30214

WiSe 16/17: Longitudinal Data Analysis

Reinhard Pollak

Hinweise für Studierende

Requirements for credit: Five short assignments

Zusätzl. Angaben / Voraussetzungen

Please register in advance (email to instructor) as the number of participants is limited.

Kommentar

Does marriage increase happiness? Does poverty alter an individual’s party affiliation? Does further education pay off in higher earnings? To answer these kinds of questions, it is necessary to have data with repeated measures and it is necessary to analyze these data with longitudinal data analysis techniques. This course will offer an introduction to the analysis of quantitative longitudinal data. The course will start out with a short recap of ordinary least square regression analysis, with regression diagnostics, and with the specific nature of longitudinal data. The main part of class is devoted to a selection of the most relevant analytical models for longitudinal data: fixed-effect, random-effect, and hybrid models for both, linear and categorical dependent variables. The course will also give a short introduction in sequence analysis and event data analysis. The seminar will include lectures introducing to the new modeling techniques as well as computer lab sessions, in which students will apply these techniques with Stata and learn how to set up the data and interpret the empirical results. This course wants to attract students in sociology who are already well familiar with linear regression models and data handling. Schließen

Literaturhinweise

Rabe-Hesketh, Sophia and Anders Skrondal (2011): Multilevel and Longitudinal Modeling Using Stata; College Station: Stata Press

16 Termine

Regelmäßige Termine der Lehrveranstaltung

Do, 14:30 - 16:30 WZB (Reichpietschufer 50) 16 Termine

Do, 20.10.2016 14:30 - 16:30

WZB (Reichpietschufer 50)