S651: Topics In Quantitative Sociology: Longitudinal & Panel Data Analysis
This special topics course covers applied statistical techniques for the analysis of repeat observations over time. The course draws on a range of models from various social science disciplines. The course begins with a review of the general linear regression model for continuous dependent variables and maintains a primary emphasis on models for continuous outcomes. Topics include (1) instrumental variables approaches, (2) error components and dynamic panel models in econometrics, (3) multilevel growth curve models, and as time permits (4) an introduction to event history models.
Instructor: Patricia McManus - firstname.lastname@example.org
Other Contact(s): Patricia McManus - email@example.com
Sequence: Prerequisites: two semesters of linear and nonlinear regression
Prerequisites: See above: regression
Algebra Required?: Matrix algebra is used for notation in course notes and texts.
Calculus Required?: No calculus expected, never needed.
Day(s) per week offered: 1X 2-hr Lecture 1X 2hr/Lab
Recommended follow-up classes: Advanced Panel Data, Latent Growth Models, Survival/Event History Analysis
Substantive Orientation: Social sciences, policy analysis, education, applied health sciences
Statistical Orientation: Observational data; Applied statistics
Books used: a) Course notes; b) Wooldridge, Jeffrey M. Econometric Analysis of Cross Section and Panel Data, 2nd Edition. Cambridge: MIT Press. c) Rabe-Hesketh, Sophia and Anders Skrondal. 2008. Multilevel and Longitudinal Modeling Using Stata. 2nd Edition. College Station: Stata Press. Similar text: Gelman and Hill "Data Analysis Using Regression and Multilevel/Hierarchical Models (Cambridge University Press 2007).
Applied/Theoretical: Applied with theory as context. No derivations, but students are expected to have a solid theoretical understanding of linear regression and its application to observational data
Software Used: GLLAMM, Stata
How the software is used: Data analysis
Problem Sets: Yes
Data Analysis: Yes, in every problem set
Exams: No exam
Keywords: regression models, panel data analysis, longitudinal data analysis, growth curves
Comments: This is an applied course: students are also required to submit two short critiques of empirical research articles in their field, along with a copy of the article.