Comparative Methodology: social scientists real people like you and me

Comparative Methodology

1. Social scientists: real people like you or me Timothy Lim

2. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases

3. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases "Why… Y?"

4. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases "Why… Y?“ Our very simple model of everything in the universe: X --> Y

5. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases How have other authors attempted to answer this question? Might X 1 cause Y? Might X 2 cause Y? Might X 3 cause Y? (If you're a college student, consider 3-5 hypotheses. If you're a social scientist, consider them all)

6. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Where has Y occurred? Identify one or two dozen cases. Was X 1 - X 3 present in these cases?

7. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Are there two very similar cases (i.e., the US and Canada) with different outcomes? Consider a Most Similar Systems (MSS) design.

8. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Most Similar Systems:   Case 1 Case 2 X1=Wealth Poor Poor X2=Ethnic diversity High High X3=Former colonial power Britain France Y=Stable Democracy Yes No

9. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Are there two very different cases (i.e., the US and Cape Verde) with similar outcomes? Consider a Most Different Systems (MDS) design.

10. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Most Different Systems:   Case 1 Case 2 X1=Wealth Poor Rich X2=Ethnic diversity High Low X3=Former colonial power Britain Britain Y=Stable Democracy Yes Yes

11. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Does causality seem very complex, with multiple Xs present in all the cases? Consider using the Method of Agreement (MoA).

12. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases Method of Agreement (MoA):   Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 X1=Wealthy Yes       Yes Yes X2=Ethnically diverse Yes Yes Yes     Yes X3=Former British colony       Yes   Yes Y=Stable Democracy Yes No No Yes No Yes

13. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases As you can see, choosing your research design and case selection go hand in hand. You have to do them in conversation with one another. BUT take note that the research design determines what cases we can select , and not the other way around.

14. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases In an idealized sequence, we choose to study populism, which leads us to a Most Different System (MDS) design, and we select the cases of Evo Morales in Bolivia and Viktor Orbán in Hungary Populism -> MDS -> Bolivia and Hungary

15. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases In practice, however, we do often choose our phenomenon because we're interested in a specific case. We are interested in studying Bolivian populism, which leads us to an MDS design, which leads us to Hungarian populism Bolivian populism -> MDS -> Bolivia and Hungary This is fine, as long as we're self-conscious about it.

16. An example comparative politics research workflow Formulate research question Read up on existing hypotheses and theories Learn about the universe of cases Choose a research design Time for case selection Determine your methodology Analyze your cases To be continued (in Poli 99: Political Science Research)