One burgeoning method to analyze the interactions between psychological variables from intensive longitudinal data is to visualize them as a temporal network, which illustrates how different psychological components influence one another from one timepoint to the next. However, there are no accepted best practices for the data collection, methodology, and analyses for temporal networks. We thus conducted a systematic review to audit the methodological practices and statistical analyses of this emerging field of research. We included studies that comprised intensive time-series data, investigated psychological variables with human subjects using temporal network analyses at a group level, and were published in peer-reviewed international journals. We identified 26 studies and extracted numerous variables relating to data collection, temporal network estimation and visualization, open science practices and research quality (for our preregistration, see https://osf.io/9va82/). During the presentation, we will discuss the most commonly used analytical approaches reported in these studies. Moreover, because most studies rely on reanalyzed data and do not include information about the development or validity of the ambulatory assessment items, we also formulated a set of guidelines to help the field to move forward.
Blanchard, A., Contreras Cuevas, A. M., & Heeren, A. (2021). Auditing the research practices and statistical analyses of the temporal network approach to psychological constructs: A scoping review. Society of Ambulatory Assessment Conference 2021, Zurich, Swizterland. https://hdl.handle.net/2078.5/109168