A tutorial on analyzing ecological momentary assessment data in psychological research with Bayesian (generalized) mixed-effects models

Abstract

In this tutorial, we introduce the reader to Bayesian modeling of ecological momentary assessment (EMA) data as applied in psychological sciences. We discuss scientific and practical advantages of the Bayesian approach over frequentist methods. We demonstrate how Bayesian statistics can help EMA researchers to (1) incorporate prior knowledge and beliefs in analyses, (2) fit models with a large variety of outcome distributions that reflect likely data-generating processes, (3) quantify the uncertainty of effect size estimates, and (4) quantify the evidence for or against an informative hypothesis. We present a workflow for Bayesian analyses and provide illustrative examples based on EMA data, which we analyze using (generalized) linear mixed-effects models to test whether daily self-control demands predict three different alcohol outcomes. All examples are reproducible, with data and code made available at https://osf.io/rh2sw/. Having worked through this tutorial, readers should be able to adopt a Bayesian workflow to their own analysis of EMA data.

Characterizing the Momentary Association Between Loneliness, Depression, and Social Interactions: Insights from an Ecological Momentary Assessment Study

“…All models were estimated using the brms package (version 2.17.0; Bürkner, 2021) for R (version 4.1.1; R Core Team, 2022). We assessed model convergence and fit by inspecting potential scale reduction factor (𝑅 " ) values, effective sample sizes, trace plots, and posterior predictive checks (see Dora et al, 2022). R code for all analyses can be found on this project's OSF page (https://osf.io/7rkfm).…”

Zero inflation in intensive longitudinal data: why is it important and how should we deal with it?

“…For instance, Dora, Piccirillo, et al (2022) conducted a meta-analysis on the relationship between affect and alcohol in EMA research and reported that the number of drinks reported in EMA studies is often zero-inflated. This phenomenon occurs because individuals tend to consume alcohol on only 20-40% of days, even in studies involving regular drinkers (Dora, McCabe, Lissa, Witkiewitz, & King, 2022;Dora, Piccirillo, et al, 2022). In our review conducted in May 2023, which encompassed 15 published articles from the last three years and was executed using Google Scholar with the keywords "alcohol use disorder" and "ecological momentary assessment", we observed that only one article (Tovmasyan, Monk, & Heim, 2022) acknowledged the presence of zero-inflation.…”