Paper Summary
Share...

Direct link:

Detecting Sigmoidal Trajectories in Structured Latent Curve Models: An Examination of Fit Measure Performance

Sun, April 30, 8:15 to 10:15am, Henry B. Gonzalez Convention Center, Floor: River Level, Room 7C

Abstract

This simulation study examined structured latent curve models under varying conditions to provide recommendations on how to properly format longitudinal research when there is an a priori hypothesis of sigmoidal growth. It also provides recommendations for model identification when completing post hoc analysis of existing data. Sigmoidal models were generated using the first order Taylor series approximation method detailed in Browne and du Toit (1991) and analyzed as sigmoidal, linear, quadratic, and cubic. Thirteen fit measures were assessed to determine their performance in identifying true sigmoidal models over competing incorrect models. Manipulated design factors for this study include sample size, repeated measures, inflection point, and rate of change at the inflection point. Results and implications of the findings are discussed.

Author