Causal Inference (Fall 2025)
Instructor: Olgahan Çat
Term: Fall 2025
E-mail: olgahan@brown.edu
Website: olgahan.com/TBD
Office Hours: TBD — Sign up: Calendly
Class Hours: Wednesday, 4–6:30pm
Office: TBD
Classroom: TBD
Main Reference:
- Cunningham, Scott. Causal Inference: The Mixtape. Yale University Press, 2021.
Course Overview
This PhD seminar introduces core principles and cutting-edge methods of causal inference in social sciences. Topics include experimental and observational strategies for identifying causal effects, drawing from statistics, econometrics, and applied political research. Students will critically assess empirical work and develop their own research designs, culminating in a research paper proposal suitable for journal submission.
Objectives
By the end of the semester, students should be able to:
- Understand key concepts such as counterfactuals, identification, and the potential outcomes framework.
- Evaluate causal claims in political science research.
- Design and analyze randomized and observational studies.
- Gain hands-on experience with methods including matching, instrumental variables, difference-in-differences, and regression discontinuity.
- Get familiar with recent literature using cutting-edge methods.
- Develop a publishable empirical paper using causal inference tools.
Grading
-
Discussions (10%)
Present a published paper applying the method of the week (5–6 minutes). -
Final Paper Proposal (40%)
Develop a proposal for an original research paper using causal inference methods.Guidelines:
- Research Question: State your question, significance, and contribution.
- Theory and Hypotheses: Conceptual framework, causal claim, treatment/outcome/mechanism.
- Research Design: Identification strategy (experiment, IV, DiD, RDD, matching), validity threats, and solutions.
- Data and Measurement: Dataset description, variable definitions, identification assumptions.
- Preliminary Analysis (if applicable): Descriptive stats, visualizations, feasibility.
- Timeline: Outline analysis and writing plan.
- References: Bibliography of relevant literature.
Length: 10 pages (double-spaced, not including references)
Due Date: Last week — TBD
Check-in: No later than Week 7 — TBD - Final Exam or Alternative Assignment (30%)
Choose between:- Option A: Cumulative in-class exam.
- Option B (Recommended): Alternative assignment (grant proposal, IRB application, revise & resubmit, pre-registration, or detailed methods appendix). Must be approved by Week 10.
-
In-Class Presentation: Cutting-Edge Causal Inference Method (10%)
Deliver a 10–12 minute presentation on a recent methodological advance. Include motivation, mechanics, assumptions, strengths/limitations, and an applied example. - Final Paper Presentation (10%)
15-minute presentation on your final research paper with 10 minutes of Q&A.
Course Schedule
Week 1: Overview of causal inference
- Syllabus
- Cunningham: Preface & Chapter 1: Causality and Data
Week 2: Foundations — Potential Outcomes & DAGs
- Chapter 2: Potential Outcomes
- Chapter 3: Directed Acyclic Graphs (DAGs)
Week 3: Unconfoundedness & Matching
- Chapter 4: Matching
Week 4: Panel Data & Fixed Effects
- Chapter 5: Panel Data and Fixed Effects
Week 5: Instrumental Variables (IV)
- Chapter 6: Instrumental Variables
Week 6: Difference-in-Differences (DiD)
- Chapter 7: Difference-in-Differences
Week 7: Regression Discontinuity Design (RDD)
- Chapter 8: Regression Discontinuity
Week 8: Synthetic Control Method
- Chapter 9: Synthetic Control
Week 9: Experiments
- Chapter 10: Experiments
Week 10: Empirical Issues
- Chapter 11: Randomization inference, multiple testing, clustering, missing data, balance checks
Week 11: Frontiers in DiD, IV, and Other Methods
- Chapter 12: Recent Advances and Frontiers
- Student Presentations, Part 1
Week 12: Student Presentations, Part 2
Week 13: No Class (Thanksgiving Recess)
Week 14: Exam or Alternative Assignment
Week 15: Reading Period — Working on final papers
Last updated: August 31, 2025