Identifying Number of Curves in Longitudinal Growth Models

• In Event: AERA Roundtable Session 16
  In Roundtable Session: Applications, Demonstrations, and Evaluations of Statistical Models and Methods (Table 16)

Abstract

Researchers who can design their studies to collect longitudinal data will usually assess each individual on an identical set of occasions, e.g. smoking behavior at age 15, 16, 17, 18, 19, and 20 years of age. Statistical analysis of this longitudinal data generally has two basic characteristics: 1) each individual is measured on the same number of occasions (balanced design) ; and 2) each individual is measured at fixed time intervals. The outcome variable is also often assumed to be time invariant. In this paper, we demonstrate a modeling strategy for determining the number of curves in the longitudinal data structure and the use of a time-varying predictor.

Author

• Randall E. Schumacker, University of Alabama