Conferences

ESRA2009:

Conference main page

Overview of sessions

Time table

Warsaw 2009: Presentations and short courses

Growth Mixtures Models with Panel Data: Recent Applications with Adolescents’ Delinquent Behaviour

Session: Comparing and Evaluating Autoregressive, Latent Trajectory, Autoregressive Latent Trajectory, and Continuous Time ALT Models (II)

Authors:

Jost Reinecke; University of Bielefeld , Germany
Luca Mariotti; University of Münster, Germany

Abstract:

Growth curve models as structural equation models have been extensively discussed in various fields (e.g., Duncan, Duncan, Strycker, Li & Alpert, 2006; Curran & Muthén, 1999). Recent methodological and statistical extensions have taken into consideration unonserved heterogeneity in empirical data. Muthén extended the classical equation approach to mixture components, i.e. categorical latent classes (Muthén, 2001, 2002, 2004).
The paper will discuss applications of growth mixture models to data from one of the first panel studies in Germany which explore deviant and delinquent behavior of adolesents (Reinecke, 2006a, 2006b). Observed as well as unobserved heterogeneity will be considered with growth mixture models using the program Mplus (Muthén & Muthén, 2006). Special attention will be given to the distribution of the substantive dependent variables as count measures (Poisson distribution, zero-inflated Poisson distribution, cf. Nagin, 2005). Different model specifications with respect to substantive questions will also be emphasized.