Longitudinal Analysis with Panel Data

About

Workshop contents and objectives

This workshop focuses on longitudinal data, which observe the same persons, firms or other entities at different points in time. Such data is crucial for studying stability, change, and causal mechanisms. During the course, participants will learn how to organize longitudinal data, measure change, and to analyse it using statistical methods.

The main focus for data analysis will be fixed effects models. While cross-sectional models compare different units (for example the happiness of married and unmarried people), fixed effects models analyse changes within units over time (for example happiness of a person before and after marriage). By implicitly controlling for unobserved time-constant variables, these models are well suited to identifying causality. Other models covered include pooled OLS, multilevel/random effects models, growth models and cross-lagged models for longitudinal data.

The focus will be on understanding the mechanics of the different models, so that participants will be able to choose suitable modelling strategies for their research questions and the characteristics of the data (e.g. number of waves, number of observations, type of units).

We will also address how data quality may impact statistical analysis (e.g. missing values, attrition). Using real data from the Swiss Household Panel, the course combines statistical tools with practical challenges such as limited statistical power due to small N, few time points, or attrition.

Workshop design

The course consists of an equal share of lectures and discussion, as well as exercises using real data. These exercises consist of individual hands-on tasks and small group activities, in which the methods learned will be applied to real data. Groups will replicate and extend the analysis of recently published papers using panel data. Alternatively, participants may analyse and present their own data. Each group will present their findings in plenary sessions on the last day.

With two instructors available, there will be ample opportunity for individual counselling. Both have extensive practical experience in collecting, preparing and analysing panel data and are happy to share their knowledge and provide guidance on your current data analysis.

Detailed lecture plan (daily schedule)

Day 1 a) Introducing longitudinal data and application examples (Swiss Household Panel).
b) Data preparation, descriptive analysis, attrition analysis, and variance decomposition.
Day 2 a) Regression models for longitudinal data: Pooled regression and change-score models.
b) Approaching causality with longitudinal data using Fixed Effects and First Difference models.
Day 3 a) Multilevel models: Random effects, Hybrid model.
b.) Hands-on exercises and group work.
Day 4 a) Growth models in a multilevel framework.
b) Participant Presentations I.
Day 5 a) Advanced models: Cross-lagged models, dynamic models, handling missing data.
b) Participant presentations II and course wrap-up.

Class materials

Prepared student data sets with example syntax and exercises.
Power Point presentations.
A selection of application examples of longitudinal data analysis from scientific journals in the social sciences (sociology, psychology, economics and political science). We will replicate and expand the analysis of some of these papers in the group work.
The compendium “Stata Data Management” - written by the instructors - provides an introduction into longitudinal data analysis based on data from the Swiss Household Panel.

**The Summer School cannot grant credits. We only deliver a Certificate of Participation, i.e. we certify your attendance.**

If you consider using Summer School workshops to obtain credits (ECTS), you will have to investigate at your home institution (contact the person/institute responsible for your degree) to find out whether they recognise the Summer School, how many credits can be earned from a workshop/course with roughly 35 hours of teaching, no graded work, and no exams.

Make sure to investigate this matter before registering if this is important to you.

Course leader

Ursina Kuhn is a senior researcher at FORS and a member of the Swiss Household Panel. Oliver Lipps is a survey methodologist at FORS, Lausanne, and member of the Swiss Household Panel team. In addition, he is a lecturer in survey methodology and survey research at the Institute of Sociology at the University of Bern.

Target group

graduate students, doctoral researchers, early career researchers, experienced researchers

Prerequisites

We will use the Stata software and assume some knowledge of Stata with cross-sectional data. For students not sufficiently proficient in Stata we offer an online two-day preparatory course shortly before the workshop introducing Stata. Experienced R-users can conduct the exercises with R, making use of AI tools to translate Stata code to R.

Fee info