. The second is the value of
each employee’s total sales for the past month recorded in pounds sterling. As you feel
that sales performance is likely to be affected by aptitude, you decide that sales performance is your dependent variable and so should be plotted against the vertical axis of
your scatter graph. Aptitude is the independent variable, and so you plot this on the
horizontal axis (Figure 7.6). Each point (cross) on your scatter graph represents an

190 Chapter 7 Analysing data
Presenting a trend
Ahmed’s research project was looking at long-term economic trends. In particular, he
was interested in changes in countries’ balance of payments since the end of the
Second World War and the relative importance of different goods and services over
the years. Using the UK Government’s website (www.statistics.gov.uk) as one of his
sources of secondary data, he downloaded the Balance of Payments time series data
set free of charge. This contained a summary of the UK’s total trade in goods recording the value of imports, exports and the balance of trade for the years 1946–2015.
To show the trend in the balance of payments for trade in goods, he used a line graph.
This showed a clear decline over time, this decline being most rapid between 2011
and 2013.
10
0
Billion pounds at current prices
UK Total Balance on Trade in Goods 1946–2015
–90
–80
–70
–60
–50
–40
–30
–20
–10
–100
–110
–120
–130

1951
1956
1961
1966
1971
1976
1981
1986
1991
1996
2001
2006
2011
Year

1946
Source: © Office for National Statistics (2016). Contains public sector information licensed under the Open
Government Licence v3.0.
Research in practice 7.2
employee. The closer these points on your scatter graph are to an imaginary straight
line (not shown on Figure 7.6 as it is imaginary!), the stronger the relationship between
the two variables. Looking at your graph, you decide that there appears to be a strong
relationship between an employee’s aptitude test result and their sales performance.
As we will see later, you decide to explain this cause-and-effect relationship
statistically.
How to describe data using statistics
We’re sure that you’ve already studied statistics at some time in the past. However,
we wouldn’t be surprised if you had forgotten at least some things. We don’t intend
to give you a lecture on how to calculate lots of different statistics in this section.
After all, there is no real need to do this, as you will be using either a spreadsheet or

7.3 Analysing data quantitatively 191
some other statistics software to calculate the statistics for you. Rather, we will talk
about the statistics you should use to describe different types of data and, when
you have calculated the statistics using a spreadsheet or statistical analysis software, what each one actually means. The statistics we will look at are summarised
in Table 7.2.
Let’s start by looking at the statistics you can use to describe a data variable. When
you do this, you’re describing one of three things:
The central tendency, which is the common, middle or average value.
The dispersion, which is how the data values are spread, or dispersed, around the
central tendency.
The trend, which is how the data values change or move in one direction over
time.
How to describe the central tendency and dispersion
As you can see from Table 7.2, if you have categorical data variables, such as the answer
to your student questionnaire question ‘In which faculty are you studying?’ the only
Sales in £’000
Aptitude test result (out of 100)
Employees’ sales last month by aptitude test result
Source: Employee database 20##
(data relates to 30 employees)
10
0
0 10 20 30 40 50 60 70 80 90 100
20
30
40
50
60
70
80
Figure 7.6 Scatter graph
Definitions
central tendency: the value for a variable that represents the common, middle or average. Statistics
that describe data in this way are called ‘central tendency measures’.
dispersion: the value for a variable that represents how the data are spread or dispersed around the
central tendency. Statistics that describe data in this way are called ‘dispersion measures’.
trend: the movement or change in one direction of the values for a variable over a period of time.
192 Chapter 7 Analysing data
Definition
normal distribution: the symmetrical distribution of data values around the mean for a quantitative
variable forming a bell-shaped curve. In a normal distribution the values of the mean, median and
mode are the same.
Mean sales last month for the organisation’s 30 employees were £46,600. As the mean,
median and mode are virtually the same, this suggests these data are normally distributed.
Consequently the standard deviation of 18.46 indicates that 95 per cent of sales fell within
the range £10,318 to £82,682, the complete range being £68,000.
statistic you can use to describe the central tendency is the mode. This describes the
category that occurs most often, for this question the Faculty of Business and Law. You
can’t describe the dispersion or the trend for categorical data.
For numerical data, you can describe the central tendency, the dispersion and, where
the data have been collected over time, the trend. The mean and the median are used normally to describe the central tendency for quantitative data. However, you might use the
mode to describe the most common value. Let’s use statistics to describe the variable
‘employee sales last month’ which we used for the scatter graph in Figure 7.6. Table 7.2 tells
you that you can calculate the mode, median, mean, range, inter-quartile range and standard deviation for this numerical data variable. You decide to do this using IBM SPSS Statistics
and get the output shown in Figure 7.7. Although your output does not give you the interquartile range, you could easily work it out by subtracting the 25th percentile from the
75th percentile. You use the output to describe employee sales in your project report:
Table 7.2 Describing a variable using statistics
For . . . use the . . . to describe the . . . which represents the . . .
categorical data mode central tendency category that occurs most often
numerical data mode
or the value that occurs most often
median
or
the
middle value when all the data values are
put in rank order
mean average value
numerical data range
or the dispersion difference between the highest and lowest
data values when they are put in rank order
inter-quartile
range
or the
difference within the middle 50% of data
values when they are put in rank order
standard
deviation
1
extent the data values differ from the
mean (for normally distributed data, 95%
of values will lie within plus or minus 1.96
standard deviations of the mean)
numerical data index
number
2
trend relative change in a series of data values
over time
1 only use if the data are normally distributed.
2 only use if the data have been collected over time.
7.3 Analysing data quantitatively 193
How to describe the trend
You will have seen in Figure 7.7 that we did not calculate index numbers for the variable
‘Employee’s sales in £’000’. This is because these data were not collected over time and so
can’t be used to show a trend. However, we can do this for the data used to draw the line
graph in Research in practice 7.2. Let’s say Ahmed is now focusing his research on the trend
for the ‘Balance of payments – trade in goods’ since 1981. Ahmed represents the value of
his data for this starting date, known as the base year, on his spreadsheet using the number
100 (Figure 7.8). Increases in subsequent years would be represented by positive index
numbers greater than 100 and decreases by positive index numbers less than 100, providing the balance of payments was in surplus. As you can see in Figure 7.8, this was not the
case for the years 2006 onwards. For these years, the deficit in the balance of payments is
represented by a negative index number. As the deficit is greater than the initial surplus in
Figure 7.7 Descriptive statistics output in IBM SPSS Statistics
Source: SPSS © 2016 International Business Machines Corporation. Reproduce courtesy of IBM.
Figure 7.8 Describing the trend using index numbers
194 Chapter 7 Analysing data
How to examine and assess the significance of interdependences
(relationships) using statistics
We will now look at the statistics you can use to examine and assess ‘relationships’
between two or more data variables when you have selected your data from a larger population using probability sampling (Chapter 6.2). When examining a ‘relationship’
using statistics, you’re calculating a test statistic to look at one of five things:
The independence between two variables.
The relationship between two variables.
The difference between two or more variables.
The explanation of one (dependent) variable by one or more other (independent)
variables.
The prediction of one (dependent) variable by one or more other (independent)
variables.
You’re also assessing how likely the outcome you see in your sample data (the ‘relationship’ represented by your test statistic) is to have occurred by chance to establish
whether you can use this to infer the characteristics of the population from which it
was selected. This means as well as a test statistic, you have to work out both a statistic and the probability (likelihood) of this test statistic or one more extreme occurring by chance alone. The process of assessing the significance of findings is known
as
significance testing, the classical approach to assessing statistical significance
being
hypothesis testing. If this probability is small (usually 0.05 or lower), then you
Definitions
significance testing: the process of assessing statistically how likely it is that the characteristics
observed in a sample have occurred by chance.
hypothesis testing: the classical approach to assessing the statistical significance of findings from a
sample.
hypothesis: a tentative (usually testable) statement about the ‘relationship’ between two or more
variables; often referred to as H
1.
null hypothesis: a statement about the ‘relationship’ between two or more variables that
the researcher hopes to reject, thereby accepting the opposite; often referred to as H
0 or Ha, the
alternative hypothesis.
Using 1981 as the base year it can be seen that the balance of payments for trade in goods
went into deficit increasing rapidly up to 2008, the index number of –3182.38 in
2008 indicating the deficit for this year was over 31 times the surplus in 1981. The index
number of –2901.00 in 2009 indicates a decline in the deficit from the previous year (index
number -3182.38). However, subsequent to 2011 the trade gap in goods has grown year on
year, particularly between 2011 and 2012.
the base year (1981), the negative index numbers are greater than 100. Concentrating
mainly on the last 10 years, Ahmed interprets the index numbers in his project report:

7.3 Analysing data quantitatively 195
have a relationship that is statistically significant. Statisticians refer to this as ‘rejecting the null hypothesis and accepting the hypothesis’. When you reject a null
hypothesis, you’re rejecting a testable statement such as ‘There is no relationship
between . . .’ and accepting a hypothesis such as ‘There is a significant relationship
between . . .’
As you can see in Table 7.3, different statistics are used to test each of the five things
we listed above. You will also see that the statistic you use depends on the data you
have, and that there are far more statistics to choose from for numerical data than for
categorical data. This means you need to think carefully about which statistic you use
and why. Table 7.3 should help you in this!
Table 7.3 Examining and assessing the significance of interdependencies (relationships)
between variables using statistics
For . . .
use . . . (symbol
in brackets)
to examine
and assess the
significance of . . .
which, if the null hypothesis is
accepted, represents the . . .
categorical data chi-square
test (
χ2)
independence association between two
variables and the probability of
this or one more extreme
occurring by chance
categorical ordinal
(ranked) data
Spearman’s rank
correlation
coefficient (
ρ) or
correlation strength of the relationship
between two variables and
the probability of this or one
more extreme occurring by
chance
Kendal’s rank
correlation
coefficient (
τ)
correlation (when
data contains tied
ranks)
strength of the relationship
between two variables and
the probability of this or one
more extreme occurring by
chance
numerical data split
into two groups using
a categorical variable
independent
groups
t-test (t)
difference size of the differences between
two groups relative to the
variation in the sample and
the probability of this or one
more extreme occurring by
chance
numerical data split
into three-plus groups
using a categorical
variable
Analysis of
variance
(ANOVA) (
F)
difference size of the variation in the
sample means relative to the
variation in the sample and the
probability of this or one more
extreme occurring by chance
pairs of numerical data
for two variables
measuring the same
feature under different
conditions
paired
t-test (t) difference size of the differences between
the two variables relative to the
variation in the sample and
the probability of this or one
more extreme occurring by
chance
(
Continued)
196 Chapter 7 Analysing data
How to examine and assess the significance of independence
(association)
Let’s say you, in your questionnaire to a sample of students, collected data on each student’s gender (male or female) as well as the faculty in which she or he was studying.
These are both categorical data variables. You look at Table 7.3 and see that, as these
data are not ranked, the only statistic you can use to examine and assess the association
between gender and faculty is the chi-square test. You use this statistic to test the null
hypothesis ‘The faculty in which students are studying and their gender are independent’ using IBM SPSS Statistics. Your output is shown in Figure 7.9.
Your output gives you both a crosstabulation of the data and another table headed
‘Chi-Square Tests’. The first row of the ‘Chi-Square Tests’ table contains the (Pearson)
chi-square statistic (11.716) and the probability (Asymptotic Significance) of this or one
more extreme occurring by chance (.008). Next to the chi-square statistic is a reference
to a footnote which states ‘0 cells (0.0%) have expected counts of less than 5 . . .’ This is
important, as for the statistic to give usable results, no more than 25% of cells can have
expected values of less than five. Fortunately, this is not the case and, since the probability of this chi-square statistic or one more extreme occurring by chance is less than
0.05, you can reject your null hypothesis, ‘The faculty in which students are studying
and their gender are independent’, and accept the hypothesis ‘There is a significant
association between the faculty in which students are studying and their gender’. You
use the table and the chi-square statistic to interpret the relationship between gender
and faculty in your project report:
For . . .
use . . . (symbol
in brackets)
to examine
and assess the
significance of . . .
which, if the null hypothesis is
accepted, represents the . . .
numerical data Pearson’s
product moment
correlation
coefficient (
r)
correlation strength of the relationship
between two variables and
the probability of this or one
more extreme occurring by
chance
Coefficient of
determination
(
r2)
explanation strength of a cause-and-effect
relationship between a
dependent and one or more
independent variables and
the probability of this or one
more extreme occurring by
chance
Regression
equation
(
y = a + b x)
prediction formula to predict the
values of a dependent
variable, given the values of
one or more independent
variables
Table 7.3 Continued
7.3 Analysing data quantitatively 197
How to examine and assess the significance of interdependencies
(relationships)
The only other categorical data statistics we highlight in Table 7.3 are for ranked
(ordinal) data. You will see in this table that there are two possible statistics that can
be used to measure the strength of the relationship between two ranked variables:
Kendall’s rank correlation coefficient;
Spearman’s rank correlation coefficient.
Both these statistics assume your sample has been selected at random, but they do not
need your data to be normally distributed. If your data for a variable contains tied ranks,
let’s say two or more students came equal second in an examination, it is better to use
Kendall’s rank correlation coefficient. Otherwise, you should use Spearman’s rank correlation coefficient.
Like all correlation coefficients, the value of both Spearman’s and Kendall’s rank correlation coefficients will always be somewhere between -1 and +1 (Figure 7.10). This
Figure 7.9 Examining and assessing the significance of association between variables
using a chi-square test
Source: SPSS © 2016 International Business Machines Corporation. Reproduce courtesy of IBM.
There is a significant association between the faculty in which students are studying and their
gender (chi-square = 11.716,
df = 3, p = 0.008). Students in both the Arts and Humanities
and Social Sciences faculties are more likely to be female, while those in the Physical Sciences
faculty are more likely to be male. Students in the Business and Law faculty are more evenly
divided between genders.

198 Chapter 7 Analysing data
0
Negative correlation Independent Positive correlation
Perfect Strong Moderate Weak None None Moderate Strong Perfect
Very
Strong
Very
Strong Weak
–1 –0.8 –0.6 –0.35 –0.2 0.2 0.35 0.6 0.8 1
Figure 7.10 How to interpret correlation coefficients
Source: Developed from Hair et al. (2009).
means that if your value is outside this range, an error has been made in the calculation!
If your correlation coefficient has a value of +1, this means that your two variables are
related perfectly, the positive value meaning that as one variable increases, so does the
other. If your correlation coefficient has a value of -1, this also means your two variables
are related perfectly. However, a negative value means that as the values of one variable
increase, the values of the other variable decrease. Beware: it is extremely unusual to
find two variables that are related perfectly. As you can see in Figure 7.10, a correlation of
zero means your two variables are independent, not being related at all. You have probably worked out by now how to interpret other values of a correlation coefficient from
Figure 7.10. However, to allow you to check whether or not you have got it right, let’s
look at two different values: a correlation coefficient of -0.23 represents a weak negative
correlation; a correlation of coefficient of 0.7 represents a strong positive correlation.
When you interpret a correlation coefficient, you also need to see how likely the relationship between these variables in your sample is to have occurred by chance. As with
the chi-square test, this is called significance testing and represents the probability
(likelihood) of your statistic or one more extreme occurring by chance. Once again, if
this probability is small (usually 0.05 or less), then you can say your relationship is statistically significant. Again, this is referred to as ‘rejecting the null hypothesis and
accepting the hypothesis’. When you reject a null hypothesis for a correlation coefficient, you’re rejecting a statement such as ‘There is no correlation between . . .’ and
accepting a hypothesis such as ‘There is a significant correlation between . . .’
If you haven’t done so already, now read Research in practice 7.3. This looks at the
use of statistics in a journal article a student is reading for her extended essay. Towards
the end of this Research in practice, the student, Samantha, looks at how the author of
the article has used Pearson’s product moment correlation coefficient. As the data used
to calculate Pearson’s product moment correlation coefficient are numerical, she feels
rightly that the correct statistic has been used. For each potential relationship between
brand consciousness and the six socialization factors, the article’s author has given
both the statistic (
r) and the probability (p) of that relationship or one more extreme
occurring by chance. Where this probability is less than or equal to 0.05, the author has
implicitly rejected the null hypothesis ‘There is no relationship between . . .’ and
accepted the hypothesis ‘There is a significant relationship between . . .’ This practice of
not stating either the null hypothesis or hypothesis often happens in academic journal
articles, so don’t worry too much if they are not actually stated in an article you’re reading. Although the article’s author has not provided an indication of how to interpret
the strength of the significant correlations, Samantha recognises that they are not particularly strong. If you look again at Research in practice 7.3, you will see that the

