An Investigation of Power Analysis Approaches for Latent Growth Modeling

• In Event: Roundtable Session 101
  In Roundtable Session: Extensions and Issues in Growth Modeling

Abstract

This investigation studied the impacts of variations in specified parameters, design features, and model misspecification in simulation-based power analysis for testing between-population difference in slopes and compared power estimates across Monte Carlo, Satorra-Saris, and MacCallum-Browne-Cai (MBC) methods. Sample size, effect size, and slope variance markedly influenced power estimates; level-1 error variance and number of repeated measures (given fixed study length) had little impact on power. Misspecification in level-1 error structure had little influence on power, whereas misspecifying the form of the growth model as linear rather than quadratic dramatically reduced power for detecting differences in slopes. Power estimates based on the Monte Carlo and Satorra-Saris techniques were very similar and often differed markedly from those based on the MBC technique.

Authors

• Bethany L. Van Vleet, Arizona State University - Tempe

• Marilyn S. Thompson, Arizona State University