Formulation of a questionAssignment
• Questions can be related to an particular
observation or open ended:
– ‘Why are people irritated by virtual agents?’
– ‘How can I develop a faster compression algorithm for
video files?’
• Looking up and evaluating evidence from
– previous experiments
– personal scientific observations
– assertions
– the work of other scientists.
Formulation of a question
• If the answer is already known
– We can use a different question that builds on the
previous evidence
• Finding a good research question isn’t easy,
but the outcome of the research depends on
it. Take care with this step.
Hypothesis
• A conjecture, based on knowledge obtained
while formulating the question, that may explain
the observed behavior.
• Can be specific or broad:
– ‘Virtual agents annoy us because they are intrusive’
– ‘The application of graph theory can help us to
develop faster compression algorithms for video files’
– ‘Virtual agents have the potential to improve social
integration for the elderly’
Statistical Hypothesis
• A conjecture about some population.
– E.g. Students, Apple users, Managers.
• Null hypothesis
– Conjecture that the hypothesis is false
• Alternative hypothesis
– Desired outcome (result is better than they would be
by chance)
• A Scientific hypothesis must be falsifiable
– Can identify a possible outcome of an experiment that
conflicts with predictions deduced from the
hypothesis
Hypothesis
• Francis Crick and James D.
Watson hypothesized that
‘DNA has a helical
structure’.
Prediction
• Involves determining the logical consequences of
the hypothesis.
• One or more predictions are then selected for
further testing.
– The more unlikely that a prediction would be correct
simply by coincidence, then the more convincing it
would be if the prediction were fulfilled.
– Evidence is also stronger if the answer to the
prediction is not already known, due to the effects of
hindsight bias.
– Ideally, the prediction must also distinguish the
hypothesis from likely alternatives.
Hypothesis
• This prediction was
determined using the
mathematics of the helix
transform, which had been
derived by Cochran, Crick
and Vand (and
independently by Stokes).
• This prediction was a
mathematical construct,
completely independent
from the biological problem
at hand.
• If DNA had a helical
structure, its X-ray
diffraction pattern would
be X-shaped.
Testing
• To determine whether observations of the real
world agree with or conflict with the
predictions derived from an hypothesis.
– If they agree, confidence in the hypothesis
increases; otherwise, it decreases.
– Scientific controls, double blind testing etc.
– TBC in Experimental Methods section…
Analysis
• Determining what the results of the
experiment show and deciding on the next
actions to take.
– Incorporate results form other experiments.
– In the case of a population study – X2 or t-test.
– Null hypothesis is proven, develop new
hypothesis.
– No sufficient evidence, further experiments or
alternate predictions.
Analysis
• When Watson saw the
detailed diffraction
pattern, he immediately
recognized it as a helix.[
• Watson and Crick then
produced their model,
using this information
along with existing
information about
DNA’s composition and
molecular interactions
such as hydrogen
bonding.
Analysis
Process
Not all steps take place in every
scientific inquiry (or to the same
degree), and are not always in the
same order.
“Invention, sagacity, [and]
genius” are required at every step
[Whewell, 1837].
Process Guideline
1. Define a question
2. Gather information and resources (observe)
3. Form an explanatory hypothesis
4. Test the hypothesis by performing an experiment and
collecting data in a reproducible manner
5. Analyze the data
6. Interpret the data and draw conclusions that often
serve as a starting point for new hypothesis (i.e. back
to 3)
7. Publish results
8. Retest (frequently done by other scientists)

Replication
• Experiments are repeated to confirm the
results.
– If the results are not the same then there is a
problem with the experiment.
• It must be possible for others to repeat the
experiment.
– Requires a well documented procedure.
– Especially for significant or surprising results.
External Review
• Evaluation by experts.
• Peer review
– Unbiased anonymous criticism.
– Scientific journals, conferences.
– Better journal or conference ->
work perceived as being of a higher standard or
better quality.
Data Recording and Sharing
• Scientists must record all data very precisely in
order.
– reduce their own bias
– aid replication by others
• Must share data and samples that are difficult
to obtain.