7.3 Analysing data quantitatively 199
Using t-tests, analysis of variance and correlation
Samantha was writing an extended essay on brand consciousness. Within her essay, she
was particularly interested in the factors that influenced people’s awareness of different
brands and how this differed according to their demographic characteristics. During her
search of the literature, she had found a 2011 article by Zaharah Ghazali in the
International Journal of Management and Marketing Research. This article used a sample of 230 students at a Malaysian university to investigate how socialisation factors
such as the media, parents and peers all influenced students’ brand consciousness for
clothing. Within the article, Ghazali also looked at how brand consciousness differed
according to the students’ demographic characteristics.
Ghazali’s article outlined how the data for each of the variables had been collected
using a questionnaire, which Samantha considered had an excellent (88%) response rate.
The article also stated that the questions used to collect the data were all derived from
previous questionnaires, although Samantha noted the source for some of the questions
was not always referenced clearly. Variables used in the analysis included the following:
Gender – categorised as male or female.
Ethnicity – categorised as Malay, Chinese, Indian or other.
Socialisation factors:
– exposure to television media – measured on a numerical scale;
– exposure to radio media – measured on a numerical scale;
– exposure to online media – measured on a numerical scale;
– exposure to movie media – measured on a numerical scale;
– exposure to parents’ influences – measured on a numerical scale;
– exposure to peers’ influences – measured on a numerical scale;
Brand consciousness – measured on a numerical scale.
Ghazali had undertaken a variety of statistical analyses using these variables, such as:
Independent sample t-tests for gender differences in brand consciousness.
One-way analysis of variance for ethnicity differences in brand consciousness.
Pearson’s correlations between exposure and each of the six socialisation factors with
brand consciousness.
Based on the first of these, Ghazali (2011) commented in her article that there was a
significant difference in brand consciousness between genders stating the
t-test value
and its significance (
t = -2.495, p = 0.013). This Ghazali interpreted as indicating that
females were more likely to be highly brand-conscious than males.
Samantha felt that the
t-test was the correct statistic to use. She looked at the table in
the article that showed the results of the
t-test and noted the mean brand consciousness
score for males (2.98) and females (3.24). This provided further support for Ghazali’s
statement that females were more likely to be highly brand-conscious than males.
Interpreting the one-way analysis of variance statistic (
F = 8.732, p ≤ 0.001), Ghazali
(2011) commented there was a significant difference in brand consciousness between
Research in practice 7.3
200 Chapter 7 Analysing data
negative correlation between exposure to movies and brand consciousness (r = -0.369,
p < 0.01) and the positive correlation between exposure to peer influence and brand
consciousness (
r = 0.486, p < 0.01), although both significant, might only be interpreted
as a moderate relationship. In contrast, the relationship represented by the positive
correlation between exposure to parental influence and brand consciousness (
r = 0.134,
p < 0.05), although significant, might be interpreted as ‘none’, owing to the very low
value of the correlation coefficient. None of the remaining correlations with brand
consciousness are assessed as significant, the value of
p for each being greater than 0.05.
How to examine and assess the significance of differences
Research in practice 7.3 also shows how two statistics, the independent groups’ t-test
and the analysis of variance, can be used to analyse differences when a numerical variable is split into two or more groups. In the article Samantha is reading, the author split
the numerical variable ‘brand consciousness’ into two groups using the categorical variable ‘gender’, and then uses the independent groups’
t-test statistic and associated p
value to assess whether the difference between males and females in her sample is
ethnic groups. This she interpreted as students from her category of ‘other ethnic
groups’ (including those of Siamese, Singaporean and Indonesian origin) being more
likely to be highly brand-conscious than students of Malay, Chinese and Indian origin.
Samantha felt that analysis of variance was the correct statistic to use. She looked at
the table in the article showing the results of the analysis of variance. This gave the
mean brand consciousness score for ‘other’ (3.95), as well as the Malay (3.09), Chinese
(3.10) and Indian (3.13) ethnic groups. This supported Ghazali’s interpretation that students in the other ethnic groups were more likely to be highly brand-conscious than
Malay, Chinese and Indian students.
Finally, using Pearson’s product moment correlation coefficient, Ghazali (2011) discussed the relationship between brand consciousness and socialisation factors such as
media socialisation, parents and peers. She noted the various factors differed in their
relationships: movie-viewing having a negative relationship with brand consciousness
(
r = -0.369, p < 0.01), while parental influence (r = 0.134, p < 0.05) and peer influence
(
r = 0.486, p < 0.01) both showed positive relationships.
Samantha considered Ghazali’s discussion of the findings in relation to the correlation
statistics presented in the associated table in the article. She felt that the use of Pearson’s
correlation was appropriate for numerical data. Although these statistics had been interpreted correctly, she felt that none of the correlations were particularly strong. She
noted that, while there were correlations between brand consciousness and exposure to
each of movie media, parents’ influences and peers’ influences were statistically significant, the correlations with exposure to television, radio and online media were not.
Although this point was made by Ghazali in her article’s discussion section, Samantha
decided to note down the remaining correlation statistics with brand consciousness and
their significance levels for use in her extended essay. These were exposure to television
media (
r = -0.081, p > 0.05), exposure to radio media (r = -0.020, p > 0.05), and exposure to online media (r = 0.067, p > 0.05).
7.3 Analysing data quantitatively 201
significant. The independent groups’ t-test was used because the author was examining
differences between two groups. It is worth noting here that a paired
t-test was not used
because the data in the two variables were not collected in pairs, one from each group.
Although a paired
t-test is calculated in a slightly different way, its interpretation is
exactly the same. As the
p value of 0.013 was less than 0.05, the probability of this test
statistic or one more extreme occurring by chance alone in the population was considered significant. A null hypothesis such as ‘There is no difference in brand values
between males and females’ could be rejected, and a hypothesis such as ‘There is a significant difference in brand values between males and females’ could be accepted. Once
again, neither the null hypothesis nor the hypothesis was stated in the article.
The author of the article also split the numeric variable ‘brand consciousness’ into
four groups, using the categorical variable ‘ethnicity’, and then used the analysis of variance statistic to see if the difference in brand consciousness between Malay, Chinese,
Indian and other ethnic groups was significant. Looking at Table 7.3, you can see that
analysis of variance was used because brand consciousness was now split into more
than two groups. The probability of the recorded
F ratio (F = 8.732) or one more extreme
occurring by chance in the population was less than 0.001 (
p ≤ 0.001). This means a null
hypothesis such as ‘There is no difference in brand values between ethnic groups’ could
be rejected, and a hypothesis such as ‘There is a significant difference in brand values
between ethnic groups’ could be accepted. As previously, the null hypothesis and
hypothesis were not stated in the article.
How to explain and predict cause-and-effect relationships
You’ve already looked at a cause-and-effect relationship using a scatter graph to display
the relationship between all sales employees’ aptitude test results and their sales performance for the past month (Figure 7.6). When you plotted these data on your scatter
graph, you plotted aptitude test result as the independent variable and sales performance in £’000 as the dependent variable. This scatter graph suggested there was a
cause-and-effect relationship between employees’ aptitude (the cause) and their sales
performance (the effect). As we stated you would, you now test this statistically by calculating the coefficient of determination (Figure 7.11). As your data relate to all sales
employees, the population, there is no need to calculate the likelihood (probability)
that this coefficient of determination has occurred by chance.
Figure 7.11 Coefficient of determination and regression equation
202 Chapter 7 Analysing data
At the start of this chapter, we said that we would focus on analysing text data qualitatively,
only talking about the analysis of non-text data such as audio recordings and the audio
parts of video recordings where these have been transcribed and so could also be analysed
as text data. If you’re collecting your own data, perhaps by interviewing, you will probably
have already begun to think about what these interviews are telling you. This is quite normal. We would expect you to begin to analyse your qualitative data before you’ve collected
them all. Doing this will allow you to follow up initial insights suggested by early interviews
in later interviews, as well as to recognise when you have reached data saturation (we talked
about this in section 6.3). In this section, we will start by looking at how you prepare text
data so that they are suitable for qualitative analysis. As you do this, you will begin to
immerse yourself in your data and so begin to understand it better. Next, we will talk about
how you develop propositions and then how you use these propositions to build or test
theory by looking for patterns in your data. Finally, we will look at how you assess the
credibility and dependability of qualitative research findings reported by others.
How to prepare your data
If you’re analysing your data qualitatively, you may or may not be using specialist
qualitative data analysis software. Although in some universities computer-aided
qualitative data analysis software (
CAQDAS), such as QSR International’s NVivo, is
7.4 Analysing data qualitatively
Definition
CAQDAS: a general term for all computer-aided qualitative data analysis software such as NVivo,
ATLAS.ti and MAXqda.
Your coefficient of determination of 0.83 means that 83% of the variation in your
dependent variable, sales in the past month, can be explained by the independent variable, aptitude test result. This means that the aptitude test is a very good predictor of
sales performance for the past month as it’s rare to obtain a coefficient of determination
larger than 0.8. Note: the value of a coefficient of determination must always be between
zero and one. If it is outside this range, then there has been an error in the calculation!
The regression equation you’ve calculated in Figure 7.11 can be used to predict the
value of your dependent variable, sales for the past month, if you know the value of the
independent variable, aptitude test. This means, for example, if a potential employee
scores 70 in an aptitude test taken as part of the selection process, you can use the regression equation to predict what their sales in £’000 would have been for the past month:
Sales in £’000 = –34.58 + (1.11 × 70)
= –34.58 + 77.7
= 43.12
You could therefore predict that this potential employee would have made sales of
£43,120 for the past month.

7.4 Analysing data qualitatively 203
used by students for undergraduate and master’s research projects, this is not always
the case. In addition, some researchers prefer to analyse qualitative data manually.
As a consequence, although you will be expected to use a spreadsheet or statistical
analysis software to analyse quantitative data, you’re less likely to be expected to use
CAQDAS to analyse qualitative data. However, the things you need to do to prepare
your text data for qualitative analysis are similar whether or not you intend to use
CAQDAS.
Research in practice 7.4 examines an annotated extract from a student’s interview transcripts, showing you how to prepare an audio recording as text for qualitative analysis.
Annotated interview transcript
As part of her research project on employees’ reactions to change, Xue had conducted
and audio-recorded semi-structured interviews with 12 participants, each of whom
signed a consent form. She found the transcription of the interviews very time consuming
as, although each only lasted on average half an hour, word-processing each audio
recording took between three and four hours. She emailed her first transcript to her
supervisor and asked him to see if there were any ways it could be improved. The extract
below from the start of the interview includes her supervisor’s comments relating to preparing the transcript for analysis. It does not include any of the supervisor’s comments
about Xue’s interviewing skills.
Research in practice 7.4
[organisation name] Interview 2
XUE: Thank you for agreeing to be interviewed. Now, just to
give a little bit of context, we’re going to be talking about you
in relation to your work and the changes at [organisation
name] in the interview. But first, can you give me an idea of
which bit of [organisation name] you work in and what sort of
job you do? (…) Can you give me a quick overview?
PHIL: I work in Financial Services in the payroll section and I
work within the control section of the payroll section running
payrolls and thatsort of thing (…)
XUE:
So what is your job title Phil?
PHIL: Well (…) at the moment it is Project Officer (…) or
something like that, I mean (…).
XUE:
Basically what you are doing is running the payrolls?
PHIL: Well we’ve gone on to a new payroll system in the last
twelve months and are also involved in setting up (……) I WAS
involved in the project in setting it up (…) and NOW I have just
sort of moved back into the payroll section, and running
payrolls, dealing with queries, dealing changes to the system,
that sort of thing at the moment ((appears to be unhappy
about this)). There is a restructuring going on in the section,
so the job title and what I do could, again, be changed into
something else.
Include more
details such as
date, time and
place of interview
I presume names
have been
changed to
preserve
anonymity
Use of italics
makes questions
easy to see
I presume more
dots inside the
single brackets
means a longer
pause
I presume the
double bracket
means this is
your description
inside the brackets
Use of capitals
makes names
easy to see
I presume capitals
means this word
was spoken more
loudly than the
others
I presume dots
inside the single
brackets mean
a pause
DON’T FORGET
TO SAVE EACH
INTERVIEW AS A
SEPARATE FILE
I know there are
no typographical
errors here, but
check there are
none in all your
transcripts

204 Chapter 7 Analysing data
As you can see from how Xue has prepared her transcript, when preparing data as text
for qualitative analysis, you need to do the following:
Include details of the date, time and place where the data were collected.
Anonymise both the organisation’s and the respondents’ names, using the alternatives
consistently.
Consistently use italics to signify questions asked.
Consistently use capitals to highlight the names of the interviewer and the respondents.
Consistently use (. . .) to show a pause in speech, the number of dots showing the
relative length of the pause.
Consistently use CAPITALS within the transcript to show those words that were
spoken more loudly than others.
Consistently use (( )) to enclose your description of what is happening such as the
participant’s tone of voice, facial expressions or other visual cues.
Make sure there are no typographical errors and that words are spelt consistently
throughout.
Save each interview transcript as a separate file.
Having just read this list, you will see that being consistent when preparing qualitative
text data such as transcripts for analysis is crucial. If you’re going to analyse web-based
interview data where questions and responses were automatically captured as they were
typed in, it is unlikely that these data will have been recorded consistently or in precisely
the right format for CAQDAS. If you’re going to analyse other forms of text data, including secondary data such as online business news reports, these data are also unlikely to
be in quite the right format for CAQDAS. Although you will not need to anonymise secondary text data that are available publicly such as news reports, you will still need to
add full details of when it was collected and the source where you found it. It is also likely
that such data will contain some typographical errors and inconsistencies that you will
need to correct. You will therefore need to spend time preparing your text data for qualitative analysis by following the points outlined above, whatever its source.
You only need to read this paragraph if you’re going to analyse your data using
CAQDAS! If you are, you need to find out if the software you are thinking of using will
enable you to do the analysis you are thinking of doing. Fortunately, most CAQDAS
software websites, for example QSR International’s NVivo 11, provides considerable
detail on what the software can do. Many also offer you the option to download a free
trial version. Do take this opportunity, especially if the software is not available to
download from your university as part of their site license.
How to develop propositions to build or test theory
We are sure that sometime in your life you have at least had a go at completing a jigsaw puzzle. When you started to make the puzzle, you probably looked first at the picture on the
box lid. Let’s say it was a photograph of the Sydney Opera House in Australia, although
unlike Figure 7.12, it would be in colour. You then tipped all the pieces out of the box onto
a table and turned them picture side up before sorting them into different groups or

7.4 Analysing data qualitatively 205
categories that had similar characteristics. To help you with your sorting, you looked at the
picture to get some ideas for categories. These would have included categories such as part
of the ‘Opera House’, the ‘sky’, the ‘sea’, ‘vegetation’ as well as the ‘edge of the puzzle’.
Inevitably, you could have put some of the puzzle pieces into more than one category, such
as being at the ‘edge of the puzzle’ and also part of the ‘sky’, or being part of the ‘sea’ and
part of the ‘Opera House’. You then fitted pieces in each category together, completing different parts of the puzzle such as the sky, the river or the Opera House. You then used those
pieces that were in more than one of your categories such as the ‘sky’ and ‘Opera House’ to
join together the different parts of the puzzle you had already completed, before fitting the
remaining pieces. Finally, you compared the puzzle picture with that on the box lid.
You’re no doubt wondering why we’ve spent so much time talking about completing
a jigsaw puzzle. The answer is simple! Analysing qualitative data and, in particular, looking for patterns by categorising data is similar to completing a jigsaw puzzle. Let’s start
with the lid of the puzzle box. In addition to the picture of the completed puzzle, it normally says how many pieces there are in the puzzle and how difficult it is to complete.
This is similar to your critical review of the literature (section 2.2), as the lid provides an
overview of significant information about the puzzle and a clear idea of how the pieces
should fit together. The information on the box lid therefore represents the existing
knowledge about how the puzzle pieces relate to each other. From this existing knowledge you develop
testable propositions. For example, looking at the picture on the box lid
Figure 7.12 Sydney Opera House
Source: © Mark Saunders, 2015.
Definition
testable proposition: an assertion that something is true that can be tested using the data
available.

