Lecture 9: Your research plan
DATA5207: Data analysis in the social sciences
Dr Shaun Ratcliff
1
The research plan
Some advice
• What we are looking for in this first stage is for you to
show that you have thought about the question, the data
you are using, and the approach you are going to take.
• You need to set out a hypothesis / working theory you
plan to test based on the dataset you are looking at and
the literature that informs it.
• Then outline the methodology you intend to use to test
the hypothesis.
• The questions are deliberately broad to allow you freedom
to explore aspects of the datasets that interest you.
2
The research plan
Some advice
• What we are looking for in this first stage is for you to
show that you have thought about the question, the data
you are using, and the approach you are going to take.
• You need to set out a hypothesis / working theory you
plan to test based on the dataset you are looking at and
the literature that informs it.
• Then outline the methodology you intend to use to test
the hypothesis.
• The questions are deliberately broad to allow you freedom
to explore aspects of the datasets that interest you.
2
The research plan
Some advice
• What we are looking for in this first stage is for you to
show that you have thought about the question, the data
you are using, and the approach you are going to take.
• You need to set out a hypothesis / working theory you
plan to test based on the dataset you are looking at and
the literature that informs it.
• Then outline the methodology you intend to use to test
the hypothesis.
• The questions are deliberately broad to allow you freedom
to explore aspects of the datasets that interest you.
2
The research plan
Some advice
• What we are looking for in this first stage is for you to
show that you have thought about the question, the data
you are using, and the approach you are going to take.
• You need to set out a hypothesis / working theory you
plan to test based on the dataset you are looking at and
the literature that informs it.
• Then outline the methodology you intend to use to test
the hypothesis.
• The questions are deliberately broad to allow you freedom
to explore aspects of the datasets that interest you.
2
Make sure you look at the additional
information, plan guide, readings and other
information on the Canvas Assessment
information page.
3
The research plan
Some advice
1. Formatting.
• PDF.
• Hide chunks.
• Think about the structure of the paper.
2. Theory.
• What are the causal mechanisms?
• How do they work?
• Can you falsify your ideas?
3. Literature.
• At least 10 texts beyond set readings minimum. Don’t
have to all be academic, but you should have a good
selection of academic texts.
• What research has been done in this area? Do you agree or
disagree with their ideas? 4
The research plan
Some advice
4. Written expression.
• Clear communication is vital.
• Get a friend/classmate to read a draft. They cannot write
for you, but can provide feedback.
5. Model building
• You don’t need to have done your analysis, but you should
have checked the data and thought about this.
• Your model is designed to test a theory. Use theory to
select variables.
• Make sure variable selection makes sense (DV and IV).
Think about confounding factors and controls.
• Have you clearly explained why the model you are using
makes sense?
• These are all decisions. There is no correct default, but you
must justify decisions. 5
The research plan
Some advice
1. Introduction (150 to 200 words - suggestion only).
2. Literature and theory (200 to 300 words).
3. Data and methodology (200 to 300 words - suggestion
only).
4. Expectations and limitations (~100 words).
• What do you expect to find? Are there any limitations to
your study as you currently envisage it?
5. Bibliography (not counted in word limit).
6
Research plans are due
11:59pm, this Friday 28
April (AEST).
7
No labs this week
• Rather than labs this week, tutors will be running
additional office hours.
• Take advantage of this. Talking one-on-one with your tutor
can be useful for those of you with questions not
answered in this lecture (or elsewhere).
• Tim posted details in an announcement on canvas.
• If their normal office hours don’t work for you, ask if they
can meet with you another time. They might not always
be available when you want to meet, but it’s okay to ask.
8
No labs this week
• Rather than labs this week, tutors will be running
additional office hours.
• Take advantage of this. Talking one-on-one with your tutor
can be useful for those of you with questions not
answered in this lecture (or elsewhere).
• Tim posted details in an announcement on canvas.
• If their normal office hours don’t work for you, ask if they
can meet with you another time. They might not always
be available when you want to meet, but it’s okay to ask.
8
No labs this week
• Rather than labs this week, tutors will be running
additional office hours.
• Take advantage of this. Talking one-on-one with your tutor
can be useful for those of you with questions not
answered in this lecture (or elsewhere).
• Tim posted details in an announcement on canvas.
• If their normal office hours don’t work for you, ask if they
can meet with you another time. They might not always
be available when you want to meet, but it’s okay to ask.
8
No labs this week
• Rather than labs this week, tutors will be running
additional office hours.
• Take advantage of this. Talking one-on-one with your tutor
can be useful for those of you with questions not
answered in this lecture (or elsewhere).
• Tim posted details in an announcement on canvas.
• If their normal office hours don’t work for you, ask if they
can meet with you another time. They might not always
be available when you want to meet, but it’s okay to ask.