Integrated PLS-SEM-Latent Growth Curve Model: A New Conditional Time Invariant Method for Analysing Panel Survey Data

This study introduces an integrated methodological framework that combines Partial Least Squares Structural Equation Modelling (PLS-SEM) with a conditional time-invariant Latent Growth Curve Model (LGCM) to analyse panel survey data. The integration addresses limitations of existing PLS-SEM approaches in longitudinal research by enabling the simultaneous evaluation of measurement validity, growth trajectories, and predictors of change over time. A disjoint two-stage approach is applied to estimate measurement models and obtain latent scores, which are subsequently used to model growth factors (intercepts and slopes). The proposed method is illustrated using three waves of the Midlife in the United States (MIDUS) study, focusing on life satisfaction and its predictors. Results indicate that psychological well-being and social well-being significantly predict both baseline levels and growth in life satisfaction, while income influences only baseline levels. These findings demonstrate the capacity of the integrated model to disentangle inter-individual differences and developmental patterns in longitudinal data. The findings indicate that: The integrated PLS-SEM-LGCM framework supports simultaneous analysis of constructs, trajectories, and predictors. Empirical validation with MIDUS data demonstrates its ability to capture variability in life satisfaction. Implementation in R through the custom pls_growth function enhances reproducibility and accessibility.