206 Chapter 7 Analysing data
(Figure 7.12), a testable proposition would be that the pieces of sky are at the top of the
picture. Another would be that the sea is at the bottom of the picture. Pieces that can be
put in two or more categories show the interrelationships within the theory, such as
the relationship between the Opera House and the sky. You test these propositions (and
others) by categorising the pieces and making the puzzle. If the picture on the box is
the same as the puzzle picture you have made, then the propositions you have developed are supported!
What you have done in using the picture and other information on the puzzle box
lid is analyse your data (the puzzle pieces) deductively (we talked about both deductive
and inductive approaches in more detail in section 5.3). This is because you started
with what was already known about the puzzle (the picture and other information on
the box) and used this as a guide to develop your series of testable propositions. You
then designed your analysis (categorised your pieces and made the puzzle) to test these
propositions. If you’re analysing qualitative data deductively for your project report,
you’re looking for things in the data to answer research questions and test the propositions that you have developed using your understanding of the academic literature.
Like the picture on the puzzle box lid, the literature on which your research questions
and propositions are based give a clear direction of the categories that are important
and, where a piece falls into two or more categories, how they should fit together. By
systematically categorising your data and seeing if particular patterns occur, you’re testing your propositions. Remember, as we pointed out in section 6.4, as you collect and
analyse your data, you may get new insights into your testable propositions that you
did not initially expect and so may need to revise them.
If your puzzle had no picture on the box lid or any other information, then you
would have had to analyse your data inductively. You would have started without a clear
theory from which to develop testable propositions of what the puzzle looked like, the
number of pieces and how they were related. You would have done this by sorting the
pieces into different categories that had similar characteristics. However, as there was
no information on the box, you would have had to use ideas you generated by looking
at (analysing) the puzzle pieces to decide on your categories. You would have found that
some of the categories you first thought of were less useful than others and may well
have decided later not to use them. If you’re analysing qualitative data inductively for
your project report, you do not have (or need) a clearly defined theoretical framework.
Rather, you’re identifying possible propositions (theories) from your data and then
developing questions to test them. Like a puzzle about which you have no information,
your theory about how the pieces fit together will develop as you systematically analyse
your data and collect more data, and the most relevant patterns become clearer.
Look at the first two paragraphs of Research in practice 7.5. In his research, Mark
developed three testable propositions from the literature. Each proposition was carefully worded, often as a possible explanation, to enable it to be tested. Reread Mark’s
first proposition:
1. Students adopt particular habits when they write.
7.4 Analysing data qualitatively 207
Research in practice 7.5
Developing testable propositions and categorising data
Mark was working with a group of students who, like many, found writing their project
report difficult. During his preparation, he used a number of academic sources on writing, including the book
Writing for Social Scientists (Becker, 2007). Using this, he developed three testable propositions:
1 Students adopt particular habits when they write.
2 Students adopt these habits because they:
(a) find writing difficult,
(b) are easily distracted from writing,
(c) are afraid that what they write will be wrong.
3 The habits students adopt differ between male and female students.
He decided to test these propositions and asked each of the students to describe in writing how he or she wrote. He explained he was interested not in the students’ scholarly
preparations, but in the details of what they actually did when they were writing. The
students’ word-processed descriptions were subsequently anonymised and formed
Mark’s data. Mark also developed an initial set of categories to attach to this data:

Habit Unsure what these could be. Develop categories from the data.
Reason – difficult
– distraction

– afraid wrong

Gender – male
– female

The following extract of a student’s description shows these categories attached to units
of data, each of which is one sentence of the description:

Gender male Male student
Habit I always compose directly into a word processor. I usually have

hand-written notes to work from which give me an overall
structure, but the thoughts are typed directly in. Before starting,
I make sure that my desk is relatively clear, other than for the
notes and articles I’m going to use. I also make sure I have a
fresh cup of tea. The computer is in a room which I always use
for writing and I feel comfortable there. To be honest, I find it
difficult to write anywhere other than this room. The word
processor software is set to my preferences: Times New Roman
12, and no auto spell corrections. I’ve turned off the underline
for misspelled words because it puts me off. I don’t have any
distractions in the room where I write – no music or television,
although there is a telephone. I nearly always start early in the
morning (7.30) and work through until about 10.30–11.00
before I have a break. I like to have full days for writing as it
helps me organise my thoughts . . .
Habit
Habit
Habit
reasoncomfort/difficult
Habit
Habit
reasondistraction
reasondistraction
Habit
Habit
208 Chapter 7 Analysing data
This suggests three possible explanations for students adopting particular habits when
they write: (a) they find writing difficult, (b) they are easily distracted from writing and
(c) they are afraid that what they write will be wrong. The data Mark collects in the form of
students’ descriptions of how they write will enable each possible explanation to be tested.
When testing these explanations, it will be important that Mark also looks for alternative
explanations, such as other reasons why students might adopt these habits. It may be that
some students do not even adopt particular habits when they write! If you do not look for
alternative explanations, you will probably only notice evidence that supports your own
opinions, resulting in your findings being biased and subjective. By testing your own
propositions and looking for alternative explanations through looking for clear patterns,
you will be able to develop clearly justified, credible conclusions from your data.
Finally, look at Mark’s third proposition:
2. Students adopt these habits because they:
(a) find writing difficult,
(b) are easily distracted from writing,
(c) are afraid that what they write will be wrong.
Now look at Mark’s second proposition:
3. The habits students adopt differ between male and female students.
Like Mark’s first proposition, this can be tested using the data Mark has collected. As
with his other propositions, he will test this by looking for patterns in the data.
How to build or test theory by looking for patterns
Our earlier jigsaw puzzle example emphasises that, whether you’re using a deductive or
an inductive approach to look for patterns to build or test your propositions, you need
to take the following steps:
1 Develop meaningful categories or codes to describe your data.
2 Decide on the unit of data that is appropriate for your analysis and to which you will
attach relevant categories.
3 Attach relevant categories to units (pieces) of your data.
Definitions
categorising data: developing meaningful categories and attaching those categories that are relevant
to specific units of data.
unit of data: a predetermined piece of data such as a line of a transcript, sentence, paragraph or response.
Although this proposition does not offer a possible explanation, whether or not the
students adopt particular habits when they write, it can be tested by looking for patterns in the data Mark has collected.

7.4 Analysing data qualitatively 209
If you’re using a deductive approach, your categories will be based on terms used
in the literature, often being drawn from existing theory. In contrast, if you’re using
an inductive approach, your categories will emerge from your data. For both deductive and inductive approaches, it is likely that you will refine your categories to
ensure they are meaningful as you look for patterns during your analysis to test your
propositions. Your choice of the unit of data to which you will attach a relevant category or categories depends on what works! You need units of data that are large
enough to highlight where there are relationships between categories, and yet not
so large as to be described by all your categories. Depending on the analysis, we’ve
found units of a line of transcript; a sentence and an individual response can all
work well.
Let’s now look at the categorising of the student’s description in Research in
practice 7.5. In this research, Mark created hierarchical categories to test three propositions he developed from the literature. However, other than the source of categories,
the process he has used is the same for both deductive and inductive approaches. For
his first category, ‘habit’, Mark was unsure what these might be and so decided to
develop more detailed sub-categories from the data. This means that for these detailed
sub-categories, he will be working inductively. Although the extract shows only part
of one student’s description, it implies, as suggested in proposition one, students do
have habits associated with how they write. Possible sub-categories of ‘habit’ that
Mark could use to group his data suggested by the extract include ‘computer use’,
‘refreshment’ and ‘room’. If these sub-categories also proved suitable for categorising
other students’ descriptions, they would reveal that the habits were common. It is
likely that further new sub-categories would be revealed by the other students’ descriptions of how they write. For his second category, ‘reason’, Mark worked deductively
using the literature to provide three sub-categories: ‘difficult’, ‘distraction’ and ‘afraid
wrong’. The first two of these have been used when categorising units of data in the
extract. In addition Mark has introduced a new sub-category: ‘reason – comfort’ based
on the data. This indicates he is looking for alternative explanations. Mark’s use of
the category ‘gender – male’ provides descriptive data about the student and will
allow him to test his third proposition by looking for differences in patterns between
male and female students.
We’ll now look at how categorising these data has already begun to suggest a
possible pattern with which to test the second proposition. Proposition two offered
three reasons why students adopted these habits associated with writing. If we look
at the student’s description, we can see that the ‘reason’ sub-categories ‘difficulty’
and ‘distraction’ are both in close proximity to the category ‘habit’. This provided
Mark with an indication that there may be a pattern. Looking at the actual data,
you can see that the ‘reason’ for the ‘habit’ of using a particular room is because it
is ‘difficult’ to write anywhere else and, towards the end of the extract, because
there are no ‘distractions’. Obviously from such a short extract, Mark can’t be sure
that this is a pattern. Analysis of the other students’ descriptions of how they write
will enable him to test the proposition further and confirm whether a similar pattern occurs.

210 Chapter 7 Analysing data
How to assess the value of qualitative research reported by others
You’re probably now thinking: this is all very well but, when I read articles reporting
qualitative research, how do I know that patterns that are discussed and the insights
offered are of value? Your question is a good one! As with you, the researcher’s biases,
preferences and so on will influence at least to some extent the way the data have been
collected, analysed and interpreted. This means that when you read an article you need
to carefully assess whether the author’s claims are credible and dependable, being justified by methods used and supporting evidence that has been offered. To do this, we
suggest you use the following questions:
1 Is the method used to collect the data explained clearly (section 6.4)?
2 Is the method used to analyse the data explained clearly (section 7.4)?
3 Are the findings explained thematically?
4 Are the findings related explicitly to research objectives, questions or propositions?
5 Do the arguments and the evidence presented support the claims made? For
example:
(a) Are the quotations used placed in context, and do they support points made?
(b) Are data included that do not support the claims made and an explanation
offered as to why?
Assessing findings from interviews
Simone’s research project was an extended literature review about how ethnic minority
small businesses (EMSBs) managed their customer relationships. In her search of the
Business Source Premier database, she had found an article by Altinay, Saunders and
Wang (2014) that explored the influence of culture on trust judgements in the development of customer relationships by EMSBs. Altinay and colleagues had analysed data
from 134 face-to-face interviews with Turkish entrepreneurs working in London. Each
interview covered a wide range of issues including relationship marketing practices and
how relationships with both ethnic and British mainstream customer groups were managed, and the entrepreneur’s understanding of trust and its importance in customer
relationship development.
Simone felt that the article outlined the data collection and analysis clearly, providing
sufficient detail for her to see precisely how both had been undertaken. This gave her
confidence in the findings. The authors had organised their findings to address their
research objectives on the influence of culture thematically using three dimensions of
trust judgement outlined widely in the literature (benevolence, honesty and competence) to provide a clear structure.
As part of the findings about how ethnic minority small businesses saw the link
between acting benevolently and managing customer relationships Altinay et al. (2014:
67) had written:
Research in practice 7.6
7.4 Analysing data qualitatively 211
Simone felt that this extract, along with the rest of the findings section on the link
between the three components of trust and customer relationship development, provided credible evidence of the role of trust in the development of customer relationships
by ethnic minority small businesses. She also recognised that, although the quotations
used in the article provided credible supporting evidence regarding how the ethnic
minority small businesses EMSBs interviewed had operationalised trusting behaviours,
there was no need to include them in her extended literature review. Her supervisor
agreed and suggested she just refer to the key findings, where appropriate, when discussing particular themes in her review.
They considered relationship development as a process of social exchange, something
they saw as a weakness of their large counterparts. The vast majority of EMSB owners
explained how they tried to ensure that customers felt there was a family atmosphere,
indicating this was the most important element of Turkish hospitality. A retail shop owner
emphasized the importance of being sincere and demonstrating good intention: ‘
Our
biggest strength is that we have a lot to offer from our hospitable culture. When the
customers go to the bigger stores, they do not know the employees very well. Employees
do not talk to them properly. They sometimes do not even know the products well.
Whereas with us, I talk to customers informally, I talk to them about football, traffic,
weather, family and develop an informal relationship
.’ . . .
During our interviews, most participants highlighted the importance of demonstrating
friendship in business transactions. They stated that this mindset inspired them in their
relationship development with customers.
Starting dialogue with customers was perceived to be an effective way of
demonstrating friendship and good intentions. One grocery shop owner explained: ‘
We
are very good at developing customer relationships. We take the initiative and start off
conversation with our customers about anything
.’ An accountant’s comments
summarized the “mentality” dominating the trust development with customers: ‘
We
treat them as “people” not as customers. We value them with their characters and
personalities
.
In line with arguments by Carson and Gilmore (2000) and Altinay and Altinay (2008),
our findings revealed that the value sets of the business owner determine the marketing
orientation of firms. Since relationship development was well embedded in the values of
Turkish EMSB owners who saw this as a competitive advantage over large counterparts,
they equated relationship marketing to their firms’ marketing philosophy . . .”
Source: Altinay et al., (2014: 67–68) Copyright © 2014 John Wiley & Sons. Reproduced by permission of
the publisher
If you haven’t done so already, now read Research in practice 7.6. This looks at a journal article using qualitative research a student is reading for her extended essay. The
student, Simone, feels that the methods used to collect the data and to analyse the data
have been explained clearly in the article (questions 1 and 2 from the list above), and
we’ll believe her! We’re also told that the findings have been organised thematically,
each of the authors’ research objectives being considered (questions 3 and 4 from the
list above). This suggests the findings are based on a dependable method.

212 Chapter 7 Analysing data
Using the extract from the article, we can now consider whether or not we agree with
Simone’s assessment that the article ‘provides credible evidence’ to support the claims
made. To do this, we’ll use the final question (5) from the list above. Look at the first
quotation in the extract. Although this quotation is more than four lines in length, it
has not been indented. If you’re including quotations in your project report, check
your university’s regulations. Often quotations of five lines or more are indented rather
than just being included within the main text of the paragraph. This first quotation
provides credible evidence from a retail shop owner to support the authors’ claim that
the vast majority of owners tried to ensure their customers felt there was a family atmosphere (question 5a). As you read on, you will see that subsequent quotes from a grocery
shop owner and an accountant offer further insights regarding how this atmosphere
was developed and the importance of friendship.
You will notice that all of the data in this extract does supports the claims being made.
This would suggest that the answer to question 5b, ‘Are data included that do not support
the claims made . . .?’ is ‘No’! You will have to take it from us that such data were included in
this article, illustrating clear differences in how ethnic minority small businesses developed
relationships with different customer groups. In particular, these data highlighted differences in those trust components that were important for relationships with other ethnic
minority groups and those that were important for relationships with mainstream British
nationals (Altinay et al., 2014). Following this discussion, we can say we agree with Simone’s
assessment and that the article ‘provides credible evidence’ to support the claims made.
The data that you and others analyse can be divided into two types: quantitative data
that are numerical or whose values had been measured in some way, and qualitative
data that are not numerical and have not been measured in some way.
Quantitative data can be split into categorical data (nominal and ordinal) and
numerical data (interval and ratio).
Qualitative data can be divided into text data and non-text data (audio, video and
still image).
When preparing quantitative data for computer analysis, you should enter it as a
data matrix. Secondary data can often be downloaded in this format. You should
check data for errors before undertaking analyses.
Tables and diagrams are used to present quantitative data. The tables and diagrams
you use depend on the type of data you have and what you want to show.
Statistical analyses are used to describe data and to examine and assess relationships.
The statistics you use depend on the type of data you have, what you want to describe,
examine or assess and whether you can meet the assumptions of the test. For data
that is a sample from a larger population, you will need to work out the probability of
the finding or one more extreme occurring by chance.
Summary
Thinking about analysing data 213
Qualitative data are often analysed in text form. When preparing qualitative data for
analysis as text, you should transcribe it as a word-processed document, ensuring
you use transcribing conventions consistently. You can either analyse qualitative
data manually or use computer-aided qualitative analysis software.
You can undertake the process of qualitative data analysis both deductively and
inductively.
In both inductive and deductive qualitative analysis, you can use propositions to
develop theory. This involves looking for patterns in your data and testing alternative explanations for these patterns using your data. To look for patterns, you
need to:
develop meaningful categories or codes to describe your data;
decide on the unit of data that is appropriate for your analysis and to which you
will attach relevant categories;
attach relevant categories to units (pieces) of your data.
It is important that, in analysing data, the arguments you offer and the findings you
present support claims you’re making clearly and logically.
As you continue to read and note the literature, use your knowledge about quantitative and qualitative data analysis to help you assess the value of other researchers’
findings (section 2.7).
Think about the analyses you will need to undertake to answer your research question
and meet your objectives.
If you’re going to use or are using secondary data (Chapter 4), use your knowledge
about different types of data and how to prepare data for analysis to help you assess
which data analysis techniques will be most suitable. Make notes about the reasons
for your choices, as you will need this level of detail to write the method section of
your project report.
If you’re going to collect or have collected your own (primary) data, use your knowledge about different types of data and how to prepare data for analysis to help you
assess which data analysis techniques will be most helpful. Make notes about the reasons for your choices, as you will need this level of detail to write the method section
of your project report. If possible, make your assessment of techniques before you
collect your data. You may find you need to change parts of your collection methods,
such as questionnaire questions (Chapter 6).
Also, make notes explaining how the analyses you have chosen will mean you can
answer your objectives. You will need this level of detail to write the method section
of your project report.
Thinking about analysing data
214 Chapter 7 Analysing data
Altinay, L., Saunders, M.N.K. and Wang, C. (2014). The influence of culture on trust judgments
in customer relationship development by ethnic minority small businesses.
Journal of Small
Business Management,
52(1), 59–78.
Becker, H.S. (2007).
Writing for Social Scientists (2nd ed.). Chicago: University of Chicago.
Eurostat Press Office (2016). Organic crop farming on the rise in the EU.
Eurostat News Release
208/2016
. Available at: http://ec.europa.eu/eurostat/documents/2995521/7709498/5-
25102016-BP-EN.pdf/cee89f9e-023b-4470-ba23-61a9893d34c8 [Accessed 21 November
2016].
Ghazali, S. (2011). The influence of socialization agents and demographic profiles on brand
consciousness,
International Journal of Management and Marketing Research, 4(1), 19–29.
Hair, J., Black, W., Babin, B. and Anderson, R. (2009).
Multivariate Data Analysis (7th ed.).
London: Pearson.
Office for National Statistics (2016).
Balance of Payments Time Series Dataset. Available at:
https://www.ons.gov.uk/economy/nationalaccounts/balanceofpayments/datasets/balanceofpayments [Accessed 20 November 2016].
Saunders, M., Lewis, P. and Thornhill, A. (2016).
Research Methods for Business Students (7th ed.).
Harlow: Pearson.
References
Chapter 8
Writing and presenting the research
proposal
When we sat down to plan this text, we thought long and hard about the order in which the
chapters should appear. The most difficult decision was where to place this chapter. It had to
appear at the beginning of the text or at the end. We could have put it at the beginning.
Setting out the recommended content of the research proposal could have acted as a route
map from which you could plot your path through the text. This would have ensured you
were familiar with the role of, say, the literature review and the details you should include
about your proposed methods. We decided against this. We thought that you should be
familiar with the major concepts and techniques that the conduct of research involves before
you think about your proposal. In short, we felt that you would be better equipped to write
an effective research proposal having read and understood the previous seven chapters.
Writing an effective research proposal is a vital part of the research process. We explain
the reasons for this in the first section of this chapter. Perhaps you feel it is unnecessary to
elaborate on this. After all, your university may require you to produce a written research
proposal as the assessment vehicle for the research methods module, so this reason is sufficient. But the research proposal is a vital document for reasons other than assessment
necessity. Indeed, you may be reading this text to help you prepare a research proposal
which will be a vital starting point in your research. For example, you may be embarking
upon a research project required by your work organisation. Whatever the purpose, writing
an effective research proposal is vital.
We then consider at which point in the research process you should be writing your
research proposal before considering the most important topic addressed in the chapter:
the content of the research proposal. The chapter ends with two discussions which we
hope you will find useful. The first is on the writing style you should adopt when compiling
your research proposal. This is something many of us have found difficult, and confusing,
in our academic careers and is something worth thinking about. Finally, we return to the
subject of assessment and ask just what is it that the assessor(s) of your research proposal
will be looking for when that all-important judgment is made.
8.1 Why you should read this chapter
216 Chapter 8 Writing and presenting the research proposal
The overall purpose of the research proposal is to present and justify a research idea you
have and to explain the practical way in which you think this research should be conducted. In effect, it is a ‘contract’ between you and your reader(s) which sets out precisely what it is you aim, or would aim, to do and the way in which you will do it. You
can also think about this contract as a personal document which binds you to your
stated intentions in the event that you are tempted to break the contract! So think
about the research proposal as an essential discipline measure in a process which has
the potential to become extremely undisciplined!
Now let’s look at some of the more specific reasons the research proposal is so
important.
It clarifies your ideas and helps you organise those ideas
Another reason we decided to place this chapter at the end of the book is that you will
realise now that assembling your research ideas is a very difficult process. Moreover,
assembling them in a way that they will make sense to an audience which is far less
familiar than you with those ideas is even more difficult.
We have already made the point in Chapter 2 that writing can be the best way of
clarifying our thoughts. This is a valuable purpose of the research proposal. There is
nothing quite so frightening as sitting in front of a screen with just the heading
‘Research proposal’ to concentrate the mind. The very blankness of the screen forces
you to come up with some ideas. If all else fails, we always suggest to our students to do
what we do ourselves (and, in fact, what we did in planning this chapter): to turn to
Rudyard Kipling! There is a section in one of his 1902
Just So Stories where he wrote:
I KEEP six honest serving-men
(They taught me all I knew);
Their names are What and Why and When
And How and Where and Who.
Source: Kipling (1902, reprinted 2007).
To use Kipling to help you in framing your research proposal, you can build the content
around the answers to:
What are the research questions I am seeking to answer?
What are my research objectives?
Why is the research I propose significant?
When are the key dates in the research process?
How will I go about collecting the necessary data to answer the research questions?
Where will I be conducting my research (e.g. organisation, sector, country)?
Who are the key participants in the research process (e.g. gatekeepers,
respondents)?
8.2 The importance of the research proposal
8.2 The importance of the research proposal 217
Not only will this technique clarify your thoughts but it will help you to organise your
ideas into a clear statement of your research intent. Clarity, in this sense, means organising your ideas in such a way that they are coherent and rigorous. Coherence means
that they make sense, are plausible and can be understood easily: by rigorous, we mean
that the ideas will stand up to searching analysis and can be defended by you in the face
of demands for you to justify their inclusion in your proposal. Your assessor will be
looking for this standard of work in your proposal.
It serves as a route map to guide you through the research process
If you are going on to do the research that you have detailed in your research proposal,
the proposal will be a very useful check for you to ensure that you are doing what you
said you will do. This may sound an obvious point, but few of us have not had our imaginations fired by a great idea during the research process only to be deflated by a supervisor who says ‘yes, that’s a super idea, but look again at your research objectives;
explain to me how this new idea fits’. Too often the response is ‘mmm . . . I see what you
mean’ when it is apparent that the new idea does not fit.
We have found that some students benefit from putting up on their study wall a copy
of their research question(s) and objectives in large, bold type so that they can revisit
them frequently to ensure they are on the right track.
Defining research questions
Freya was a final-year undergraduate who had to prepare a research proposal for her
research methodology module assessment. For her topic, she had chosen to explore the
extent to which work experience during a undergraduate course impacted upon job
performance in the graduate’s first job. Her interest in the topic stemmed from her own
spell in a large non-profit organisation which she felt prepared her well for her entry into
paid employment. So her overall research question was ‘To what extent and in what
ways does the degree student placement affect the job performance of the newly
recruited graduate?’
As a result of a lot of reading of reports, particularly in the educational media, Freya
generated the following research objectives:
1 To establish whether there is a link between the experience of a placement by the
undergraduate and the performance in the first career job.
2 To discover whether the type of placement (e.g. duration, amount of responsibility) is
associated with level of job performance.
3 To identify the variables which may be most strongly associated with job performance (e.g. quality of experience gained, personal confidence, closeness of relationship between placement duties and first-job duties).
Freya’s proposal went on to detail her research strategy and methods that she would
adopt. Her main data collection method was that of interviews with two samples of UK
Research in practice 8.1

