A magnifying glass for the study of coupled developmental changes: Combining psychological networks and latent growth models

Abstract

In recent years, methodological advances for analyzing developmental data are coming thick and fast. Two of the most popular and rapidly developing frameworks are (i) longitudinal structural equation modeling and (ii) network modeling. The present paper outlines the incremental gain in what we can learn from data about co-developing skills and challenges when using these two frameworks in tandem. First, we discuss the proposed analytic paradigm in the context of fundamental questions in developmental psychology. Second, we present two different paths to formalize such questions, introducing, first, a recently developed network model for longitudinal panel data and, second, the notion of growth parameter networks based on latent growth curve models. Used in tandem, they can provide new insights into the longitudinal co-development of developmental domains. Specifically, we focus on integrating growth parameters from latent growth curve models into networks and analyzing them as such. Third, we illustrate these analytic steps with an empirical example using longitudinal data from the Millenium Cohort Study (N=7623). As illustrated and discussed in the real data example, the proposed approach offers a magnifying glass to the study of coupled developmental changes. Teasing apart the processes underlying the heterogeneity of childhood development can, in turn, add to substantive developmental theory.

Transdiagnostic symptom dynamics during psychotherapy

“…As such, the findings from this study can only be taken as potential causal associations and inferences about intra-individual processes of change during psychotherapy in this sample relate to a hypothetical average person. Identification of subgroups combined with latent growth network modelling 52 may offer additional insights into the change mechanisms during psychotherapy, help us understand how different subgroups respond to therapy, and what specific factors may contribute to better outcomes.…”

Transdiagnostic Symptom Dynamics during Psychotherapy

“…As such, the findings from this study can only be taken as potential causal associations and inferences about intra-individual processes of change during psychotherapy in this sample relate to a hypothetical average person. Identification of subgroups combined with latent growth network modelling 52 may offer additional insights into the change mechanisms during psychotherapy.…”

More alike than different: Novel methods for measuring and modeling executive function development

“…Further, as methods for appropriately modeling longitudinal data emerge, network analysis provides an avenue for understanding the potential reciprocal relationships among EFs over time 32 . For example, in a separate study we are examining how growth in performance on individual tasks are connected.…”

“…We intentionally did not explicitly model the dependency of multiple observations per student that occurs with longitudinal data in either analytic approach. While it is possible to model this dependency using factor analysis, development of network models that can handle longitudinal data are still in their infancy (though see 32 ). To keep the general modeling strategy consistent so that inferences would be comparable, we treated all observations as independent in both approaches.…”