Latent transition analysis with random intercepts (RI-LTA).

Area of Science:

• Statistics
• Psychometrics
• Longitudinal Data Analysis

Background:

• Latent Class Transition Analysis (LTA) models are widely used for analyzing change over time.
• Standard LTA models can be overly restrictive, potentially limiting the accuracy of findings.
• There is a need for more flexible models to better capture complex longitudinal data patterns.

Purpose of the Study:

• To introduce and evaluate an alternative Latent Class Transition Analysis (LTA) model incorporating random intercept variation.
• To demonstrate that this enhanced LTA model offers superior data fit and parameter estimation compared to the standard LTA.
• To highlight the benefits of separating between-subject and within-subject variation for clearer interpretation of longitudinal data.

Main Methods:

• Development and application of a random intercept Latent Class Transition Analysis (LTA) model.
• Comparative analysis of the proposed model against the standard LTA using existing literature examples.
• Exploration of model variations including Mover-Stayer analysis, measurement invariance, and covariate inclusion.

Main Results:

• The proposed random intercept LTA model demonstrates a significantly better fit to the data compared to the standard LTA.
• The enhanced model provides more accurate estimates of transition probabilities.
• Separating between-subject variation from within-subject transitions leads to clearer data interpretation and extraction of novel insights.

Conclusions:

• The standard Latent Class Transition Analysis (LTA) model is unnecessarily restrictive.
• An alternative LTA model with random intercept variation offers substantial improvements in data fit, estimation accuracy, and interpretability.
• This enhanced LTA methodology provides a more powerful tool for analyzing longitudinal data and understanding individual change processes.