400009 SE Experiments in the Social Sciences (2024S)

The scientific experiment is a central way to advance knowledge in the empirical social sciences. By randomly assigning participants to experimental groups, experiments allow us to make claims regarding causality. Hence, if our aim is to establish causal inferences, wherever possible it is recommended to conduct experiments.
In this course we will discuss the scientific experiment, how and when it can be used, what pitfalls to avoid, and how to interpret results. We will focus on statistical power, design of stimuli, and open science (with a focus on Registered Reports). We will also focus on how best to analyze results.
We will explore these topics by reading and discussing texts, through inputs by me, short summaries of the texts presented by you, and hands-on analyses in R. Throughout, we will engage with your phd projects and your individual perspectives.
Each student will give a short summary of one of the papers we discuss. Everyone is expected to participate actively in the discussions.
In the hands-on part of the seminar, we will analyze actual data/experiments. We will use the software R. Although not needed, a basic understanding of R is recommended. If R is new, I recommend reading introductory texts or watching online tutorials. Here are some helpful materials:
- https://r4ds.had.co.nz
- https://github.com/jobreu/r-intro-gesis-2021
- https://github.com/ccs-amsterdam/r-course-material
- https://www.youtube.com/watch?v=BvKETZ6kr9Q
1. Introduction
2. Basics: What’s an Experiment?
3. Planning an Experiment
4. Experimental Designs
5. Open Science & Registered Reports
6. Statistical Power
7. Analysis: t-Tests
8. Analysis: ANOVAs
9. Analysis: Regressions
10. Analysis: Multilevel Analyses
11. Conclusion