Exercise 1: Questions about your dissertation


Note: Not all of these questions will be relevant to your project. If they are not, you don’t need
to answer them. If there are other issues that you think are more relevant, feel free to tell me.


1.       Briefly describe your research question.

2.       What are you trying to explain? What are your dependent variables or outcomes?

3.       What are the main existing theories that explain this phenomenon?

4.       Do you have a new theory or hypotheses that explains this phenomenon? If so, what is it
and how does it work? Or are you mainly trying to adjudicate between existing theories?

5.       How do you plan to conceptualize or measure your main variables?

6.       How would you describe your methodology - a case study, small-N comparison, experiment,
analysis of survey data – and why is it appropriate for answering your question?

7.       What sorts of evidence do you plan to gather?

8.       Which of the following types best fit your project?

a.       Effects of causes: X à

b.       Causes of effects: à Y

c.       Description

d.       Typology/Concept Formation

9.       If you have a number of different variables, can you draw an arrow diagram connecting all
of them?

10.   What are your cases and what was the principle you used for choosing these cases.

11.   What is the variation in your cases that allows you to draw inferences? Is it variation
across cases? Within cases? Over time?

12.   What are the mechanisms connecting your key independent and dependent variable – ie, what are
the intermediate steps between the cause and the outcome?

13.   What are your main worries about your project? Could any of the following issues be a
problem?

a.       Omitted variables: Other factors might affect X and Y?

b.       Endogeneity: Could Y cause X? How?

c.       Selection biases: Do you have variation on your dependent variable? Are any cases missing
– whether cross-sectionally or spatially? Can you justify these omissions?

d.       Mechanism: Will you have any evidence for the steps in your mechanism?

e.       Measurement/conceptualization: Are your measures valid and reliable?

14.   If you had god-like powers and could conduct any sort of study to answer your question, what
would the study look like? (For example, if you wanted to assess foreign aid on economic growth,
you could randomly assign foreign aid to developing countries.)

15.   Is there anything else you want to tell me or would like advice about?