218 Chapter 8 Writing and presenting the research proposal
students, one who had completed a work placement and one who had not done so. She
also suggested interviews with managers to whom the first-job graduates were
responsible.
Freya was praised for her proposal which raised an interesting question that would be
of considerable use to future students. However, her assessor criticised the lack of consistency between her objectives and her data collection methods. In particular, her assessor thought that the language of her objectives was couched in precision (e.g. duration of
placement, amount of responsibility and, most importantly, actual job performance)
whereas Freya’s data collection method was more consistent with subjective impressions.
In short, there was a mismatch between her objectives and method. The overall comment of her assessor was ‘you may have been wiser to stress in your objectives the quality of experience of the graduate in the first job rather than the performance: the first
suggesting subjective impressions, the second, objective measurement’.
It shows you have a good knowledge of the existing work
Your research proposal is your opportunity to demonstrate your expertise on the subject you have chosen. This will largely result from your close acquaintance with the relevant literature. From this you will gain the insights that will help you to produce an
interesting proposal. Later in this chapter we talk more about the place of the literature
in the research proposal. At this stage, we must stress that there is a world of difference
between the proposal which merely lists some references at the end and that which
clearly uses the literature to inform the choice of topic, the question(s) and objectives
and the research approach, strategy and methods.
If you have read thoroughly in preparation for your research proposal, this will shine
through because it will demonstrate that you are up to date with the current news and
debates concerning your topic. This will be at two levels: first, the news level where the
latest developments in, say, outsourced marketing are reported; and second, where academic research has been done to reach a deeper level of understanding of the topic. Of
course, current knowledge of the second level means that you don’t submit a research
proposal which merely replicates research done by other researchers, risking quite justifiable criticism from your assessor. But even if your proposal is similar to that already
done by another researcher, you can demonstrate novelty and knowledge of current
literature by building on existing research and, perhaps, approaching the research from
a different perspective, or in a different context.
It demonstrates to the assessor(s) that the research is viable
If you have to submit a research proposal without having to actually conduct the
research, there is a danger that you will not give sufficient thought to the practical
issues which would be faced by the researcher. The focus here is on resources.
As a management and business expert, your assessor will be mindful of the cost
implications of your proposal. This relates to the direct costs involved, such as travel
and possible accommodation expenses, and indirect costs, most particularly your time.

8.2 The importance of the research proposal 219
You will have other things to do in your degree programme, so submitting a proposal
which implies that it will take up more of your time that you can reasonably afford to
devote is not a good idea. It is not only the amount of time that is important, but the
length of time of the proposed research from start to finish that you should consider. It
is no good suggesting a research project which, say, plots the effect of management
coaching on longer-term job performance if the timescale is longer than you have
allocated.
But perhaps the most important resource of which you may underestimate the
importance is data availability. There may be little point in developing a proposal which
examines the importance of the vision that Sir Richard Branson brings to the Virgin
Group if you cannot get access to the man himself! In the example shown in Research
in practice 8.1, the proposal specifies the role of job performance. This assumes that job
performance data will be made available to the researcher. This is problematic, given
issues of confidentiality.
Two further points about the viability of the research are relevant for those of you
who are not only submitting your proposal for assessment but are going to conduct the
research you have included in that proposal. The first relates to the extent to which you
indicate that you are capable of doing the research. This may be implied in all that has
gone before in terms of evidence of quality of thought. But you also have the opportunity to show that you are creative and innovative, not only in the research objectives
you set but in the way in which you set out to achieve them. A genuine interest in the
research you propose is something that will be clearly evident in what you have written.
The second point about the viability of the research concerns the practical planning for
the research from your university’s perspective. A clear and timely proposal is often
used to decide who would be the most suitable supervisor for you.
It ensures that your research meets the requirements
of your university
It is likely that your proposal may have to contain your assurance that you have considered the requirements of your university. An example of these requirements is the university which states in its code of practice that an ethics committee needs to consider all
research proposals which actively involve human participants. In another UK university, the research proposal must clarify exactly what is meant by guaranteeing anonymity and confidentiality to research participants when they participate in the research.
Here, anonymity refers to concealing the identity of the participants in all documents
resulting from the research. Confidentiality is concerned with the right of access to the
data provided by individual participants and, in particular, the need to keep these data
secret or private. In addition, many ethics committees require researchers to state the
steps they will take to ensure protection of respondents’ identities and ensure that the
information collected is stored securely. We go into the subject of research ethics in
more detail in section 3.8.
For those of you who are going to conduct the research you have outlined in your
proposal, you are justified in thinking that a proposal which is deemed acceptable

220 Chapter 8 Writing and presenting the research proposal
implies that the research itself promises to be successful. Obviously, this cannot be
guaranteed. But it is reassuring to know that at least you started your research journey
with a suitable destination and journey plan.
In our view, the proposal could be started at day one of your research module. At that
stage, it may bear little or no resemblance to the finished article, but there are enormous benefits to be gained from starting early, not the least of which is committing
thoughts to writing. Once again, we make the point that writing is a great way of clarifying your thoughts. At the first stage of thinking about your research, you will probably not be able to note any more than outline topics in which you are interested. As
your thoughts develop, you may add some initial research questions and then follow
these with research objectives. All the time, expect that you will amend what you have
written. But do keep a copy of the various main versions of your proposal document.
You never know when you may need to resurrect a previous idea. Treat your proposal as
work in progress rather than seeing it as a ‘one off’ – something which you write up and
submit after your thinking, discussing and planning stage.
We have always encouraged our students to adopt this approach to the development
of their research proposals. This has proved to be beneficial to us too. When we sit and
discuss our students’ research ideas, it is so helpful to have a shared document to form
the basis of that discussion, even if it is in a very early stage of development. Also, we
keep copies of earlier drafts which show us how thoughts have developed and become
more concrete. We also have the ability to look back at ideas which may have become
‘lost’ and may be usefully reconsidered.
The earlier the research proposal is discussed with your supervisor, the better. Often
the supervisor will suggest that the scope of what you are considering is too ambitious.
This will usually be because the work proposed is not viable for one of the reasons we
have explained in section 8.2. Whatever amendments you need to make, and for whatever reason, the sooner you begin preparing your research proposal, the better.
8.3 When you should write your research proposal
To a great extent, the content of your research proposal will be governed by the format
required by your university and by the proposal content itself. What we offer here is a
guide to the sections that you will most likely need to include.
Research overview
You may think of this first section as an abstract or an executive summary. As such it is a
brief statement of what you intend to do for the research. It should be no more than a
8.4 What you should include in your research proposal
8.4 What you should include in your research proposal 221
paragraph or two and should describe the proposal content to busy readers in no more than
a few words.
Title
Try here to reflect as accurately as possible the content of the proposal. The economist
Adam Smith may not have written a research proposal for his famous book, but it seems
to us that you can’t be more precise than the title he chose:
An Inquiry into the Nature and
Causes of the Wealth of Nations
(1904).
You may want to leave composing the title until you have finished the body of the
proposal. Alternatively, this may be your start point. If this is the case, don’t be afraid to
amend the title as your work progresses.
Introduction to the research
This is an important part of the proposal. You should place the proposed study in a context which will assist your assessor in understanding why it is you have chosen this particular topic. You may wish to demonstrate the topical relevance of your proposal if it is
concerned with something that is generating current debate. An example here would
be the way the retail banks are trying to overcome their negative public image through
their marketing strategies. Alternatively, you may wish to concentrate upon a problem
being experienced by an organisation with which you are familiar. Here, an example
may be the difficulty the organisation has with retaining key employees.
In either of these two cases, your assessor will be keen to see if you are knowledgeable
about the topic and can relate it to existing theory. So in the case of the retail bank proposal you may show how this is drawn from, and/or may contribute relevant marketing
theory. Similarly, you may use the literature on employee retention strategy theory in
the second example.
It will help to persuade your assessor of the quality of your proposal if you can explain
why you are interested in the topic. This may be because you have worked in the particular organisation or have studied the retail banks in another module.
If your proposal is the precursor to your going on to conduct the research, it will be
helpful if you show that you are enthusiastic about the research topic. This will help
convince your assessor you have sufficient commitment to sustain your effort over the
life of the research.
In short, the real value of this background section is to convince the assessor that the
research is worth pursuing; not just by you as part of your course, but by a competent
researcher who will add to an understanding of the particular topic you are studying.
Research Question(s)
In section 1.7 we explain what is meant by research questions and research objectives
and the difference between the two. To recap, the research question may be one overall
question, or a number of questions that the research process will address which are often
the forerunner of research objectives. Research objectives are clear, specific statements

222 Chapter 8 Writing and presenting the research proposal
that identify what the research process seeks to achieve as a result of doing the research.
So the movement from research questions to research objectives is a developmental one:
objectives follow questions to provide precision to that which is more general.
This progression from research questions to research objectives is reflected in the
order we suggest for the sections in the research proposal. We propose that the literature review section separates them. The reason for this is that we see the research question as directing and leading you into the relevant literature, whereas the research
objectives are developed as a result of careful consideration of the literature. Of course,
the process is not that straightforward. Our model implies that reading generally on
the topic of your interest may fire your imagination sufficiently to raise workable
research questions; and having developed a research question(s), you then embark on
more specific reading to develop your question(s) into objectives. The reality is that
the three processes of reading, question development and objective development may
take place in no coherent order or, more likely, simultaneously without you really
sensing that there is a particular order. There are some examples of research questions
in section 1.7.
Literature Review
Chapter 2 explains in detail the role of the critical literature review in doing research
and ways in which this may be accomplished. A critical literature review is something
that you would normally include in a final written project report. Here we are more
concerned with that which you should include in your research proposal. It is important to show in your proposal that you are knowledgeable about the literature that
relates to your research topic. More specifically, you should use this opportunity to
explain how your proposal relates to the academic debate which is being conducted in
the literature. You will be expected to show a clear link between the previous work that
has been done in your field of research interest and the content of your proposal. Put
simply, you should show in your review of the literature where your research question(s)
came from and how your research objectives will move the debate forward by, say
applying a new perspective or setting your research in a new context: the literature is
both a point of departure and a signpost pointing to your destination.
This all suggests that it is insufficient in your research proposal just to provide an
overview of the key literature sources from which you intend to draw. Clearly you
should include references to key articles and texts, but you must also show relevance to
your research.
A research proposal about power
James was a part-time student. His full-time job was in the fire service where he led a
team of firefighters. His organisation had recently been through a major change programme which had been only partly successful. James had noticed that in some parts of
Research in practice 8.2
8.4 What you should include in your research proposal 223
his organisation, the programme had been implemented more effectively than others. In
particular, he felt that this might relate to the leadership style adopted by the manager
in charge of the individual departments.
James’s initial theory was that a leadership style which was based on the principle of
‘leading by example’ was more likely to yield positive results than one rooted in
autocracy.
James had studied leadership as part of his course and had decided to examine his
organisation’s change programme as his research project. He was interested in the idea
of power and how this related to leadership and particularly attracted by the typology of
power used by Raven and French (1959), still a key influence in the power literature
although written half a century ago.
The section below is from James’s research proposal, where he outlines the way in
which he plans to incorporate the idea of power into his study.
This study will draw on the theory of power as noted by Raven and French (1959), and
reviewed more recently by Elias (2008). Power is defined here as the ability to influence
others to believe and act in such a way that those in power would wish.
Raven and French argue that power manifests itself in five main forms. These are:
• Coercive Power
• Reward Power
• Legitimate Power
• Referent Power
• Expert Power
1. Coercive Power
This form of power is based upon the idea of coercion. This means that someone is
forced to do something, typically against their will. The main objective of coercion is
compliance.
Coercive power can lead to unhealthy behaviour and dissatisfaction at work.
Leaders who use this leadership style rely on threats in their management styles.
Often these threats can relate to dismissal or demotion.
2. Reward Power
This form of power is based on the idea that as a society we are more inclined to do
things well when we are getting something in return for this. The most popular forms
of reward are pay rises, promotions or compliments. The problem with this form of
power is that when the reward does not have enough perceived value to others, the
power is weakened. One of the frustrations when using rewards is that they often
need to be bigger than the last time if they are to have the same effect. Even then,
when they are given regularly, employees can become accustomed to the rewards
and as a result, they will lose their effectiveness.
3. Legitimate Power
Legitimate power is usually based on a role. People traditionally follow the one person
with power which is solely based on their position or title. This form of power can

