Longitudinal Data Analysis

Longitudinal data are an important tool for social researchers, as they can help address questions related to temporal ordering, change over time, and sometimes even causal relationships. Learning how to collect and analyse longitudinal data is therefore a desirable skill to a quantitative social scientist’s methodological toolkit.

In this short course, you will learn how to leverage panel data to address two common questions in the social sciences: causal inference and reciprocal relationships. First, we will discuss how to use longitudinal data to infer causal effects based on the potential outcomes framework, with a focus on the difference-in-differences estimator. Second, we will discuss how longitudinal data can be powerful at depicting reciprocal relationships and controlling for reverse causality using cross-lagged panel models. In both cases, we will learn the intuition of the basic setup, understand some of its strengths and limitations, and discuss some recent developments.

Adopting a hands-on approach, we will also draw on real-world applications using R and practical exercises.

Session outcomes:

Students will learn how to leverage panel data to address two common questions in the social sciences: causal inference and reciprocal relationships.

This training session will be delivered in person at the University of Leeds.

This training session will be recorded, and the recording will be made available on the WRDTP website.

Places are limited, so please only book a ticket if you can guarantee your attendance.

Bookings will close at 9am on Wednesday 2nd April.