Segmented Change in Latent Variables: A Second-Order Structured Latent Curve Model for Time-Structured Count Data

• In Event: Beyond Traditional SEM: Model Fit, Heterogeneity, and New Paradigms

Abstract

First-order and second-order structured latent curve models using piecewise functions have been applied with greater frequency to continuous repeated measures data whose underlying development occurs in distinct phases or segments. As an extension of this framework, this research considers piecewise change of a latent construct over time where the latent construct is itself measured by multiple indicators that are counts at each measurement occasion. The model will be developed and its myriad of interesting features will be highlighted and discussed in the context of fitting the model to empirical expressive language data (grammatical morpheme counts) drawn from multiple distinct corpora available in the Child Language Data Exchange System (CHILDES).

Authors

• Jeff Harring, University of Maryland

• Marian Strazzeri, US Food and Drug Administration

• Nan Bernstein Ratner, University of Maryland