224 Chapter 8 Writing and presenting the research proposal
easily be overcome as soon as someone loses their position or title. This form of
power can be an ineffective way to persuade and convince other people.
4. Referent Power
Leaders in this form of power are often seen as a role models. Their power is often
treated with admiration. This power emanates from a person that is highly liked and
people identify strongly with them in some way. Leaders with referent power often
have a good appreciation of their environment and therefore tend to have a lot of
influence. But individual responsibility flowing from referent power is heavy.
5. Expert Power
This form of power is based on in-depth information, knowledge or expertise. The
leader who has a particular expertise within an organisation can often persuade
employees, who trust and respect them, to do things for them. This expertise is
greatly appreciated and forms the basis of this type of leadership.
James’s plan was to construct an initial theoretical model which related the variables
of organisational change effectiveness and power in leadership. He proposed to define
indicators of each and collect data noting the extent to which these indicators were
present in the change programme.
References
Raven, B. H. & French, J. (1959). The bases of social power. In D. Cartwright (Ed.)
Studies in social power (pp. 150–167). Ann Arbor, MI: Institute for Social Research.
Elias, S. (2008). Fifty years of influence in the workplace: The evolution of the French
and Raven power taxonomy,
Journal of Management History, 14(3): 267–283.
Research objectives
We made the point earlier in this section that the research objectives are developed as a
result of careful consideration of the literature. Make sure that they:
relate to and are developed from your research question(s);
relate to and are developed from your review of the literature;
are specific, measurable, achievable, realistic and timely (see Chapter 1 and
Table 1.7).
Method
This will be one of the more detailed sections of your research proposal. It flows directly
from your research objectives and shows how you will go about achieving them. In the
method section, you have the opportunity to show the assessor the extent to which you
really understand the research process through ensuring this close connection between
research objectives and method. For example, if one of your objectives is ‘to establish
the influence of price promotions on supermarket bread sales’, you will need to think
about ways in which you may quantify the effect price has on sales. This may be done,

8.4 What you should include in your research proposal 225
for example, through the examination of sales records. Whether it is possible to collect
the data is an important consideration here: that is a potential assessor concern that
you must address in the proposal. This example also raises the issue of validity and reliability (see section 5.6). A key question that will be going through your assessor’s mind
will be ‘are the methods being proposed here likely to deliver credible results that can
lead to sound, valid conclusions?’ It is vital that you think this through thoroughly
when preparing the method section of your proposal. You should justify your choice of
methods in the light of the question about credibility. So in defence of any questioning
of your research methods, you can answer, ‘I chose to collect data this way because I feel
this will provide valid and reliable data’.
Your method section can be divided into two parts: research design and data collection. In the research design part of the section, you should explain your overall strategy,
for example a case study or a survey strategy, alongside two other aspects of your
research design. These are, first, the location of your research and, second, the research
population from which you will select your sample (section 6.2). If your research is to
be carried out in one organisation, then the location is straightforward and needs no
further elaboration. However, if your research topic is not concentrated in one organisation, then you will need to detail the locations. So, for example, a study of the introduction of a new distribution strategy by Internet retailers may mean choosing a sample
of retailers among whom you are carrying out a comparative study. In this case, you
would need to explain why you chose the participant organisations. Your reasoning
here will be judged against the extent to which your choice of organisations is consistent with your research objective and the need to provide credible data. You may be proposing a project which is even more generic, say, a study of the effect of changing
consumer preferences for music subscription services on the music industry. Here you
would need to explain why you chose this sector.
Secondly, in this part of the section on research method, you will need to describe
and justify the population from which you propose to collect data. This may be a small
sample of employees that you propose to interview, a large sample that will receive a
questionnaire (Chapter 6) or a number of companies about which you can access secondary data (Chapter 4).
The method section should also include an explanation of the way in which you
intend to carry out the research. It could involve, say, questionnaires, semi-structured
or unstructured interviews, analysis of secondary data or a combination of data collection techniques. Again, it is essential to explain why you have chosen these techniques.
This explanation should reflect upon whether this is the most effective way of meeting
your research objectives and providing credible data.
The data collection section should be much more detailed about how specifically the
data are to be collected (Chapter 6). For example, if you are using questionnaires, you
should specify your population and sample size and how your sample will be selected.
You should also clarify how the questionnaires will be distributed, the likely response
rate, and how the data will be analysed. If you are using interviews, you should explain
how many interviews will be conducted, how long they will last, whether they will be
audio-recorded and how they will be analysed. You should show your assessor that you

226 Chapter 8 Writing and presenting the research proposal
have thought carefully about all the issues regarding your method and their relationship to your research objectives and data credibility. Don’t worry about the necessity to
provide precise detail. It is normally not necessary in the proposal to include precise
detail of the method you will employ, for example the content of an observation schedule or questionnaire questions. In this section, you will also need to include a statement
about how you are going to adhere to your university’s ethical code (see section 3.8).
Timescale
This is a useful part of your research proposal. Clearly breaking down the research process into a series of steps will show you whether it is reasonable to expect the various
tasks to be done within the timescale. For example, it is no good taking three months to
do your literature review if your project has to be submitted in two months. More tellingly, it will show clearly how much time is to be available for data collection and analysis. The time required here is often far more than anticipated. This is for a number of
reasons, not the least of which is the degree of dependence that you may have upon
your respondents from whom you are collecting your data. Analysing the data also
often takes far more time that anticipated (Chapter 7). This is particularly so if you have
a limited amount of experience with any software that is being used for analysis.
Devising a timescale is also an important part of the research proposal because it
enables your assessor to assess the variability of the work you propose. You can follow
all the advice in this chapter, and in this text, and produce a beautifully crafted, intellectually coherent research proposal only to be marked down with the damning comment ‘an excellent proposal but one which proposes research that could not possibly be
done within the timescale’. Discussing your plans with your supervisor should avoid
this happening. Experience of reading many proposals usually gives supervisors a good
idea of what is practicable in a given timescale.
Table 8.1 An example timescale for a research project.
Task To be completed by:

Begin research idea formulation and first coverage of literature.
Main part of literature research completed, research questions and
objectives defined.
Research proposal submitted.
Make arrangements for data collection.
Literature research finished and review written.
Secondary data research.
Primary research and analysis.
Draft written report.
Revised draft written report.
Final submission of written report.
1/10/2017
20/11/2017
30/11/2017
31/12/2017
31/12/2017
31/12/2017
20/2/2018
31/3/2018
15/05/2018
30/5/2018

8.5 The style you should use to write your research proposal 227
Resources
It may be that you are asked to list the resources required for the completion of the
project you propose. This also will assist your assessor in assessing the viability of your
proposal. We explained in section 8.2 that you should detail direct costs involved,
such as travel and possible accommodation expenses and indirect costs, most particularly your time.
We also mentioned in section 8.2 the importance of data availability. Your assessor
will be concerned about approval from any organisations in which you are planning to
conduct your research and may well require written evidence of this. You will also need
to convince your assessor that the response rate to any questionnaire that you send is
likely to be satisfactory.
You should also be able to convince your assessor that you can undertake the analysis
of your data satisfactorily. This is often an aspect of the research process that is ignored
when the proposal is submitted. It is important that you describe the resources you
have available for data analysis purposes. This may include computer software and the
appropriate skills to perform the analysis, or help in learning these skills in an appropriate time.
References
You may be tempted to impress your assessor (and fellow students!) with a long list of
references at the end of your proposal. This is not necessary. Do bear in mind the point
we made in section 8.2 that simply listing references to those sources which you have
consulted is not what is required. You should list only those references that clearly use
the literature to inform the choice of topic, the question(s) and objectives and the
research approach, strategy and methods. Be careful to ensure that your references are
in exactly the format required by your university (section 2.7 and Appendix 1).
We have included a section a number of dos and don’ts on writing style in this chapter.
It will obviously be very relevant to those who are writing the project report. We hope it
will also help those of you who are doing the research proposal without progressing to
the full project report following the conduct of the research. The points we make in this
section are sufficiently generalised to meet all research writing needs.
Do write clearly and simply
You are unusual if you enjoy the style in which much of the material you have to read
for your university work has been written. We hope you find the content interesting (it
would be worrying if you didn’t!) but the style, well that’s a different matter. We agree
with the American academic C. Wright Mills (1970: 239–40), who, you can see from the
The style you should use to write your
8.5 research proposal
228 Chapter 8 Writing and presenting the research proposal
quote that follows, was unimpressed by the writing of many of his academic
colleagues.
The . . . lack of ready intelligibility [in scholarly writing], I believe, usually has little or
nothing to do with the complexity of the subject matter, and nothing at all to do with
profundity of thought. It has to do almost entirely with certain confusions of the academic writer about his own status . . . To overcome the academic prose you first of all have
to overcome the academic pose . . .
This is not to say that you should trivialise what you write in your research proposal
or report. Far from it. You will be dealing with a serious subject for a serious purpose
so your writing should reflect this. But you can be serious and clear at the same
time!
An example will illustrate our point. Consider the following two statements.
1. The research proposed here is seeking to gain an understanding of the ways in which those
factors which are closely associated with the relationship between consumer preferences
for security and economy are reconciled when the consumer purchasing decision for
delivery is made in the Internet seller auction sector.
2. When sellers on Internet auction sites decide which carrier to use, they consider both
security and cost. In this research I aim to establish how these two factors influence the
purchasing decision.
Both statements say the same thing. But the second uses 34 words rather than 50. This
is 32% fewer words! So the ‘wordy’ research proposal which takes 1,000 words could do
the same job for 660. Just consider the benefit to you of asking the assessor to do 32%
less work! In addition to this obvious advantage, you will be conveying the same meaning in a much clearer way, so not only will your assessor have to read fewer words but
those which are read will be digested much more easily.
Do write simple sentences
Not only does the second statement in the example above use fewer words while conveying the same meaning but it uses two short sentences rather than one long one.
Look at your writing and break up those long sentences. Use the simple rule: one idea –
one sentence.
Research writing is often difficult because you are writing about ideas and facts
and usually working out relationships between the two. Long, complicated sentences
usually mean that you aren’t sure about what you want to say. This is quite understandable. But shorter sentences are much better for explaining complex information because they break the information up into smaller chunks which are easier to
follow.
One more golden rule: avoid the embedded clause! Consider these two sentences.

8.5 The style you should use to write your research proposal 229
Sentences with lots of clauses and exceptions confuse the audience by losing the
main idea in a jungle of words. Don’t put all your ideas in one sentence; separate
your ideas and make each one the subject of its own sentence, like in sentence 2
above.
Do be careful with spelling and grammar
Spelling and grammatical errors detract from the quality of your work and make it look
less credible and authoritative. It’s too easy to say that the word processing software you
use will do this job for you. It won’t tell you that you have used ‘practice’ and ‘practise’
or ‘moral’ and ‘morale’ in the wrong context. Try to get a friend who is a good speller to
check your work so you don’t make these errors.
Common grammatical errors can irritate the assessor. All assessors have their pet
hates! So avoid simple mistakes such as referring to one interviewee as ‘they’; calling
lots of data ‘it’ rather than ‘they’ (a single piece is ‘datum’); using clichés such as ‘the
real world’ to refer to that which exists outside the university. Again, persuade a critical
reader to check your writing for you. You may not avoid all the grammatical pitfalls, but
at least you may not fall into the obvious ones.
Don’t use jargon
All disciplines have their jargon, and business is as guilty as most. The meaning of terms
like ‘buy-in’ are clear enough to us, but nonetheless they have the potential to irritate
your assessor. There are perfectly good words to use which are correct English and the
irritated assessor may ask ‘what is wrong with “gain agreement”’?
Jargon should not be confused with technical terms. Some technical terms are
perfectly valid. Here it is useful to put a glossary of technical terms in the appendices. However, do not assume that your assessor will have your level of knowledge of
the subject and, in particular, the context. Try to put yourself in the position of the
assessor. As part of a research writing workshop, we ask students to assess research
proposals. Some of these are written by students who use their own organisation as
a context for the research. The students assessing these proposals are usually amazed
at the assumptions that their fellow students make about the assessor’s prior
knowledge.
1. While confidentiality is of the utmost importance, it is necessary to bear in mind the
practicalities of organisational life, therefore respondents will be interviewed individually
insofar as this is possible, although privacy may not always be possible for the same reasons
of practicality.
2. I plan to interview the respondents individually in as setting which will be private and
therefore conducive to confidentiality. This depends upon the availability of suitable
rooms.

230 Chapter 8 Writing and presenting the research proposal
Do beware of using a large number of quotations from the literature
In your research proposal you are unlikely to use many quotations from your literature
sources. Yet, one or two may lend authority. However, it is important that you use these
to illustrate, or emphasise a point that you are making rather than insert it to make your
proposal ‘look good’. We should stress here that the point that you are making must be
your point. If you are using a quotation that is largely unexplained, it is that author’s
point alone. In general, it is better to explain other people’s ideas in your own words. In
that way, you can really gain an understanding of what the author means and convince
your assessor of your understanding.
When it comes to the project report we feel that quotations from the literature
should be used sparingly. Sometimes we read draft projects that consist of little more
than a series of quotations from books and journal articles that a student has linked
together with a few sentences of her or his own. This is unacceptable. It tells us very little about the student’s understanding of the concepts within the quotations. All it
shows is that the student has looked at the book or journal article and (hopefully!)
acknowledged sources correctly. Using quotations in this way means that line of argument becomes disjointed and difficult to follow. That doesn’t mean that you should
never use quotations. But we advise you to use them in moderation to create maximum
impact in supporting your storyline.
Do be careful when using personal pronouns
Often project reports are written in a rather dry and unexciting style. This is partly
because the convention has been to write impersonally (e.g. ‘it was decided to interview a small sample . . .’). Here, the writer is distanced from the text. Some writers
anonymise themselves by referring to themselves as ‘the author’ or ‘the researcher’.
Our view is that, where appropriate, your writing is much livelier if you use the first
person (‘I decided to interview a small sample . . .’). We say, ‘where appropriate’. This is
because your research strategy and methods may dictate your choice of personal pronoun. It may be useful to ask yourself, ‘Am I inside or outside the data collection process?’ By ‘inside’ we mean that the researcher is an intrinsic part of the data collection.
This may be the case with interviews. However, with questionnaires, the researcher is
‘outside’ the data collection process. Many academics, but not all, use the broad rule;
only if you are inside the data collection process is it appropriate to use the personal
pronoun.
Do check this with your supervisor. There may be conventions which you are
expected to observe. Some researchers think that excessive use of ‘I’ and ‘we’ casts doubt
on your ability to stand outside your data and to be objective.
Do be careful when using tense
Rather like the use of person, there are no clear rules about the correct use of tense in
academic writing. We usually recommend that you use the present tense when referring to previously published work (e.g. ‘Smith notes in his earlier article . . .’) and the

8.6 How your research proposal will be judged 231
past tense when referring to your present results (e.g. ‘I found that . . .’). Although there
are exceptions to this rule, it serves as a useful guide.
Do be careful when using gender
No doubt you will already have been warned about use of language that assumes the
gender of a classification of people. In business and management, the most obvious
example is the constant reference to managers as ‘he’. This is not only inaccurate in
many organisations, it also gives offence to many people of both sexes. It is easy to avoid
falling into this trap. Instead of ‘I propose to interview each senior manager in his
office’, it is more appropriate to write, ‘I propose that each senior manager’s office will
be the setting for the interview’. (In an early draft of another book we have written, we
referred to a particular writer as a ‘master craftsman’ before deciding it would be more
suitable to change this to ‘an expert in the field’!)
In the event of your research having an international dimension, it is a good idea to
be aware of any country-specific or national guidelines on the non-discriminatory use
of language.
Do preserve anonymity
It is likely that you have given your participants or respondents (and the organisations)
from whom you collected data an undertaking that you would not disclose their identity in your written work. You will need to conceal their identity in your research proposal or report. Normally you can do this by inventing pseudonyms for organisations (as
in Research in practice 8.2, where AAA and BBB were used) and not to name individual
participants or respondents. It may make your work a little less interesting to read, but in
such cases, you have no choice. Anonymising may also allow you do be rather more
critically evaluative than you could have been with named participants or respondents.
Do successive drafts of your work
In section 8.3 we recommended that you continually revise your research proposal,
keep each successive draft and treat the report as ‘work in progress’. This point applies to
all you’re writing. Your style and content will be refined, sharpened and clarified with
each successive version.
Your module guidelines should make clear the assessment criteria against which your
research proposal will be judged. In this final section, we offer three guidelines. These
may be in your published assessment criteria, and they summarise much of the material already covered in this chapter in particular and this text in general. We offer them
here to emphasise their importance.
8.6 How your research proposal will be judged
232 Chapter 8 Writing and presenting the research proposal
The extent to which the components of the proposal fit together
The main components of your research proposal should fit together in one coherent
and seamless whole. These are:
1 The introduction. This should contain an explanation of the reasons for conducting
your research and include a summary of the previous published research, covering
relevant theories in the topic area.
2 Your research question(s) and objectives, which should be based on the material in
the introduction.
3 Your research design, including the proposed method, which should flow directly
from these research question(s) and objectives.
4 The time that you have allocated, which should be a direct reflection of the methods
you employ and the time available.
5 The resources that you need, which should be a direct reflection of the methods you
employ and your own skills.
The absence of preconceived ideas
Your research should be a voyage of discovery. As a broad rule, if you know the answer to
the research question already – look for another topic! Having preconceived ideas will
stunt your creativity and lessen the enjoyment you derive from the research process.
The viability of the proposal
This is the answer to the question: ‘Can this research be carried out satisfactorily within
the timescale and with available resources?’
We end this section on the content of the research proposal with a complete proposal (Research in practice 8.3) for you to consider. This should not be seen as ideal.
Indeed, you should review its content in the light of the points made in this chapter to
conduct your own assessment.
Research proposal
Research in practice 8.3
Research proposal
Mr C.D. Eff, MBA (part-time)
Knowledge Management in High-Technology Manufacturing SMEs:
Key Strategies to Maintain Competitive Advantage
Research Question
This research is designed to answer the question ‘What are the key features of a
Knowledge Management Strategy which can contribute to competitive advantage for
UK knowledge-based small and medium sized enterprises (SMEs) in the high-technology
manufacturing sector, and why are these features evident?’

