Quantitative Methods 2

Brenda van Coppenolle and Jan Rovny

This is an intermediate level course of quantitative methods for social scientists. It builds on Quantitative methods 1, in which students learn descriptive statistics, the logic of inference, and OLS regression analysis, as well as basic programming in R. This course, aims to extend this knowledge by introducing the logic of maximum likelihood estimation. The course then applies MLE to a broad spectrum of practical questions, particularly to the analysis of categorical or ordinal dependent variables, time-series analysis, multi-level modeling, and data reduction techniques. The course further extends students' knowledge of R programming with a focus on data management and graphing. The course is based on a mixture of lectures and practical lab sessions. Students will thus be expected to participate in lectures, as well as lab sessions where they will work on practical problem sets. Besides these problem sets, and a final exam, the students will write a final paper applying the methods learned in class to a research topic of their interest.

Syllabus Final paper instructions Example student paper

Session 1: Research Design [BvC]

Session 2: Review of OLS and the Logic of Maximum Likelihood Estimation [JR]

Lecture OLS R script OLS Lecture MLE R script MLE

Session 3: Application of MLE — Logit and Probit [JR]

Lecture R script Example research

Session 4: Application of MLE — Ordinal Logit and Multinomial Logit [JR]

Lecture R script 1 R script 2 Example research

Session 5: Common Problems, Common Solutions, and Multi-Level Models [BvC]

Session 6: Difference in Difference [BvC]

Session 7: Instrumental Variables [BvC]

Session 8: Regression Discontinuity Design 1 [BvC]