Speakers: Cathy LaBrish, York University
Department of Psychology
Title: Growth Mixture Models - Real Life Trials and Tribulations
Abstract: Memory researchers often employ measures designed to assess the rate at which individuals learn new information. This learning rate is often assessed using a "learning over trials" task. In a learning over trials task, individuals are presented with the same new information repeatedly during the same testing session. The rate at which individuals acquire new information is thought to be predictive of whether or not individuals later develop conditions in which memory is particularly impaired (e.g., Alzheimer's disease). My recent research has focused on determining whether the learning over trials portion of the Rey Auditory Verbal Learning Test (RAVLT) is potentially predictive of later development of Alzheimer's (DAT). To test this hypothesis, I have been attempting to estimate a growth mixture model (GMM) in which rates of change in the number of words recalled on the learning over trials portion of the RAVLT at baseline assessment is predictive of membership in a group later diagnosed with DAT. I report on my progress to date as well as next steps, with particular emphasis on issues relating to checking of the assumptions underlying a GMM as well as those specific to the estimation of these models (e.g. selection of starting values, identifying local maxima).
Suggested Readings:
Bauer, D. J., & Curran, P. J. (2003). Distributional Assumptions of Growth Mixture Models: Implications for Overextraction of Latent Trajectory Classes. Psychological Methods, 8, 338–363.
Hipp, J. & Bauer, D. J. (2006). Local Solutions in the Estimation of Growth Mixture Models. Psychological Methods, 11, 36-53.
Jung, T. & Wickrama, K. (2008). An Introduction to Latent Class Growth Analysis and Growth Mixture Modeling. Social and Personality Psychology Compass, 2/1, 302–317.
Ram, N. & Grimm, K. (2009) Growth mixture modeling: A method for identifying differences in longitudinal change among unobserved groups. International Journal of Behavioral Development 2009, 33, 565–57.