8.6 How your research proposal will be judged 233
Knowledge Management
Growth of the field of knowledge management (KM) was began in the 1990s and has
been ongoing (Hislop, 2013). This has been driven by the evolution of a ‘knowledge
economy’ where businesses must continually adapt and use their knowledge effectively
to maintain competitive advantage. Teece (2001) attributes superior performance to the
ability of firms to be good at innovating, protecting intangible and difficult to imitate
knowledge assets and using those assets effectively.
Knowledge management has often been described as comprising three elements: people,
processes and technology (Edwards, 2011). (It is important to stress that the term
processes refers to the business processes of the organisation concerned, not just to its
knowledge management processes.)
These three elements link together, each of them having a reciprocal relationship with
each of the other two. For example, people help design and then operate processes,
while processes define the roles of, and the knowledge needed by people.
Tidd et al. (2001) state there are five main types of KM strategy. These are ripple (bottomup continuous improvement), integration (functional knowledge into processes),
embedding (coupling systems, products and services), bridge (novel combination of
existing competencies) and transfer (existing knowledge in new context). The optimum
knowledge strategy is likely to be a mix of these and will depend on the structure and
culture of the organisation, the market environment and the availability of resources.
Knowledge Sharing and Organisational Memory in High-Tech SMEs
For large organisations, organisational design and formal procedures form the basis for
knowledge integration. In contrast, for high-technology SMEs, the role of senior management
is far more important in the integration of technological knowledge and devising strategies
to exploit knowledge competencies. Tidd et al. (2001: 56) state that it is the manager’s ‘level
of technical and organisational skills’ which determine whether they ‘will be able to develop
and commercially exploit a firm-specific technological advantage’. As firms grow, it is clear
that organisational design, processes and culture need to be developed to maintain and
foster the environment of knowledge sharing which was inherent in the smaller organisation.
The five main learning activities described by Garvin (1993) are ‘systematic problem
solving, experimentation, learning from experience, learning from others and transfer of
knowledge quickly and efficiently throughout the organisation’. For high-technology
SMEs practice is often lacking in the application of past experience and the systematic
recording and sharing of often tacit knowledge throughout the business. Rather as
suggested by Hutchinson and Quintas (2008) these firms manage knowledge in informal
ways which are both structured and deliberate.
Research Objectives
The aim of this research is to understand the characteristics of the firms and their products
which determine the success of different types of KM strategy. The research objectives are:
1. To identify KM strategies in operation in SMEs in the high-technology manufacturing
sector.
2. To analyse the importance of KM and understand how the strategies and practices in
operation contribute positively or negatively to competitive advantage.

234 Chapter 8 Writing and presenting the research proposal
3. To understand the practices which enable the development of organisational memory
and why these are evident.
4. To understand the characteristics, structure and products of a firm which determine
the success of different types of KM strategy and why these are significant.
5. To develop a list of key practices and a route for implementation of a successful KM
strategy for an SME in the high-technology manufacturing sector.
Research Methods
Research Approach and Design
The analysis of the research will be largely quantitative, highlighting patterns. The
findings will be compared to theory in order to describe the patterns which exist. An
inductive approach will be used in order to develop theory as a result of the research
findings, although there will be an element of deduction in that the structure of the
research will based on the academic literature.
The research will be based on two case study organisations using:
1. structured interviews with a senior manager in both of the case study organisations;
2. questionnaires to all employees in both of the case study organisations.
The cases chosen all have the following characteristics: located in the SME hightechnology manufacturing sector, possessing a significant R&D function as a basis for
product development and are UK based. The method of selection of case studies was
through personal networking and sampling to find a number of organisations with the
required characteristics. Effort was also made to select organisations with different
characteristics such as markets, sizes and life-spans. Within the case studies the
respondent will be senior managers who understood the KM practices and business
environment. I intend to interview one management participant from each organisation.
The questionnaire respondents will be either all, or at least 10, of the employees working
in research or related areas, depending on the size of the company.
A semi-structured interview will be held with a senior manager in each case study
organisation to gain company information such as number of employees, number of
research staff, annual turnover, market sector and products. The interviews will include
open ended questions to discuss the KM practices in order to gain an understanding of
the approach taken, the structure and effectiveness of the KM practices and how and
why certain strategies work whilst others are less effective. Each respondent will be asked
to answer identical opening questions, although follow-up questions are likely to differ.
The questionnaire delivered to employees in each organisation will be used to establish the
practices and individual perceptions in the organisation. The aim is to gain an understanding
of how the KM strategies work through gaining the opinions and understanding of the
employees. The questionnaire will have closed questions with set responses in order to
map the KM characteristics of the organisation. There will also be a few open questions for
participants to make comments on the various aspects of KM in their organisation.
My approach to the design of the questionnaire is as follows:
• Preliminary framework built on the review of theory from academic literature prior to
design of questionnaires and structured interviews
• Pilot interview – restructure the questions as necessary

8.6 How your research proposal will be judged 235
• Pilot use of questionnaire – restructure the questions as necessary
• Structured interviews – notes taken during interview and audio-recording to produce
a full record immediately after interview
• Questionnaire explained and handed out/collected during session
Data Collection and Analysis
I intend to perform research in my own organisation, called here AAA, for which I have
been granted access to carry out the interviews and questionnaire. In addition, I believe
that I require one further organisation to make a comparison of the KM practices and
their effectiveness. I have contacted five organisations and have obtained a verbal
agreement for access from one of these (called BBB). I am therefore confident that I will
be able to gain the required number of participating organisations.
In order to make the process run smoothly and to obtain a speedy and efficient response
to the questionnaires I intend to run one or two sessions in both organisations where all
of the participants will be present and during which I will hand out and collect the
questionnaires. This will both ensure a high response rate and that the correct participants
answer the survey. In addition, it will enable me to explain clearly the aims of the
research, control how the survey is administered and make clear the type of information
required for the open-ended part of the questionnaires.
Analysis of the structured interviews will be mostly qualitative, with the interviews used
to understand the business environment, characterise knowledge strategies and canvas
individual opinion on their effectiveness. Quantitative analysis of the questionnaires will
be used to map the characteristics of both organisations and discover patterns in the
responses within organisations. Consideration of the data and patterns found in both
organisations will be used in conjunction with academic theory to try to explain the
findings and answer the research objectives.
Validity and Reliability
I will attempt to achieve internal validity through the use of multiple sources of evidence,
structured interviews and questionnaires. The design of questions will be based on
understanding of the theory from the literature and pilot testing of the interview and
questionnaire will be used to make sure questions are understood as intended.
I have ensured external validity through the use of multiple cases to examine whether
findings can be generalised across both organisations. There is no requirement to make a
statistical analysis of the results for generalisation here as the approach chosen examines
practices and their effectiveness. The aim is to explain the findings and explore
generalisability through a comparison of findings with theory.
Reliability is pursued through structured interviews with questions derived from the
literature in order to examine cases in same way. In addition, the surveys reinforce the
interviews and obtain views from a wider group. The questionnaires will all be
administered at the same time and in controlled manner with explanation to ensure
participants all understand research in same way.
Research Ethics
Respondents will be given a clear written description of the purpose, scope and intended
outcomes of the research. The type of information required for the research will be
clearly stated as will the policy for anonymity and confidentiality.

236 Chapter 8 Writing and presenting the research proposal
The research will be carried out in a way that will ensure confidentiality of the participant
organisations and the individual participants in the surveys. The organisations which
participate in the research will only be referred to by their pseudonyms AAA and BBB and
not be named in the project. Interview participants will not be named and the
questionnaires will be anonymous.
The interview questions and questionnaire will be designed to examine only the practices
of KM, no confidential product or customer information will be required.
Timescale and Resources
The plan for the research project timescales is shown below.
• Literature review. I have already performed a background literature search to help
formulate research ideas. I plan further extensive period of research before writing
the literature review. This will be completed by December 2016.
• Questionnaire/Interview Design. I have a questionnaire which needs adapting for
use. I intend to design the structured interview and questionnaire after the majority
of the literature review is complete, will both be piloted and their design reviewed.
This will be completed by December 2016.
• Interviews and questionnaires. I plan to visit participant organisations to carry out
data collection during January and February 2017.
• Data Analysis. This will be completed by the end of March 2017.
• Project draft completed by end of April 2017. Final submission end of May 2017.
The main resource required to carry out the research is my time, I have the support of
my employers to carry out this research and I will be able to take days out of work to
visit the other participant organisation. I have the means to visit the participants and
also to analyse the data and write up the project report.
References
Edwards, J. (2011). ‘A Process View of Knowledge Management: It ain’t what you do,
it’s the way that you do it’.
The Electronic Journal of Knowledge Management, 9 (4),
pp. 297-306.
Available online at www.ejkm.com
Garvin, D. A. (1993) ‘Building a Learning Organization’,
Harvard Business Review,
71(4) pp. 78-91.
Hislop, D. (2013). Knowledge Management in Organisations 23rd ed. Oxford: Oxford
University Press.
Hutchinson, V. and Quintas, P (2008), Do SMEs do knowledge management?: or
simply manage what they know?
International Small Business Journal, Vol. 26, No. 2,
pp 131-154.
Teece, D.J. (2001), ‘Strategies for Managing Knowledge Assets: The role of the Firm in
Structure and Industrial Context’ in Nonaka, I. and Teece D.J. (Eds.)
Managing
Industrial Knowledge: Creation Transfer and Utilization,
London: Sage.
Tidd, J., Bessant, J. and Pavitt, K. (2001),
Managing Innovation: Integrating
Technological, Market and Organizational Change
. Chichester: Wiley.
Thinking about your research proposal 237
Writing a research proposal is important because: it clarifies your ideas and helps you
organise those ideas; it shows you have a good knowledge of the existing work; it
demonstrates to the assessor(s) that the research is viable; it ensures that your
research meets the requirements of the university and your programme.
It may helpful to begin your research proposal at the start of the research process and
treat it a ‘work in progress’ by amending it as you progress through the process of
preparing the proposal.
The content of the research proposal is likely to be: research overview; title; introduction to research; research questions(s); literature review; research objectives;
method; timescale; resources required and literature references. The main points to
consider when writing the research proposal or report are: to write clearly and simply; to use short, simple sentences; exercise care in spelling and the use of grammar;
avoid jargon; beware of using large numbers of quotations from the literature; the
use of person, tense and gender and the preservation of confidentiality and
anonymity.
Among the criteria against which your research proposal will be assessed will be the
extent to which the various components of the research fit together and the absence
of preconceived ideas.
Summary
Look again at the idea of the six honest serving-men (section 8.2) to generate some
initial headings for your research proposal. Your headings may respond to the following ‘prompt questions’: What are the research questions I am seeking to answer?
What are my research objectives? Why is the research I propose significant? When are
the key dates in the research process? How will I go about collecting the necessary
data to answer the research questions? Who are the key participants in the research
process (e.g. gatekeepers, respondents)?
Study the brief given to you detailing your university’s requirements for the research
proposal and note the key sections that you should include in your proposal.
Look again at the brief given to you specifying your university’s requirements for the
research proposal and ensure that you are aware of the assessment criteria that will be
used to grade your proposal.
Consider the advice given in section 8.5. on the use of person in your writing. In the
light of this advice and that from your supervisor, decide which person is appropriate
for you to use.
Thinking about your research proposal
238 Chapter 8 Writing and presenting the research proposal
Kipling, R. (1902, reprinted 2007).
A Collection of Rudyard Kipling’s Just So Stories. London:
Walker Books.
Mills, C.W. (1970). On intellectual craftsmanship. In
The Sociological Imagination. London:
Pelican.
Saunders M., Lewis, P. and Thornhill, A. (2016).
Research Methods for Business Students. (7th edn).
Harlow: FT Prentice Hall.
Smith, A. (1904).
An Inquiry into the Nature and Causes of the Wealth of Nations (5th edn.)
London: Methuen and Co.
References
Appendix 1
How to reference
Within business and management, two referencing systems predominate, the Harvard
style and the American Psychological Association (APA) style. These are both authordate systems where all the sources are listed alphabetically in the ‘references’ or ‘bibliography’ section using the authors’ family names. If there is more than one work by the
same author or originator in this list, they are listed chronologically. If there is more
than one publication by the same author from the same year, you need to include a, b,
c etc. immediately after the year. Do not forget to ensure that these letters are consistent
with the letters used for the references in the main text.
Increasingly we read electronic versions of books and journal articles. For such versions, it is usually acceptable to reference them using exactly the same format as printed
books and journal articles, provided the copy you have read is a facsimile copy.
Otherwise you need to reference them as explained in Table A1.1. Facsimile copies have
precisely the same format as a printed version, including page numbering, tables and
diagrams, other than for the copies of journal articles, which are published ‘online
first’.
Online first refers to forthcoming articles that have been published online prior
to appearing in journals. They therefore do not have a volume or part number, and the
page numbering will not be the same as the final copy. When including an ‘online first’
copy in the list of references, you should always include the DOI (digital object identifier) as part of the reference. The DOI provides a permanent and unique identifier for
that document. Where there is no DOI, it is usual to include the document’s URL
(Uniform resource locator – usually its web address). As the URL is not permanent, the
date when it was accessed is also included in the reference.
The Harvard style is an author–date system, a variation of which we use in this book. It
usually uses the author’s or originator’s name and year of publication to identify cited
documents within the text. All references are listed alphabetically at the end of the text.
The style for referencing work in the text and in the references or bibliography is outlined in Table A1.1.
The Harvard style
Table A1.1 Using the Harvard style to reference
To cite In the text In the references/bibliography
General format Example General format Example
Books and chapters in books

(first edition) 1 author:
(Family name year)

2 or 3 authors:
(Family name, Family
name and Family
name year)
4+ authors:
(Family name et al.
year)
1 author:
(Silverman 2007)
2 or 3 authors:
(Berman Brown and
Saunders 2008)
4
+ authors:
(Millmore et al. 2010)
Family name, Initials. (year).
Title. Place of publication: Publisher.
Family name, Initials. and Family
name, Initials. (year).
Title. Place of
publication: Publisher.
Family name, Initials., Family name,
Initials. and Family name, Initials [can
be discretionary to include more than
first author] (year).
Title. Place of
publication: Publisher.
Silverman, D. (2007).
A very short,
fairly interesting and reasonably cheap
book about qualitative research
.
London: Sage.
Berman Brown, R. and Saunders, M.
(2008).
Dealing with statistics: What
you need to know.
Maidenhead: Open
University Press.
Millmore, M., Lewis, P., Saunders, M.,
Thornhill, A. and Morrow, T. (2007).
Strategic human resource
management: Contemporary Issues.
Harlow: FT Prentice Hall.
(other than first
edition)
As for ‘Book (first
edition)’
(Anderson
et al.
2014)
Family name, Initials. and Family
name, Initials. (year).
Title. (# edn).
Place of publication: Publisher.
Anderson, D.L., Sweeney, D.J.,
Williams, T.A., Freeman, J. and
Shoesmith, E. (2014).
Statistics for
Business and Economics.
(3rd edn).
Andover: Cengage Learning EMA.
(edited) As for ‘Book (first
edition)’
(Saunders
et al.
2010)
Family name, Initials. and Family
name, Initials. (eds.) (year).
Title. Place
of publication: Publisher.
Saunders, M.N.K, Skinner, D.,
Gillespie, N., Dietz, G. and Lewicki,
R.J. (eds.) (2010).
Organizational
trust: a cultural perspective.
Cambridge: Cambridge University
Press.

