Research Design and Data Analysis in Political Science
Research Design and Data Analysis in Political Science is a second-year elective course of the Master’s programme in Global Politics and Euro-Mediterranean Relations (GLOPEM) of the University of Catania, also open as an optional course to students of the Master’s programme in Data Science: 6 ECTS, 30 hours of lectures and 12 hours of laboratory, in English. It is the course on research design and methods that the Chair’s Grant Agreement undertakes to redesign. Before the Chair the course was offered in the Master’s in Data Science; under the Chair it moved to the Department of Political and Social Sciences and opened to political science students. It has been taught in each year of the Chair: in spring 2024 (academic year 2023-24, as Political Science Research Design and Methods), in spring 2025 (2024-25, as Research Design and Methods in Political Science) and in autumn 2026 (2026-27, with the current title). There was no edition in the academic year 2025-26, when the course moved from the first to the second year of the GLOPEM programme.
Overview
The course covers the methods for designing political and social science research: how to design a research project, from asking good research questions to selecting data sources, formulating and testing hypotheses, and evaluating findings, with attention to the differences between research that explores problems, develops theories or makes forecasts. Each student chooses a topic and develops a research project during the course. It is not a primarily technical course: the emphasis is on the ability to design, carry out and evaluate research, more than on analytical techniques or statistical modelling. Political science and data science students attend together, and for some laboratory sessions the two groups have separate activities.
Learning outcomes
Students identify the theories, hypotheses and methods used in empirical political science research, know the main research designs (experimental, large-N, comparative, case study and mixed) and the logic of causal inference, and understand how large and new data sources contribute to the study of political and social problems. They design a research project of their own, draw causal diagrams (DAGs) to represent a theory, and carry out basic data analysis in R. They assess which design and method suit a research question and evaluate the design of published research. They present their project in writing and in front of an audience, and they read methodological and applied literature on their own.
Structure
| Part | Sessions | Content |
|---|---|---|
| Foundations | 1–5 | The scientific study of politics; research questions and the literature review; theory; concepts, operationalisation, measurement and description; explanation and causality. First student workshop, on research questions |
| Causal diagrams | 6–7 | DAGs of published research; the DAG of each student’s own project, built in class |
| Research designs | 8–12 | Experimental designs; large-N designs; comparative designs and case selection; case studies, within-case analysis, mixed and nested designs. Second student workshop, on theory, hypotheses and causal diagram |
| Data analysis and student projects | 13–16 | Data analysis with R in three laboratory afternoons, with separate tracks for political science and data science students; third student workshop, with the presentation of the research designs |
How the course is taught
In the lectures the instructor presents the topic of the session, following the reference text. In most lecture sessions the class discusses one or two published studies, for what they find and for how they are designed. In three student workshops the students present their own research design proposal: first the research question, then theory, hypotheses and causal diagram, with a written deliverable, then the research design. Short exercises inside the lectures practise, for example, variable and case selection. The laboratory hours go to causal diagrams, drawn first on published research and then on the student’s own project and tested by peers, and to data analysis with R in three afternoons: political science students follow a guided introduction to R and the tidyverse with exercises on political science datasets, while data science students analyse the data for their own proposal or replicate a published study and report to the class.
Editions
- Autumn 2026 (2026-27): 22 attending students.
- Spring 2025 (2024-25): 10 attending students.
- Spring 2024 (2023-24): 5 attending students.
Assessment
Class participation and the three workshop presentations count for 30%, the final paper for 40% (a research design proposal of 3,000 to 5,000 words, built from the workshop deliverables and including the causal diagram and a data note from the R laboratory), and the final exam for 30% (a written design outline on an assigned topic and an oral discussion of the proposal). Data science students, for whom attendance is not compulsory, may choose a non-attending programme.
Texts
Dimiter Toshkov, Research Design in Political Science, Palgrave, 2016, is the main text. The course also uses Paul M. Kellstedt and Guy D. Whitten, The Fundamentals of Political Science Research, 3rd edition, Cambridge University Press, 2018; Nick Huntington-Klein, The Effect. An Introduction to Research Design and Causality, Chapman and Hall/CRC, 2022; Scott Cunningham, Causal Inference: The Mixtape, Yale University Press, 2021; and, in the R laboratory, Elena Llaudet and Kosuke Imai, Data Analysis for Social Science, Princeton University Press, 2022.
Syllabi
Seminars
In 2023-24 the course hosted the first seminar of the Chair, by Stefano M. Iacus (Harvard University) on how the language models behind ChatGPT work and how far we can trust them (17 April 2024), and its R block was held as a three-day crash course on R for data analysis (22–24 May 2024). In 2026-27 the course hosts the seminar of Adrià Urso Coll (FAO) on large language models and agentic AI, on 10 November 2026.