1
SOCI4022 Assignment 1
Due date Feb 27-2023
Technical notes for submission:
Your assignment should consist of two documents:
a PDF document where you present and discuss tables and figures, as well as where you
offer background information and references, all according to APA standards and
guidelines.
A syntax file, which is a modified version, with your notes and additional analyses, of the
one I posted.
I will not accept the SPSS Output as a display of your results or PDF submissions with
unformatted graphs and tables (copy-paste from output). You will be evaluated for how
you selected the important parts of the SPSS output, how you formatted them into
clean and informative tables or graphs and how well you discussed them.
When you discuss the results, blend rigor in discussing statistics, and a clear narrative
about real implications of the results. Build up a coherent and engaging story that you
must communicate effectively.
Reminders of helpful tools on data visualization:
Storytelling with data.
http://www.storytellingwithdata.com/blog
http://www.storytellingwithdata.com/chart-guide
How to Tell a Great Story.
https://hbr.org/2014/07/how-to-tell-a-great-story
Ebook available at the HIL:
Knaflic, C. (2015). Storytelling with data : A data visualization guide for business
professionals. Hoboken, New Jersey: Wiley. (2015).
2
Instructions (please read carefully)
Assignment 1 is about describing data with basic statistics and graphs, comparing
groups, and discussing results in meaningful ways.
We are using PISA 2018 data, and I created already a dataset for you to use in this
Assignment 1 folder called PISA2018_CAN.
In this assignment, you will study construct two measures of students wellbeing, sense
of belonging and bullying, and will relate them to students’ test scores in reading.
I share with you a syntax file that helps you with 80% of the analyses that are needed.
You can follow my train of thoughts in the syntax, and I annotated it as well.
Task one:
write a short paragraph, in bullet points, and explain what the analytical approach in the
syntax is. Each bullet point should explain what I do in the syntax and why the command
is useful; what it does, and why it does it.
[For e.g.: bullet 1: frequencies of four variables related to bullying-the goals are:
describe students’ behavior, look for anomalous response options, evaluate number of
missing data and consider their impact on the analyses.]
Task two:
the analyses start by requesting frequencies of variables describing the sample. It is your
job to describe the sample composition using my syntax and add an analysis by grade.
Describe the sample composition by gender and grade and show their distribution in the
sample. (½ page, plus table or graph, if needed.)
Describe students’ reading score distribution and their socioeconomic status. The syntax
offers you some visual and numerical tools to do that. (½ page plus table or graphs, if
needed.)
Look at how I constructed two indexes, one that measures sense of belonging and the
other that measures intensity of bullying experience. Describe their distributions in 15-
year old Canadian students. Use the statistics to build a narrative of learning and school
environment.
To assist you, I prepared box plots for you to discuss the distribution of these indexes,
by sex and by students’ socioeconomic status.
Improve the SPSS Output and transform the graphs and tables into powerful visual
narratives. Pay attention to details and use fine descriptive skills when you discuss what
the graphs tell us. Remember not to just speak about medians, but focus on the
distribution (symmetry, spread, interquartile range). As we are writing for also the
broader public, ensure that you explicitly embed the boxplots into a narrative about
learning and school environment for 15-year-old students in Canadian schools. Start
with discussing the results, and then connect them issues of inequalities and possibly
segregation.
3
I suggest that for Task 2 you write 5 pages, considering that at least 2 pages will be
occupied by graphs/tables.
Task three:
Move to PART 2 in the syntax. With cross tabs, I tried to describe how some groups of
students are more likely to be “poorer readers” than others.
Your responsibility is to add the appropriate percentages and test to the contingency
tables, and to calculate by hand and report in one single clean table Odds and Odds
Ratios. The “event” is negative- the student performs poorly in reading, so each
contingency (crosstabs) needs to be restructured to make correct calculations of odds
and OR. I used gender, low socioeconomic status, and being bullied as grouping
variables.
Present clean and formatted crosstabs.
Present the results for all the crosstabs (please deal with the right %-by row or column)
and compute Odds and Odds Ratios. Report in a clean table or in writing the odds and
OR for each crosstab.
Discuss inequalities and alarming issues that relate to learning issues in sub-populations
of students based on gender, socioeconomic status, and being bullied. You are welcome
to run additional analyses if you want to elaborate further.
This section should be about 2 pages, tables included. Try to condense the results into
one or two tables only.
Overall, your assignment should be about 8 pp long, use APA format, and no double
space. If you cite in the body of your text, add a reference list.