To cite In the text In the references/bibliography
General format Example General format Example
(e-book)
As for ‘Book (first
edition)’
(Saunders 2013) Family name, Initials. (year).
Title.
[name of e-book reader]. Place of
publication: Publisher.
Saunders, J.J. (2013).
The Holocaust:
History in an Hour
[Kindle e-book].
London: William Collins.
(chapter in an edited
book)
(Chapter author
family name year)
(King 2012) Family name, Initials. (year). Chapter
title. In Initials. Family name and
Initials. Family name (eds)
Title. Place
of publication: Publisher. pp. ###–###.
King, N. (2012). Doing Template
Analysis. In G. Symon and C. Cassell
(eds)
Qualitative Organizational
Research
. London: Sage. pp. 426-50.
Dictionaries and other reference books
(where author known)
As for ‘Book (first
edition)’
(Vogt and Johnson
2011)
Family name, Initials. (year).
Title. (# edn). Place of Publication:
Publisher. pp. ###–###.
Vogt, W.P. and Johnson, R.B. (2011).
Dictionary of statistics and
methodology: a nontechnical guide for
the social sciences
. (4th edn).
Thousand Oaks, CA: Sage. pp. 31–2.
(where no author or
editor)
(Publication title
year)
(
The right word at
the right time
1985)
Publication title. (year). (# edn). Place
of Publication: Publisher. pp. ###–###.
The right word at the right time.
(1985). Pleasantville, NY: Readers
Digest Association. pp. 563–4.
Reports As for ‘Book (first
edition)’
(Gray
et al. 2012) Family name, Initials. and Family
name, Initials. (year).
Title. Place of
publication: Publisher.
Gray, D.E., Saunders M.N.K. and
Goregaokar, H. (2012).
Success in
challenging times: Key lessons for UK
SMEs
. London: Kingston Smith LLP.
(no named author) (Originator name or
Publication title
year)
(Mintel Marketing
Intelligence 2008)
Originator name
or Publication title.
(year).
Title. Place of publication:
Publisher.
Mintel Marketing Intelligence. (2008).
Designerwear: Mintel marketing
intelligence report.
London: Mintel
International Group Ltd.
(continues)
To cite In the text In the references/bibliography
General format Example General format Example
(online)
As for ‘Book (first
edition)’
(Thorlby
et al. 2014) Family name, Initials. and Family
name, Initials. (year).
Title of report.
Available at http://www.
remainderoffullInternetaddress/
[Accessed day month year].
Thorlby, R., Smith, J., Williams, S. and
Dayan, M. (2014).
The Francis Report:
One year on
. Available at: http://
www.nuffieldtrust.org.uk/sites/files/
nuffield/publication/140206_the_
francis_inquiry.pdf [Accessed 20 Mar.
2014].
Journal articles
(print or facsimile)
As for ‘Book (first
edition)’
(Rojon
et al. 2011) Family name, Initials. and Family
name, Initials. (year). Title of article.
Journal name. Vol. ##, No. ##,
pp. ###–####.
Rojon, C., McDowall, A. and Saunders,
M.N.K. (2011). On the experience of
conducting a systematic review in
industrial, work and organizational
psychology: Yes, it is worthwhile.
Journal of Personnel Psychology.
Vol. 10, No. 3, pp. 133–8.
(forthcoming
published online first
as facsimile)
As for ‘Book (first
edition)’
(Saunders and Rojon
2014)
Family name, Initials. and Family
name, Initials. (year). Title of article,
Journal name. Available at full doi or
Internet address [Accessed day month
year].
Saunders, M.N.K. and Rojon, C.
(2014) There’s no madness in my
method: explaining how your
coaching research findings are built
on firm foundations.
Coaching:
An International Journal of
Theory, Research and Practice.
Available at DOI:
10.1080/17521882.2014.889185
[Accessed 6 March 2014].
Table A1.1 Using the Harvard style to reference (continued)
To cite In the text In the references/bibliography
General format Example General format Example
Magazine articles
As for ‘Book (first
edition)’
(Saunders 2004) Family name, Initials. and Family
name, Initials. (year). Title of article.
Magazine name. Vol. ##, No. ##
(
or Issue or day and/or month),
pp. ###–###.
Saunders, M. (2004). Land of the long
white cloud.
HOG News UK. Issue 23,
Oct. pp. 24–6.
(no named author) (Originator name or
Publication name
year)
(
People Management
2014)
Originator name
or Publication name.
(year). Title of article.
Magazine name.
Vol. ##, No. ## (
or Issue or day and/or
month), pp. ###–###.
People Management. (2014).
Efficiency rule was misused.
People
Management
. Mar. p. 17.
News items including newspapers and online news
(article)
As for ‘Book (first
edition)’
(Frean 2014) Family name, Initials. and Family
name, Initials. Title of article.
Newspaper name, day month year,
p. ###.
Frean, A. Credit Suisse bankers
‘assisted tax evasion’.
The Times.
27 Feb. 2014, p. 35.
(article no named
author)
(Newspaper name
year)
(
The Times 2014) Newspaper name. Title of article, day
month year, p. ##.
The Times. Budweiser’s early win,
27 Feb. 2014, p. 33.
(article published
online)
As for other News
articles
(Rankin 2014) Family name, Initials. and Family
name, Initials. Title of article.
Newspaper name, day month year.
Available at http://www.fullInternetaddress/ [Accessed day month
year].
Rankin J. Record number of women
make 28th annual Forbes’ billionaires
list.
The Guardian. 4 Mar. 2014.
Available at http://www.theguardian.
com/business/2014/mar/03/recordnumber-women-forbes-28thbillionaires-list.html?src=linkedin
[Accessed 4 Mar. 2014].
(continues)
To cite In the text In the references/bibliography
General format Example General format Example
(article from electronic
database)
As for other News
articles
(Anderson 2009) Family name, Initials. and Family
name, Initials. Title of article.
Newspaper name, day month year,
p. ### (if known). [Accessed day
month year from Database name].
Anderson, L. How to choose a
Business School.
Financial Times,
23 Jan. 2009. [Accessed 20
Mar. 2010 from ft.com].
(article from news
web site)
As for other News
articles
(Gordon 2014) Family name, Initials. and Family
name, Initials. Title of article.
News
web site
, day month year. Available at
http://www.full-Internetaddress/
[Accessed day month year].
Gordon, O. Keeping crowdsourcing
honest. Can we trust the reviews? BBC
News, 14 Feb. 2014. Available at:
http://www.bbc.co.uk/news/
technology-26182642 [Accessed
4 Mar. 2014].
Websites (Source organisation
year)
(European
Commission 2014)
Source organisation. (year).
Title of
site or page within site.
Available at
http://www.
remainderoffullInternetaddress/
[Accessed day month year].
European Commission. (2014).
Eurostat – structural indicators.
Available at http://epp.eurostat.ec.
europa.eu/portal/page/portal/
structural_indicators/introduction
[Accessed 5 Mar. 2014].
Conference papers
(published as part of
proceedings)
As for ‘Book (first
edition)’
(Saunders 2009) Family name, Initials. and Family
name, Initials. (year). Title of paper.
In Initials. Family name and Initials.
Family name (eds)
Title. Place of
publication: Publisher. pp. ###–###.
Saunders, M.N.K. (2009). A real world
comparison of responses to distributing
questionnaire surveys by mail and
web. In J. Azzopardi (Ed.)
Proceedings
of the 8th European Conference on
Research Methods in Business and
Management.
Reading: ACI,
pp. 323–30
Table A1.1 Using the Harvard style to reference (continued)
To cite In the text In the references/bibliography
General format Example General format Example
(unpublished)
As for ‘Book (first
edition)’
(Saunders et al. 2010) Family name, Initials. and Family
name, Initials. (year).
Title of paper.
Unpublished paper presented at
‘Conference name’. Location of
conference, day month year.
Saunders, M.N.K., Slack, R. and
Bowen, D. (2010).
Location, the
development of swift trust and
learning: insights from two doctoral
summer schools
. Unpublished paper
presented at the ‘EIASM 5th
Workshop on Trust Within and
Between Organizations’. Madrid,
28–29 January 2010.
Film, Video, TV, Radio, Downloads
(Television or radio
programme)
(Television or radio
programme title
year)
(
Today Programme
2014)
Programme title. (year of production).
Transmitting organisation and nature
of transmission, day month year of
transmission.
The Today Programme. (2014). British
Broadcasting Corporation Radio
broadcast, 11 Apr. 2014.
(Video download
e.g. YouTube)
(Company name or
Family name year)
(Miller 2008) Company name
or Family name,
Initials. (year). Title of audio download
.
YouTube.
Available at http://www.
remainderoffullInternetaddress/
[Accessed day month year].
Miller, L. (2008). Harvard style
referencing made easy.
YouTube.
Available at http://www.youtube.
com/watch?v=RH1lzyn7Exc [Accessed
5 Mar. 2014].
Notes: Where date is not known or unclear, follow conventions outlined towards the end of Table A1.2. Be warned, most lecturers consider citing of lectures as ‘lazy’ scholarship.
246 Appendix 1 How to reference
The American Psychological Association style or APA style is a variation on the author–
date system. It is explained in full in the latest edition of the American Psychological
Association’s (2009)
Concise Rules of the APA Style, which is likely to be available for reference in your university’s library. There are small but significant differences between the
Harvard and APA styles, and many authors adopt a combination of the two styles. The
key differences are outlined in Table A1.2.
The American Psychological Association (APA) style
Table A1.2 Key differences between Harvard and APA styles of referencing
Harvard style APA style Comment
Referencing in the text
(Lewis 2001) (Lewis, 2001) Note: punctuation
(McDowall and Saunders
2010)
(McDowall & Saunders, 2011) Note: ‘&’, not ‘and’
(Altinay
et al. 2014) (Altinay, Saunders & Wang,
2014)
For first occurrence if three to
five authors
(Millmore
et al. 2007) (Millmore et al., 2007) For first occurrence if six or
more authors; note punctuation
and use of italics
(Tosey
et al. 2012) (Tosey et al., 2012) For subsequent occurrences of
two or more authors; note
punctuation and use of italics
Referencing in the list of references or bibliography
Berman Brown, R. and
Saunders, M. (2008).
Dealing with statistics:
What you need to know.
Maidenhead: Open
University Press.
Berman Brown, R. & Saunders,
M. (2008).
Dealing with
statistics: What you need to
know.
Maidenhead: Open
University Press.
Note: use of ‘and’ and ‘&’
Varadarajan, P.R. (2003).
Musings on relevance and
rigour of scholarly research
in marketing.
Journal of the
Academy of Marketing
Science
. Vol. 31, No. 4,
pp. 368–376. [Accessed
6 Apr. 2010 from Business
Source Complete].
Varadarajan, P.R. (2003).
Musings on relevance and
rigour of scholarly research in
marketing.
Journal of the
Academy of Marketing Science,
31(4), 368–376. doi:
10.1177/0092070303258240
Note: Volume, part number and
page numbers;
DOI (digital object identifier)
number given in APA. Name of
database not given in APA if
DOI number given;
Date accessed site not included
in APA.

Altinay, L., Saunders, M.N.K. and Wang, C. (2014) The influence of culture on trust
judgments in customer relationship development by ethnic minority small businesses.
Journal of Small Business Management, 52(1), 59–78.
Barclaycard (2016). The emergence of ‘serial returners’ – online shoppers who habitually over order and take advantage of free returns – hinders growth of UK businesses.
Available at: https://www.home.barclaycard/media-centre/press-releases/emergence-of-serial-returners-hinders-growth-of-UK-businesses.html [Accessed 20
October 2016].
Baruch, Y. and Holtom, B.C. (2008). Survey response rate levels and trends in organizational research.
Human Relations, 61(8), 1139–60.
Becker, H.S. (2007).
Writing for Social Scientists (2nd ed.). Chicago: University of Chicago.
Beynon, H. (1973).
Working for Ford. London: Allen Lane.
Brinkmann, S. and Kvale, S. (2014).
InterViews: Learning the Craft of Qualitative Research
Interviewing
(3rd ed.). Los Angeles, CA: Sage.
Cassell, C. (2015).
Conducting Research Interviews for Business and Management Students.
London: Sage.
Clough, P. and Nutbrown, C. (2012).
A Student’s Guide to Methodology (3rd ed.).
London: Sage.
Corbin, J. and Strauss, A. (2008)
Basics of Qualitative Research: Techniques and Procedures
for Developing Grounded Theory
(3rd ed.). London: Sage.
Creswell, J. (2008).
Qualitative, Quantitative, and Mixed Methods Approaches (3rd ed.).
Thousand Oaks, CA: Sage.
Department for Work and Pensions (2014).
Small Employer Recruitment Practices:
Qualitative Research into How Small and Medium-Sized Enterprises Select Candidates for
Employment.
Report No 855, July 2014.
Dragons’ Den (2014). Scott Cupit and Swing Patrol on BBC’s
Dragon’s Den. YouTube.
Available at: https://www.youtube.com/watch?v=lHYnxtR1u0I [Accessed 8
November 2016].
Dunsby, M (2016).
Dragons’ Den Success Stories: Swing Patrol. Available at: http://startups.
co.uk/dragons-den-success-stories-swing-patrol/ [Accessed 8 November 2016].
Edelman (2016)
2016 Edelman Trust Barometer Global Report. Available at: http://www.
edelman.com/insights/intellectual-property/2016-edelman-trust-barometer/globalresults/ [Accessed 28 October 2016].
Ekinci, Y. (2015).
Designing Research Questionnaires for Business and Management Students,
London: Sage.
Eurostat (2014).
Methodological Manual for Tourism Statistics Version 3.1. Luxembourg:
Publications Office of the European Union.
Eurostat (2016).
Tourism: Main Tables. Available at: http://ec.europa.eu/eurostat/web/
tourism/data/main-tables [Accessed 1 November 2016].
Bibliography
248 Bibliography
Eurostat (no date). Europe in Figures. Available at: http://ec.europa.eu/eurostat/statisticsexplained/index.php/Europe_in_figures_-_Eurostat_yearbook [Accessed 1
November 2016].
Eurostat Press Office (2016). Organic crop farming on the rise in the EU.
Eurostat News
Release 208/2016.
Available at: http://ec.europa.eu/eurostat/documents/
2995521/7709498/5-25102016-BP-EN.pdf/cee89f9e-023b-4470-ba23-61a9893d34c8
[Accessed 21 November 2016].
Fenwick, D. and Denman, J. (1995) The monthly unemployment count: change and
consistency,
Labour Market Trends, November, 397–400.
Ghazali, S. (2011). The influence of socialization agents and demographic profiles on
brand consciousness,
International Journal of Management and Marketing Research,
4(1), 19–29.
Gov.UK (2015).
Data Protection. Available at: https://www.gov.uk/data-protection/
the-data-protection-act [Accessed 15 June 2016].
Hair, J., Black, W., Babin, B. and Anderson, R. (2009).
Multivariate Data Analysis
(7th ed.). London: Pearson.
Harvard College Library (2011).
Interrogating Texts: 6 Reading Habits to Develop in Your
First Year at Harvard
. Available at: http://bsc.harvard.edu/files/interrogating_texts_
six_reading_habits_to_develop_in_your_first_year_at_harvard.pdf [Accessed 12
October 2016].
Henley, J. (2016). Sweden leads the race to become cashless society
The Guardian, 4 June
2016.
Hofstede, G., Hofstede, G.J. and Minkov, M. (2010).
Culture and Organizations: Software
of the Mind
(3rd ed.). Columbus OH: McGraw-Hill.
House of Commons Environment, Food and Rural Affairs (2015).
Rural Broadband and
Digital-Only Services, Seventh Report of Session 2014–15
. London: The Stationery Office
Limited.
Hudson, S., Roth, M.S., Madden, J.T. and Hudson, R. (2015). The effect of social media
on emotions, brand relationship quality and word of mouth: An empirical study of
music festival attendees.
Tourism Management, 47(1), 68–76.
Kipling, R. (1902, reprinted 2007).
A Collection of Rudyard Kipling’s Just So Stories. London:
Walker Books.
Leenders, M.A.A.M. (2010). The relative importance of the brand of music festivals: a
customer equity perspective.
Journal of Strategic Marketing, 18(4), 291–301.
Levitas, R. (1996).
Interpreting Official Statistics. London: Routledge.
Lewin K. (1945). The Research Centre for Group Dynamics at Massachusetts Institute of
Technology
. Sociometry, 8(2), 126–36.
McSweeny, B. (2002). Hofstede’s model of national cultural differences and their consequences: a triumph of faith – a failure of analysis. Human Relations, 55(1), 89–118.
Mills, C.W. (1970). On intellectual craftsmanship. In
The Sociological Imagination.
London: Pelican.
Office for National Statistics (2016).
Balance of Payments time series dataset. Available at:
https://www.ons.gov.uk/economy/nationalaccounts/balanceofpayments/datasets/
balanceofpayments [Accessed 20 November 2016].

