Applied Panel Data Analysis

Instructors: Oliver Lipps & Ursina Kuhn

Modality: In presence

Week 2: 22 -26 August, 2022

Panel data are repeated measures of a large number of units, e.g., individuals or companies. Such longitudinal data is typically collected through surveys, but panel data can also come from other data sources, such as administrative records. The longitudinal nature of the data makes it possible to address research questions, that cannot be analysed with cross-sectional data and can be used to approach causality. However, panel data have a more complex structure and require additional skills for data preparation and data analysis.

In this course, we provide the skills to prepare panel data for analysis and apply regression models with panel data. We will put emphasis on understanding the mechanism behind the different approaches, focussing on their advantages and limitations such that participants will be able to choose and apply the appropriate model.

Workshop contents

This workshop provides an introduction to the structure, data management, and analysis of panel data. Participants will learn how to prepare panel data for different types of research questions specific to this type of data.  This includes creating wide data files, long data files, spells, and transitions that can be used for descriptive statistics and for regression models.

For data analysis, we focus on different basic regression models, namely pooled OLS, fixed effects, random effects, and first differences. The aim of this workshop is to understand the within perspective of panel data as opposed to the between perspective of cross-sectional data and to grasp the mechanics of these models.

For careful data analysis, it is important to understand how data has been collected. For panel, difficulties arise, among others, from following of units over time, non-response and attrition, and measurement changes over time (e.g. change of survey modes). We will use real data in the course and discuss the implications of these issues for data analysis. Examples are mainly from the Swiss Household Panel (SHP), which contains data over more than 20 years, covers topics from different social science disciplines and offers different kinds of data files (individual, household, one-time collected information, events). We will also discuss other types of panel data, that have different numbers and types of levels (households, schools, companies) and measurement points.

In the morning sessions, we will present theory and applications and discuss how to do research with panel data. In the afternoon sessions, participants will work on practical exercises either with data prepared by the instructors from the SHP or with panel data of their own choice. One of the instructors is available to discuss and solve individual problems.

Prerequisites

We will use the Stata software and assume a basic knowledge of Stata with cross-sectional data. For students not sufficiently proficient in Stata we offer a two-day preparatory online course shortly before the workshop introducing Stata. We will also provide syntax for R users for data preparation and to apply models treated in the course. However, we can only give limited advice for programming in R.

Bibliography

Introductory texts

  • Longitudinal Data / SHP data
    • Andress, H. J., Golsch, K., & Schmidt, A. W. (2013). Applied panel data analysis for economic and social surveys. Springer Science & Business Media.
    • Ruspini E. (1999) Longitudinal Research and the Analysis of Social Change. Special Issue of Quality & Quantity 33: 219-227. [Springerlink, access requires subscription]
    • Voorpostel et al. (2017) Swiss Household Panel User Guide (1999-2016). FORS, Lausanne. [Swiss Household Panel] Specifically chapters 2.1 (p.13), 2.5. (p.17-22)
  • Application examples using SHP data
    • Lipps, O. and F. Moreau-Gruet (2010) Change of individual BMI in Switzerland and the USA: a multi-level model for growth." International Journal of Public Health, 55(4):299-306.
    • Oesch, D. and O. Lipps (2013). Does unemployment hurt less if there is more of it around? A panel analysis of life satisfaction in Germany and Switzerland. European Sociological Review 29 (5): 955-967.
    • Fitzgerald, J., and K. Amber Curtis (2011). Partisan discord in the family and political engagement: A comparative behavioral analysis. The Journal of Politics 74 (3): 783-796.

Books and articles on panel data analysis methods

(Kind of material covered during class)

  • Kohler, U. and F. Kreuter (2009). Data Analysis Using Stata, 2nd ed. College Station, Texas: Stata Press.
  • Rabe-Hesketh, S. and A. Skrondal (2007). Multilevel and Longitudinal Modeling Using Stata. Second edition. College Station, Texas: Stata Press.
  • Singer J. and J. Willett (2003). Applied longitudinal Analysis - Modeling Change and Event Occurrence. Oxford University Press.
  • Online resources: Stata Starter Kit, UCLA Academic Technology Services, USA
  • Wooldridge, J. (2010). Econometric Analysis of Cross Section and Panel Data, 2nd Edition. MIT Press.
  • Angrist, J. and J. Pischke (2009). Mostly harmless econometrics: An Empiricist's Companion. Princeton University Press.
  • Morgan, S. and C. Winship (2015). Counterfactuals and Causal Inference. Methods and Principles for Social Research. 2nd ed. Cambridge University Press.
  • Bell, A. and K. Jones (2015). Explaining Fixed Effects: Random Effects Modeling of Time-Series Cross-Sectional and Panel Data. Political Science Research and Methods 3(1): 133-153.