Unraveling Temporal Dynamics: Utilizing Cross-Lagged Panel Models for Data Mining in Dental Health Research

The Cross-Lagged Panel Model (CLPM) is a robust statistical tool widely employed in longitudinal data analysis, enabling researchers to assess reciprocal relationships between variables over time. CLPM has significant applications in data mining within dental science, particularly in elucidating the temporal relationships between psychological and behavioral variables related to oral health. This cohort study focused on individuals aged ۶۰ years and older in Tehran, examining changes in self-efficacy scores, risk perception, and outcome expectations across three time periods: before the intervention and one and three months post-intervention. The study utilized the Health Action Process Approach (HAPA) model to guide the analysis. Both the standard lag-cross panel model and the random-origin lag-cross panel model were fitted to the data. The standard lag-cross panel model evaluates relationships between variables at different time points, capturing both lagged effects (influence of past values) and cross-lagged effects (inter-relationships between variables). This model typically employs fixed or random effects to control for unobserved individual heterogeneity but does not account for variations in initial conditions. In contrast, the random-origin lag-cross panel model enhances this framework by accommodating differences in baseline levels among individuals, thereby allowing for the examination of individual trajectories and variations in relationships from distinct starting points. This model is particularly advantageous in contexts characterized by substantial variability in initial conditions, as it effectively captures individual-specific dynamics. To assess the goodness of fit of the models, the study utilized several fit indices, including the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), and Root Mean Square Error of Approximation (RMSEA).

کلیدواژه ها:

Cross-Lagged Panel Model ، Longitudinal Data Analysis ، Data Mining ، Dental Science ، Psychological Variables