Investigating moderation effects at the within-person level using intensive longitudinal data: A two-level dynamic structural equation modelling approach in Mplus

Abstract

Recent technological advances have provided new opportunities for the collection of intensive longitudinal data. Using methods such as dynamic structural equation modelling, these data can provide new insights into moment-to-moment dynamics of psychological and behavioural processes. In intensive longitudinal data (t > 20), researchers often have theories that imply that factors that change from moment to moment within individuals act as moderators. For instance, a person’s level of sleep deprivation may affect how much an external stressor affects mood. Here, we describe how researchers can implement, test and interpret dynamically changing within-person moderation effects using dynamic structural equation modelling as implemented in the structural equation modelling software Mplus. We illustrate the analysis of within-person moderation effects using an empirical example investigating whether changes in spending time online using social media affect the moment-to-moment effect of loneliness on depressive symptoms. We provide annotated Mplus code, enabling researchers to better isolate, estimate and interpret the complexities of within-person interaction effects.

The role of student–teacher relationships in the association between negative parenting practices and emotion dynamics – Combining longitudinal and ecological momentary assessment data

“…To investigate the association between negative parenting, student–teacher relationships, NA, EI and EL, Dynamic Structural Equation Modeling (DSEM; Asparouhov et al, 2018) was used, as implemented in Mplus 8.8 (Muthén & Muthén, 2017). DSEM combines structural equation modeling, time‐series analysis and multilevel modeling (McNeish & Hamaker, 2020; Speyer et al, 2022). NA was modeled using a univariate DSEM that disaggregates the data into latent within‐ and between‐person factors.…”

Social Media Self-Effects on Adolescents’ Self-Esteem: A 100-Day Diary Study

“…To test our within-person moderation hypotheses (H2 and H3), we added withinperson moderators to our model estimation of H1 (i.e., perceived positivity of peer feedback and feedback preoccupation; Model 2 and Model 3, respectively). Before model estimation, we extracted the within-person components to include them as latent person-mean centered variables in the model (following procedures of Speyer et al, 2024). In the DSEM model, we then regressed self-esteem on the interaction term for H2 (i.e., the positivity of social media self-presentation * perceived positivity of peer feedback) and H3 (i.e., the positivity of social media self-presentation * feedback preoccupation).…”

Exploring the effect of ADHD traits on the moment‐to‐moment interplay between provocation and aggression: Evidence from dynamic structural equation modeling

“…Such effects can illuminate if, for example, individuals with higher levels of ADHD traits show greater aggressive inertia (e.g., if an individual, on having their aggression triggered, maintains their heightened level of aggression for a longer time period). Cross‐lagged effects, in contrast, refer to the lagged associations between different variables; whereby significant directional effects of variables on one another over time are identified (e.g., the effect of being provoked at one time point on behaving aggressively at the next (Hamaker et al, 2018; Speyer, Murray, et al, 2022). Individual differences in these can be explored to examine if, for example, individuals with higher levels of ADHD traits have stronger aggressive reactivity to being provoked (e.g., if feeling provoked at a previous time point can predict future aggression toward others to a greater extent for those with higher levels of ADHD symptoms).…”