Dr Jason Seawright is Professor of Political Science at Northwestern University. He is the author of three books, Party-System Collapse: The Roots of Crisis in Peru and Venezuela (Stanford, 2012) Multi-Method Social Science: Combining Qualitative and Quantitative Tools (Cambridge, 2016), and Billionaires and Stealth Politics with Benjamin Page and Matthew Lacombe (Chicago, 2018). He has also published in Political Analysis, Sociological Methods and Research, the American Journal of Political Science, Perspectives on Politics, and Comparative Political Studies, among other journals and edited volumes. His research interests include causal inference, mixed-methods research designs, political parties and party systems, populism, and political representation.

Course Content
The course will discuss mixed-method research in the sense of research designs that incorporate quantitative as well as case-study components. It will focus primarily on issues of causal inference, and the first sessions will review statistical theories of causation, and also discuss causal inference in regression and case studies. The course will then cover research designs in which qualitative components are used to test the assumptions of regression-based causal inference. Topics include looking for confounders, measurement error, mediation, and case selection. The course then turns to designs using more advanced statistical methods: matching, natural experiments, true laboratory experiments, and machine learning tools. Finally, the course considers designs in which a qualitative causal inference is improved using statistical design components.

Course Objectives
Participants should expect to develop an increased ability to notice and take advantage of aspects of the theories or hypotheses that they test which are open to consideration using multiple research designs. They will also possess a toolkit of statistical techniques and case-study ideas that facilitate successful multi-method research. These tools and abilities are particularly useful in planning and executing coherent large-scale research projects, such as book-length work in which multiple research components are intended to speak to a common underlying theory or question.

Course Prerequisites
Students should have carried out regression analysis using real social scientific data at least a few times. They should be able to read and interpret the standard regression tables routinely published in social science journals. They should also have read at least some instances of case-study research in the social sciences.

Representative Background Reading
Lieberman, Evan S. 2005. Nested analysis as a mixed-method strategy for comparative research. American Political Science Review 99 (3): 435-52.

Required texts
Seawright, Jason. Multi-Method Social Science: Combining Qualitative and Quantitative Tools(Cambridge, 2016). ISBN: 978-1107483736. This text will be provided as part of the course material provided by the Summer School.

Provisional course programme (emphasis given to topics may change)

1) The epistemological argument for mixed-method designs
2) Comparative strengths of qualitative research techniques
3) Quantitative tools for case selection
4) Designing and implementing case studies in light of prior quantitative research
5) Revising statistical analysis to test/incorporate case-study insights
6) A specific design: a paired comparison framed by prior regression analysis
7) A specific design: the quantitative comparative case study
8) A specific design: experiments nested within case studies

Day 1: Epistemology and Mixed-Method Designs
• What are mixed-method designs (examples)?
• Characterising methods in terms of assumptions and inferential scope.
• Arguments for and against mixed-method designs.

Day 2: Comparative Strengths and Weaknesses
• Strengths and weaknesses of qualitative research
• Varieties of quantitative research, with strengths and weaknesses
• Strengths and weaknesses of experiments

Day 3: The Comparative Method and Case Selection
• Mill’s methods
• Crucial cases
• Quantitative tools for case selection

Day 4: Case Studies after Quantitative Work I
• Causal mechanisms and regression models
• Measurement error

Day 5: Case Studies after Quantitative Work II
• Discovering and analysing omitted variables
• Considering dependency among cases
• Thinking about interactions

Day 6: Quantitative Work after Case Studies
• Testing generalizability
• Mechanisms and models
• Addressing measurement problems
• Testing the “importance” of omitted variables

Day 7: A specific design: a paired comparison framed by prior regression analysis
• Nested induction/inference
• Examples

Day 8: A specific design: the quantitative comparative case study
• Comparative method and comparative within-country quantitative analysis
• Strengths and weaknesses vis-à-vis pooled quantitative analysis, hierarchical models, and qualitative case studies
• Examples

Day 9: A specific design: experiments nested within case studies
• Theory, case studies, and experiments
• Basics of experimental design
• Qualitative evidence and external validity
• Incorporating experimental results in the case study

Day 10: Matching methods and case studies
• Basics of quantitative matching methods
• Using case studies to evaluate causal inferences based on quantitative matching methods
• Matching to choose cases for qualitative research