|
HU Credits:
4
Degree/Cycle:
2nd degree (Master)
Responsible Department:
Political Science
Semester:
Yearly
Teaching Languages:
English
Campus:
Mt. Scopus
Course/Module Coordinator:
Dr. Matthew Simonson
Coordinator Office Hours:
By appointment
Teaching Staff:
Dr. Matthew Simonson
Course/Module description:
This course will teach you to answer the question “Does X actually cause Y or are they merely correlated?” We will begin by looking at randomized experiments including:
• a survey experiment testing how a hypothetical candidate’s ethnicity, age, and religion affects their popularity with voters
• a field experiment that randomly assigns female police officers to districts without any
• an evaluation of programs aimed at discouraging former rebels from joining criminal gangs
We will also discuss the ethics, practicalities, and limitations of experiments—and design our own. The course then turns to “natural experiments” and “quasi-experiments” such as:
• the impact of a new policy that is being gradually phased in (e.g., comparing districts where the policy has already been implemented to those where it has not)
• the legacy of colonialism on democratic values (e.g., comparing villages falling on one side of an arbitrary colonial border to those on the other)
• how a monarch’s gender impact’s foreign policy (e.g., comparing European monarchs who, by chance, had no sons and were thus succeed by a queen instead of a king)
These natural and quasi-experiments—which tend to be partially but not completely random—require more advanced statistical tools. Therefore, we will learn R, a programming language that has become the standard among the newest generation of political scientists.
Course/Module aims:
This course aims to help you develop the tools to test causal relationships in academia and beyond. A thorough understanding of causal inference has become essential for any student seeking to publish quantitative research in a top journal. For those seeking a career outside academia, this course will give you the tools to evaluate the impact of government policies and non-profit programs. Those interested in the private sector will come away with the skills to experimentally compare products, apps, and marketing strategies.
Learning outcomes - On successful completion of this module, students should be able to:
1. Examine social phenomena through a causal lens
2. Design and conduct experiments and quasi-experiments
3. Measure the impact of interventions using advanced statistical techniques
4. Program in R at an intermediate level
Attendance requirements(%):
100
Teaching arrangement and method of instruction:
In-person lecture and discussion
Course/Module Content:
I. Theories of Causal Inferences
- Counterfactual reasoning
- Directed Acyclic Graphs
- Potential Outcomes
II. Experiments
- Simple, Complete, Block, and Cluster Randomization
- Randomization Inference
- Ethics of Human Subjects Research
- Preregistration, Transparency, and Replicability
- Survey Experiments
- Field Experiments
- Conjoint Experiments
- Encouragement Designs
- Power Analysis
III. Estimation
- Stratification
- Matching and Weighting
- Regression and weighted regression
IV. Classic Quasi-Experimental Methods
- Instrumental Variables
- Regression Discontinuity
- Panel and Fixed Effects
- Differences-in-Differences
V. Advanced Inference Techniques
- Synthetic Control and Matrix Completion
- Networks and Interference
- Causal Forests and Heterogenous Effects
Required Reading:
1. Social Science Experiments: A Hands-on Introduction by Donald Green, 2023
2. Field Experiments: Design, Analysis, and Interpretation, by Alan Gerber and Donald Green, 2012
3. R for Data Science (2nd edition) by Hadley Wickham, Mine Çetinkaya-Rundel, and Garrett Grolemund (free online), 2023
4. The Effect: An Introduction to Research Design and Causality by Nick Huntington-Klein (free online), 2022
Additional Reading Material:
Grading Scheme :
Essay / Project / Final Assignment / Home Exam / Referat 20 %
Active Participation / Team Assignment 10 %
Submission assignments during the semester: Exercises / Essays / Audits / Reports / Forum / Simulation / others 70 %
Additional information:
Familiarity with introductory statistics is required. Students are expected to be familiar with p-values, confidence intervals, and linear regression. No programming experience is needed.
|
