Longitudinal Data Analysis and Latent Growth Curve Modeling

Longitudinal data are very commonly collected in social work research. Almost all social work researchers collected data by several time points which produced longitudinal data to evaluate the growth curve. However, the longitudinal data are temporally correlated which directly violate the fundamental assumption of independence in typically regression modelling and therefore erroneous conclusions and social policy recommendations could be made if the data are not analyzed appropriately. Therefore an appropriate analysis of longitudinal data is important to capture the intra-individual growth changes and inter-individual variabilities. This workshop is then designed to show how to do longitudinal data analysis using R package “lme” and latent growth curve modelling using Mplus.

This workshop aims to address the longitudinal data analysis which is common in social work research. We will review the classical longitudinal data analysis methods from multi-level and hierarchical modelling to test for within-individual longitudinal change and between-individual variability. Then a latent growth curve modelling will be introduced to analyze multi-domain longitudinal data using Mplus. Real data on study of 405 Hong Kong Chinese women who underwent cancer surgery will be used as a real example in the class to model the evidence of rate change in their mood and social adjustment at 1, 4, and 8 months post-surgery.