Dynamical Properties and Conceptual Interpretation of Latent Change Score Models.

Pablo F Cáncer,Mar J F Ollero,Emilio Ferrer,Eduardo Estrada

doi:10.3389/fpsyg.2021.696419

Pablo F Cáncer, Mar J F Ollero + Show 2 more

Open Access

https://doi.org/10.3389/fpsyg.2021.696419

Copy DOI

Abstract

Latent Change Score models (LCS) are a popular tool for the study of dynamics in longitudinal research. They represent processes in which the short-term dynamics have direct and indirect consequences on the long-term behavior of the system. However, this dual interpretation of the model parameters is usually overlooked in the literature, and researchers often find it difficult to see the connection between parameters and specific patterns of change. The goal of this paper is to provide a comprehensive examination of the meaning and interpretation of the parameters in LCS models. Importantly, we focus on their relation to the shape of the trajectories and explain how different specifications of the LCS model involve particular assumptions about the mechanisms of change. On a supplementary website, we present an interactive Shiny App that allows users to explore different sets of parameter values and examine their effects on the predicted trajectories. We also include fully explained code to estimate some of the most relevant specifications of the LCS model with the R-packages lavaan and OpenMx.

Highlights

A Latent Change Score (LCS) model is a latent variable dynamic model for the analysis of processes that unfold over time
We described how a developmental researcher reproduced a wide variety of change patterns using a dual LCS model (McArdle, 2001)
The common feature of all these specifications is the use of time-sequential dynamics between latent constructs to simultaneously explain both the covariance structure and the changes in the means over time. This distinctive property is what differentiates LCS models from other frameworks, such as cross-lagged panel models (CLPM), which do not focus on growth or decline patterns, or latent growth curve models (LGCM), which ignore the time-lagged dynamics. All these approaches are mathematically related, and both CLPMs and LGCMs can be obtained through respecifications of the LCS model parameters

Summary

INTRODUCTION

A Latent Change Score (LCS) model is a latent variable dynamic model for the analysis of processes that unfold over time. Based on previous literature (e.g., Ferrer et al, 2010), she hypothesizes that changes in children’s reading performance will be determined to some extent by cognitive ability level For such purpose, she decides to use a bivariate dual LCS (BLCS) model, which allows examining the interrelations between two variables as they unfold over time. This specification of couplings leads to three possible types of developmental relations: (1) reading performance and cognitive ability are mutually interrelated over time (i.e., γx = 0 and γy = 0); (2) one process having a (positive or negative) impact on the changes in the other process, but not vice versa (e.g., γx = 0 and γy = 0); and (3) both processes following dynamically independent courses over time (i.e., γx = γy = 0) This flexibility allows testing hypotheses about the directional effects between processes that develop differently across groups of individuals. All these approaches are mathematically related, and both CLPMs and LGCMs can be obtained through respecifications of the LCS model parameters (see Usami et al, 2015, 2019; Serang et al, 2019)

CONCLUSION

DATA AVAILABILITY STATEMENT