234002 SE Statistics for Social Scientists 2 (2024S)

This seminar is targeted at students with an interest in applied empirical research and will prepare them to answer their own research questions by carefully choosing their data and employing appropriate survey and microeconometric methods.

The course will start by introducing students to different data types and data sources frequently used for microeconometric analyses in the social sciences. Next, we continue to discuss the most important quality criterions of these data and explore how potential flaws can be accounted for with survey methods. We then proceed to the largest part of the course, which will deepen students’ knowledge about quantitative data analysis, focusing on microeconometrics. More specifically, we will cover

• General linear models for different data types
• Decomposition methods
• Panel data analysis with fixed and random effects models
• Causal inference from instrumental variables, difference-in-difference analyses, and regression discontinuity designs
• Survival analysis

METHODS: The lecturer will introduce students to different data sources, survey methods, and estimation techniques, thereby mainly focusing on the intuition behind the respective methods. In addition, students will learn how to critically evaluate the implementation of these methods by reading and discussing topical research articles. They will apply the newly learned methods by analysing data during a take-home assignment.

LEARNING OUTCOME: After this course, students will (i) know the most important data types and sources for microeconometric analyses in the social sciences, (ii) know how flawed data can be accounted for using survey methods, (iii) be able to identify appropriate estimation techniques to analyse these data, in particular microeconometric methods, and (iv) be able to critically evaluate research designs considering data, methods, and interpretation of the results.

COURSE AND EXAMINATION LANGUAGE: English

PREREQUISITES: Students should have basic training in statistics including hypothesis testing, probability distributions, and linear regression. Basic knowledge of a statistical software (e.g., Stata or R) is necessary to complete parts of the take-home assignment (see performance components below). The software will not be covered in class.