Categorical longitudinal panel analysis
is currently chairman Department of Methodology, Faculty of Social Sciences, Tilburg University.
A short CV can be found here.
The objective of the course is to make people understand the basics
of longitudinal analyses and especially the analyses of longitudinal
categorical data, and then mainly the (loglinear) analyses of panel
data. The emphasis will be on practical problems and applications.
Topics:
- longitudinal designs: panel, trend, and cohort designs
- the core business: analyses of turnover tables
- Answering typical panel questions: comparing turnover tables among
groups, over time, among different characteristics
- Longitudinal Causal Analysis
- Unreliability: separating real change from change due to
measurement error.
- systematic and unsystematic measurement error
- categorical latent variable models (latent class analysis)
- Nonresponse, missing data, panel Attrition
Bibliography
Basic text/overview/main references
- Jacques A. Hagenaars (1990) 'Categorical Longitudinal Data; loglinear
panel, trend, and cohort analysis' Newbury Park: Sage
- Jacques A. Hagenaars (1993) 'Loglinear Models with Latent Variables',
QUASS 94, Newbury Park:Sage
- Andress, Hans-Juergen, Jacques A. Hagenaars,
Steffen Kuehnel (1997) 'Analyse von Tabellen und Kategorialen Daten;
loglineare Modelle, latente Klassenanalyse, logistische Regression und GSK
Ansatz. Berlin: Springer
Remedial Reading
- Knoke, D. and P.J. Burke (1980) 'Log-linear Modeling' QUASS 20,
Newbury park: Sage
- Reynolds, H.T. (1977) 'The Analysis of Cross-Classifications'. New
York: Free Press
- Langeheine R. (1986) 'Log-linear Modelle' Pp.
122-195 in: J. von Koolwijk and M. Wieken-Mayser (Eds.) Techniken der
empirische Sozialforschung. Band 8. Kausalanalyse. Muenchen: Oldenburg.
Prerequisites
Basic courses on social science statistics (descriptive and
inferential statistics)
knowledge of basic social science methodology, including the basics
of cross-sectional and longitudinal designs.