Scientific Theories
• This allows scientists to gain an understanding of
reality, and later use that understanding to
intervene in its causal mechanisms (such as to
cure disease).
• The better an explanation is at making
predictions, the more useful it is, and the more
likely it is to be correct.
• The most successful explanations, which explain
and make accurate predictions in a wide range of
circumstances, are called scientific theories.
Falsification
• Scientific knowledge is closely tied
to empirical findings, and always remains subject
to falsification.
– No theory can ever be considered as completely
certain.
– If falsifying evidence is found.
• A new theory is proposed
• Or (more commonly) he original theory is modified
– The strength of a theory is related to how long it has
persisted without falsification of its core principles.
Subsumption
• Confirmed theories are also subject to
subsumption by more accurate theories.
– Newton’s laws (discovered to be a special case of)
-> theory of relativity (explains deflection of light
by gravity)
– Evolution
-> Evolutionary synthesis (evolution + genetics)
Beliefs and biases
• Pursuit of experimental control and
reproducibility diminishes the effects of
cognitive biases.

Principles of Project Management
Phases Cornerstones
Initiation
Planning
Execution
Monitoring
Closing
Milestone-based
Accountability
Managing Risk
Initiation & Planning of Projects
• Assemble team (think about external experts!)
• Identify goals, establish milestones/aims
• Create a project plan (simple or detailed)
– Specific project activities & tasks
– Timing of activities
– Interdependencies among work streams
– Cost
– Important milestones
Detailed Project Plan and Timeline: Microscoft
Gantt Chart is an Industry Standard
Simplified Project Plan and Timeline
Modeled After Gantt Chart (NIH)

Accountability
• Regularly scheduled team meetings
• Create agendas for team meetings and follow
them
• Team meeting minutes with ACTION ITEMS
assigned to team members
• Review ACTION ITEMS at next team meeting
to ensure they have been completed
• Careful tracking of documentation/data

Managing Risks
• Identify risks and have
some form of mitigation for
these risks
• Supplement your team with
complementary skill sets
• Formulate endpoints and
pivot if necessary
Build optionality into your project plans and make adjustments to

Some Important Issues for Academic
Laboratories You Need to Think About
• Research data retention
• Research data organization and storage
• Industry utilizes electronic laboratory notebooks and has
extensive infrastructure to support capturing work product,
data, and inventions

Central Desktop
• Electronic Filing System for Documents &
Data
• Functionality for Tracking Project Plans &
Timelines
• Reporting and Monitoring
• A Collaboration Tool
Data Management Tools
Central Desktop Dashboard
Best Practices from Industry that Can Be
Leveraged in Academia
• Defined milestones
• Defined project needs / resources
• Mapping of timelines
• Team building with complementary team skills
• Data management and storage
• Principles of effectiveness in managing project
progression should apply equally in both settings
project advancement is fundamental to both
Summary
• Concepts
– Intellectual Rigor, Hypothesis, Contribution to
Knowledge, Scientific Method, Dissemination
• Types of Research
– Exploratory, Descriptive, Analytical, Predictive
• Approaches
– Quantitative/Qualitative, Applied/Basic,
Deductive/Inductive
….
Summary
….
• Scientific Method
– Remove bias, ‘let the data speak for itself’
– Formulation of a question, Hypothesis, Prediction,
Testing, Analysis
– Replication, External review, Data recording and
sharing
• Project management
– Planning, Accountability, Risks, Data
Resources
• Wikipedia, Scientific Method
• TED talks, www.ted.com/talks
• Google Scholar
Tutorial
• Using the Wiki
– Visit the wiki at paulspapers.com/wiki
– Apply for an account with username
FirstnameLastname (e.g. PaulCraig)
– Visit the CSE402 Research Methods page
– Add your name and create a webpage
Tutorial
• Your homepage
Homework
• Peta Kucha Presentation
– 6 minutes 40 seconds
– 20 images, 20 seconds each
– Connected seamlessly with well-developed
narration are your only rules
– Part of our continuous assessment!
Tutorial
• A Template for Pecha Kucha PowerPoints
– Open PowerPoint. In slide view, right-click on the first slide on
the left and select Layout and Blank. This creates a blank
canvas.
– Right-click again on the slide and select Duplicate. This creates
another slide just like it.
– Since the Duplicate command is already in PowerPoint’s
memory, use the shortcut Ctrl-Y to repeat the duplicate (or just
right-click duplicate again) 18 more times, for a total of 20 blank
slides.
– Use Ctrl-A to Select all slides in the left, and then go to
Animation, advance slide and set it to 20 seconds.
– You can also select transition styles and speed here. Just don’t
choose Dissolve—the simplest is the best. Maybe nothing more
than a simple fade.