A primer on intensive longitudinal psychometrics

Many intensive longitudinal studies are interested in topics that are not always amenable to direct physical measurement and instead are often theorized as latent constructs (e.g., affect, emotion, mood, behavior, personality). However, several recent reviews suggest that few empirical studies using scales to measure these constructs consider or report psychometrics when working with intensive longitudinal data (only 15-50%). Furthermore, a recent survey reported that many researchers working with intensive longitudinal data do not consider psychometrics because they either do not know how or are unaware that such methods exist. This is potentially problematic because it can be unclear whether momentary shifts in the variables represent true, construct-relevant changes or merely measurement error. Therefore, the goals of the current paper are to (a) provide an overview of the existing intensive longitudinal psychometric methods, (b) demonstrate how these methods can be applied, and (c) show how including psychometric methods can supplement and fortify intensive longitudinal analyses. Specifically, the paper covers four main topics: reliability, construct validation, measurement invariance, and measurement error modeling. The paper is also accompanied by a Shiny application to facilitate access and implementation of these methods. Supplemental materials provide in-depth walkthroughs of software applications in R and Mplus with annotated code for the empirical examples in the paper.