Bibliography 249
Office for National Statistics (2016). Labour Force Survey. Available at: https://www.ons.
gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/uklabourmarket/october2016 [Accessed 20 October 2016].
Office for National Statistics (2016).
Retail Industry. Available at: http://www.ons.gov.
uk/businessindustryandtrade/retailindustry#datasets4 [Accessed 31 October 2016].
Oxford Brookes University (2016).
Ethical Standards for Research Involving Human
Participants: Code of Practice.
Available at: https://www.brookes.ac.uk/Documents/
Research/Policies-and-codes-of-practice/ethics_codeofpractice/ [Accessed 11
November 2016].
Pricewaterhouse Coopers, University of Southern California and the London Business
School (2013).
PwC’s NextGen: A Global Generational Study. Available at: http://www.
pwc.com/us/en/people-management/publications/nextgen-global-generationalstudy.html [Accessed 27 October 2016].
Raimond, P. (1993).
Management Projects. London: Chapman & Hall.
Saunders, M.N.K. (2011). Choosing research participants. In C. Cassell and G. Symons
(eds),
The Practice of Qualitative Organizational Research: Core Methods and Current
Challenges
. London: Sage.
Saunders, M.N.K. (2012). Web versus mail: the influence of survey distribution mode
on employees’ response.
Field Methods, 24(1) 56–73.
Saunders, M.N.K., Gray, D. and Goregaokor, H. (2014). SME innovation and learning:
the role of networks and crisis event’
European Journal of Training and Development
38(1/2), 136–49.
Saunders M., Lewis, P. and Thornhill, A. (2016).
Research Methods for Business Students.
(7th ed.). Harlow: FT Prentice Hall.
Saunders, M.N.K. and Townsend, K. (2016). Reporting and justifying the number of
interviews participants in organisation and workplace research.
British Journal of
Management
, 27(4), 836–52.
Smith, A. (1904).
An Inquiry into the Nature and Causes of the Wealth of Nations (5th ed.).
London: Methuen and Co.
Sutton, R. and Staw, B. (1995). What theory is not.
Administrative Science Quarterly, 40(3),
371–84.
University of Plymouth (2016) School of Marine Science and Engineering, Project
Marking Scheme. Available at: http://www.tech.plym.ac.uk/sme/mingproject/
MScheme910.pdf [Accessed 12 June 2016].
Van Maanen, J., Sørensen, J.B. and Mitchell, T.R. (2007) ‘The interplay between theory
and method’,
Academy of Management Review, Vol. 32, No. 4, pp. 1145–54.
Wallace, M. and Wray, A. (2016)
Critical Reading and Writing for Postgraduates (3rd ed.).
London: Sage.
YouGov (2016)
Panel Methodology. Available at: https://yougov.co.uk/about/panelmethodology/ [Accessed 7 November 2016].
abduction 113–14
abstracts 44–5
action research 122–4
ad hoc survey 88
Adobe Reader 44
American Psychological Association
(APA) referencing style 239, 246
Harvard referencing style
vs. 246
analysis of variance 195
anonymity, for research subjects 82, 231
archival research 126
articles 31, 35–6, 38
abstracts of 44–5
full text of 44
non-academic 115
peer review of 37, 55
assessment criteria, for research 17
assumptions
axiological 107
epistemological 106–7
ontological 106
author-date referencing systems 50, 56, 239
axiological assumptions 107
background thinking 6
bar graphs 186–7
BBC 90
Beynon, Huw 125–6
Bing search engine 39, 40, 100
Blackwell Reference Online database 41
Boolean operators 43
brainstorming 9–10
Brinkman, S. 163
Business Source Premier database 39, 41,
44, 115
CAQDAS (computer aided qualitative
data analysis software) 202–3, 204
case study 121–2
categorical data 180–1, 184–5, 192,
195, 212
categorising data 208–9
category questions 150
cause-and-effect relationship 201–2
Census of Population 88
censuses 88
central tendency 191–2
chi-square test 195, 196–7
clarity, in research 18
Clough, P. 20, 21
Code of practice 76
coefficient of determination 196, 201–2
cognitive access 61
compiled data 86–7
concept mapping 9–10
conference proceedings 38
confidentiality 78, 82, 219
confidentiality agreements 67–8
construct validity 149
consumer focus groups 115
contacts, use of 62–3
content validity 149
continuous data 182
continuous surveys 88
convenience sampling 147–8
course assignments, past 7
covering letter, structure of 155
Cresswell, J. 26–7
critical case purposive sampling 146
critical realism 108
criticism, receipt of 73
cross-sectional study 129, 130
data
access to 15–16
categorisation of 208–9
clarity of 18
consistency in 132–3
defined 25
dispersion of 191–2
primary
see primary data
qualitative
see qualitative data
quantitative
see quantitative data
reliability of 18
secondary
see secondary data
sensitivity of 60
Index
Index 251
techniques of 35–6
and theory 25
types of 78, 180–3
units of 208–9
validity of 18
data analysis 3, 5, 36–7, 179–212
qualitative 202–12
quantitative 183–202
data collection 3, 137–8
causes for 99
combining methods of 128–9
harm caused by 79–80
and honesty 80
from Internet 80
method used for, assessment of 99–100
observation and 168–75
plausibility of methods 133
questionnaires and 148–58
sample selection 138–48
techniques 35–6, 122
data matrix 183–4
data presentation
for proportions 188–9
for relationships between variables
189–90
for summarisation 185
for trends 189
for values 186–8
Data Protection Act 1988 81–2
data saturation 165
deadlines 73
deduction 112, 113–15, 209, 213
characteristics of 112
defined 112
induction
vs. 111
Delphi technique 13–15
descriptive data 182
descriptive studies 116–18
diagrams 26
direct questions 163
direct realism 108
discrete data 182
discussion, of ideas 11, 13
dispersion 191–2
documentary secondary data 89–90
DOI (digital object identifier) 239
EBSCO
see Business Source Premier
database
email, structure of 155
Emerald Insight database 39, 41,
42–4, 115
employee survey 11–12
employers, production of reports for
11–12
enthusiasm, for research topic 15
epistemological assumptions 106–7
ethical standards 75–83
breaching of 78
code of practice for 76
defined 75
ethnography 124–6
Excel software 142, 180
experiment 119–20
explanatory studies 118
exploratory studies 115–16
external organisations
assurances of confidentiality 67–8
establishment of credibility with 68
gaining access to 59–61
external validity 134
extreme case purposive sampling 146
face-to-face interviews 157–8, 165–6
filter questions 154
flexibility, in research 16–17
footnotes systems 50
Ford Motor Company 126
fresh insights 18
Galaxy Note 7 109
gatekeepers 77, 82
defined 61
gender, usage in research proposal 231
Ghazali, Zaharah 199–200
Goldilocks test 20
Google 32, 39, 40, 100
Google Scholar database 41
grammatical errors, in research proposal
229
grand theories 26–7
grounded theory 124
Harvard College Library 48
Harvard referencing style 239–45
APA referencing style
vs. 246
heterogeneous purposive sampling 146
histograms 187–8
Hofstede, G.J. 48
homogeneous purposive sampling 146

252 Index
hypotheses 25, 108, 194, 201 see also null
hypothesis
hypothesis testing 194
IBM SPSS Statistics 37, 94, 152, 180, 193,
196
illegitimate codes 184
illogical relationships 184
in-depth interviews 127
indirect questions 163
induction 213
deduction
vs. 111
defined 113
informant 168
information gateways 100, 101
informed consent 76–7
internal validity 134
Internet
for conducting semi-structured and
unstructured interviews 167–8
and data collection 80
questionnaires 156–7
role in exploratory study 115
as source of information 37, 39, 55
Internet-mediated observation 172–5
interpretivism 109
interval data 181
interviews
face-to-face 157–8, 165–6
in-depth 127
participants in 76–7, 162
pilot testing for 160, 164–5
semi-structured
see semi-structured
interviews
structured 116, 148
telephonic 157–8, 166–7
transcripts 182–3
unstructured
see unstructured interviews
jargon, avoidance of 229
journals 7, 35
peer review of 37, 55
JSTOR online database 40, 41
Kendal’s rank correlation coefficient 195,
197–8
Key Note 91
keyword searches 39
Kipling, Rudyard 216
Kvale, S. 163
learning objectives 24
learning outcomes 74
Lewin, Kurt 26
Lewis, Philip J. 26, 47, 49, 105, 123, 150
libraries 37, 55
line graphs 189
list of references 49
list questions 150
literature
searching and obtaining 39–45
types of 37–8
usefulness of 46–7
literature reviews 7
and data collection techniques 35–6
defined 32
downloading of articles and abstracts 44
importance of 34–7
keywords and phrases in 40
and online databases 40–2
process of 32
purpose of 32–3
research projects based on 34
and research topic selection 34–5
search techniques 42–4, 55
secondary data and 35, 55
and shopping catalogue review 53–5
structure of 52–3
longitudinal studies 129, 130
marking scheme guides 75
matrix questions 151
methodology, of research 35, 133, 224–6
Microsoft Excel 142, 180
middle-range theories 26, 27
Mills, C. Wright 227
Minkov, M. 48
motivation, for research 70–1
multiple source secondary data 91–3
narrative inquiry 126–7
National Lottery Lotto game 142
netiquette 156
news media 8–9
newspapers 38, 85, 91
Nexis online database 41
nominal data 182
non-directive interviews
see unstructured
interviews
non-numeric data 85, 86
see also
qualitative data
Index 253
non-probability sampling 140, 141–2
convenience sampling 147–8
defined 141
purposive sampling 145–6
quota sampling 144–5
volunteer sampling 146–7
non-text data 182–3, 212
non-text documentary secondary
data 90
normal distribution 192
note taking, of ideas 10
null hypothesis 194, 195, 201
see also
hypotheses
numerical data 85, 181–2, 184–5, 192,
195–6, 212
Nutbrown, C. 20, 21
observations
conducting 172–5
data collection using 168–75
Internet-mediated 172–5
in person 173–4
preparation for 169–72
scenarios for 171
structured 168–9, 170
unstructured 169, 170
using videography 175
observer bias 135
observer effect 79
observer error 135
’one off ’ survey 88
online databases 32, 37, 39–44
online first 232
ontological assumptions 106
open questions 150, 155
ordinal data 181
organisational skills, development
of 58
Oxford Brookes University 76
paired
t-test 195
participants, in research
anonymity of 82, 231
confidentiality for 78, 82, 219
consent from 76–7, 162
defined 158
vulnerability of 78
past projects, usage of 6–7, 40
Pearson’s product moment correlation
coefficient 196, 200
peer review 37
personal experience 3
personal pronouns, usage of 230
physical access 60
pie charts 188–9
pilot testing
defined 156
for interviews 160, 164–5
plagiarism 32, 48, 56
plausible theory 114
populations, for sampling 138–42
Portable Document Format (PDF) 44
positivism 107–8
postal questionnaires 157
postmodernism 110
pragmatism 111
preliminary studies 12–13
primary data
availability of 16
defined 85, 101
harm caused by collection of
79–80
probability sampling 140–1
defined 140
simple random sampling 142–3
stratified random sampling 143–4
systematic sampling 143
progress reports 73
project literature
links to 17
use of 12
project objectives 18
proposition development 204–8
public domain, data in 93
purposive sampling 145–6
defined 145
varieties of 146
QSR International’s NVivo software
202–3
qualitative data 102, 180, 212
analysis of 202–12, 213
confidentiality of 78
defined 86
types of 182–3
Qualtrics software 152
quantitative data 78, 102, 212
analysis of 183–202
defined 86
preparation of 183–4

254 Index
presentation of 184–90
secondary 97
statistics and 190–202
types of 180–2
quantity questions 151
questionnaires 3, 35–6, 88, 100, 116,
120–1, 128
and data collection 148–58
defined 148
designing 152–5
distribution of 156–8
Internet, distribution of 156–7
pilot testing of 156, 160
postal, distribution of 157
purpose of 155
web-based software for 152
questions
designing 149–52
research 221–2
in semi-structured/unstructured
interviews 163–4
types of 150–1, 163
quota sampling 144–5
quotations, from literature 230
ranked data 181
ranking questions 151
rating questions 151
ratio data 181
raw data 86
reactivity 79
realism 108–9
see also critical realism;
direct realism
referencing 25, 49–51, 227, 239–46
Harvard style 239–45
regression equation 196, 201–2
regular surveys 88
relationships. statistical examination of
197–200
reliability, of research 135
research
approaches to 111–15
conclusions 131–5
flexibility in 16–17
motivation for 70–1
noting ideas for 10
objectives 224
professionalism in 65–6
reliability of 135
scope of 16–17
sources for 47–8
validity of 134
research design, selection of 104–35
research ethics
see ethical standards
research findings
believability of 131–5
consistency with data 132–3
logical flow of 132
research ideas
examples of 21
generation of 5–11
research objectives 18–27, 133
’research onion’ 105
research philosophy 106–11
defined 106
interpretivism 109
positivism 107–8
postmodernism 110
pragmatism 111
realism 108–9
research process, management of 58–83
by self-management 69–71
supervisors and 71–4
in universities 74–5
research projects 1, 4, 34, 69
research proposal 215–36
for assembling research ideas 216–17
components of 232
content of 220–7
importance of 216–20
judgement, assessment criteria for 231–6
purpose of 216
style 227–31
viability of 232–6
and viability of research 218–19
writing 220
research questions 18–27, 133
research resources 3, 15, 227
research topics, selection of
characteristics of good topics 15–18
difficulties in 4–5
importance of 2–4
role of literature review in 34–5, 222
respondents
consent from 76–7
defined 148
vulnerability of 78
reverse test 132
Russian doll principle 21
quantitative data (
continued)
Index 255
sample, selection of 138–48
non-probability sampling techniques
144–8
probability sampling techniques 142–4
reasons for 139
techniques of 139–42
sample/sampling
defined 138
non-probability 140, 141–2
population for 138–42
probability 140–1
selection of
see sample, selection of
sampling frame 139, 143–4
Samsung 109
Saunders, Mark N.K. 26, 36, 47, 49, 105,
123, 141, 150
scatter graph 189–90, 201
scatter plot 189–90
search engines 39, 40
search terms 39–40
secondary data 85–103, 212
causes for using 102
defined 85, 101
documentary 89–90
forms of 86–7, 102
and literature reviews 35, 55
multiple source 91–3
pitfalls of using 96–8, 102
potential of 93–6
qualitative analysis of 204
quantitative 97
reanalysis of 116
relevance of 16, 98–9
searching of 100–1
suitability of 98–100, 102
survey 88–9
unauthorised use of 80
usage of 102–3
self-selection sampling 147
semi-structured interviews 158–68
conducting 165–8
defined 158
face to face interview 165–6
internet-mediated 167–8
preparation for 160–2
questions in 163–4
telephonic interview 166–7
sentences, in research proposal 228–9
significance testing 194
simple random sampling 142–3
skills, of researchers 3
sloppy referencing 50
SMART test 24
Smith, Adam 221
snapshot multiple source secondary
data 91
snowball sampling 147
social actors 109
Spearman’s rank correlation coefficient
195, 197–8
specialist knowledge 3
spelling errors, in research proposal
229
SPSS Statistics (IBM) 37, 94, 152,
180, 193
statistics, and quantitative data 190–202
Staw, B. 24
stratified random sampling 143–4
structured interview 148
structured observation 168–9, 170
subject bias 135
subject error 135
substantive theories 26, 27
summarising 48
supervisors 5, 71–4
survey secondary data 88–9
SurveyMonkey software 152
surveys 11–12, 120–1
Sutton, R. 24
systematic sampling 143
t-test 195, 199–200
telephonic interview 157–8, 166–7
tense, usage in research proposal 230–1
testable propositions 205–8
text data 182, 212
text documentary secondary data 90
textbooks 7, 37, 38, 55
theory 24–7
defined 24
development, approaches to 111–15
grand 26–7
middle-range 26, 27
as research element 5
substantive 26, 27
thinking, and research topic selection 6
Thornhill, A. 26, 47, 49, 105, 123, 150
time management 69–70
time-series multiple source secondary
data 92–3

256 Index
timescale 69, 226
topicality, of research 16
trade magazines 35
trends, in data 191, 193–4
triangulation process 128–9
typical case purposive sampling 146
units of data 208–9
University of Plymouth 75
unstructured interviews 158–68
conducting 165–8
defined 159
face to face interview 165–6
internet-mediated 167–8
preparation for 160–2
questions in 163–4
telephonic interview 166–7
unstructured observation 169, 170
useful sources 47–8
validity
construct 149
content 149
of research 134
Vancouver referencing 50
variables 25, 192
interdependences between 194–202
videography
defined 173
observations using 175
volunteer sampling 146–7
Wallace, M. 49
Wikipedia 32, 40, 41–2
Wiley Online database 41
Wray, A. 49
writing style, research proposal 227–31
YouTube 90

Front cover photograph © 2017 Mark Saunders: www.pearson-books.com
Herefordshire Beacon, Malvern Hills
Are you looking for advice about your research project?
Are you not sure where to start, how to choose your research topic
or how to write your research proposal? Have you got data and don’t
know what to do with it next?
If you need any help with your research project or dissertation then this fully revised and
updated second edition is ideal for you.
Doing Research in Business and Management
brings research methods to life, covering critical literature review projects, projects
using existing data and those projects where you collect your own data. It explains
how to write your research proposal and complete your entire research project.
The concise and easy-to-read style combined with clear tables and diagrams makes it
highly accessible. ‘Defi nition’ boxes explain key terms and ‘Research in practice’ boxes
show how research is used in real life, making this the book you need to successfully
complete your project.
Mark Saunders and Philip Lewis are the best-selling authors of Research Methods for
Business Students
, which has now sold over 350,000 copies worldwide. Mark Saunders
is Professor of Business Research Methods at The Birmingham Business School,
University of Birmingham, and
Philip Lewis is a former Principal Lecturer at The Business
School, University of Gloucestershire.
Don’t forget to visit
www.pearsoned.co.uk/saunders for online tutorials on research
software, such as SPSS, and where you can search the Internet more e ciently and
e ectively with our Smarter Online Searching Guide!
Start your project with confi dence and complete it with success!
New to the second edition:
latest sources of online secondary data,
extended discussion of research philosophies and theory development,
using observation,
how to reference.