At the Frontiers of Modeling Intensive Longitudinal Data: Dynamic Structural Equation Models for the Affective Measurements from the COGITO Study

Abstract

With the growing popularity of intensive longitudinal research, the modeling techniques and software options for such data are also expanding rapidly. Here we use dynamic multilevel modeling, as it is incorporated in the new dynamic structural equation modeling (DSEM) toolbox in Mplus, to analyze the affective data from the COGITO study. These data consist of two samples of over 100 individuals each who were measured for about 100 days. We use composite scores of positive and negative affect and apply a multilevel vector autoregressive model to allow for individual differences in means, autoregressions, and cross-lagged effects. Then we extend the model to include random residual variances and covariance, and finally we investigate whether prior depression affects later depression scores through the random effects of the daily diary measures. We end with discussing several urgent-but mostly unresolved-issues in the area of dynamic multilevel modeling.

Analysis of dynamic, bidirectional associations in older adult physical activity and sleep quality

“…Default uninformative priors were used; that is, no prior assumptions were made about the associations between the PA and sleep quality variables. Model convergence was ensured by checking that the potential scale reduction was close to 1 and that the trace plots did not contain trends, spikes or other irregularities (Hamaker, Asparouhov, Brose, Schmiedek, & Muthen, ).…”

A Quantitative Exploration of Two Teachers with Contrasting Emotions: Intra-Individual Process Analyses of Physiology and Interpersonal Behavior

“…Cross-correlations describe the nature of the concurrent association between two time series (i.e., at the same point in time). However, cross-correlation analysis does not take into account relationships over time and (the in our case large) autocorrelations, which may lead to overestimation of the cross-correlations (see Hamaker et al, 2018;Sadler et al, 2009). Dynamic Structural Equation Modeling (DSEM; Asparouhov et al, 2017) is a new technique that gives insight in cross-lagged relations between variables, that is, how one variable is associated with other variables at the previous time point, while controlling for their own value at the previous time point.…”

“…The nature of this coupling, again, represents a potential predictor of macro-level outcomes such as general feelings of burnout (cf. Hamaker, Asparouhov, Brose, Schmiedek, & Muthén, 2018).…”

Annual Research Review: Ecological momentary assessment studies in child psychology and psychiatry

“…Individualized estimates can be summarized and modeled as a function of between‐person characteristics (e.g., baseline age, sex) in a two‐step approach. Similarly, Bayesian multilevel models with time series elements may allow simultaneous estimation of temporal complexity and heterogeneity (Depaoli & Clifton, ; Hamaker, Asparouhov, Brose, Schmiedek, & Muthén, ). Advances in dynamic modeling also offer strategies for handling complexities present in EMA designs, including models that can capture non‐linear trends, decaying lag functions across continuous time, and state‐space grids defining the dynamic behavior of a two‐variable system (Hamaker et al, ).…”