Latent Transition Analysis (LTA) : A Method for Identifying Differences in Longitudinal Change Among Unobserved Groups

Abstract

The latent transition analysis (LTA) model is a version of Latent Class Analysis (LCA) which is used in longitudinal data analysis. The goal of LTA is to examine the variation over time and to identify the association of repeated measures. LTA gives an elegant solution to study heterogeneous changes in longitudinal data. As a classic LCA, it assumes that the data consists of several unknown groups that have homogeneous choices. This paper aims to present a review of assess the performance of LTA to identify the differences in longitudinal differences among unobserved classes. An example of LTA application in educational assessment was developed to illustrate the process and to explore the change among the time in reading comprehension.