Enrollment and Compliance in an Ecological Momentary Assessment Study of Later Life

Abstract

This study addresses enrollment and compliance in ecological momentary assessment (EMA) research by focusing on a sample of older adults with varying levels of technological familiarity. This is noteworthy because contemporary EMA studies predominantly require participants to use smartphones to record data in real time. We employ a two-stage analysis where we estimate the probability of enrollment based on sociodemographic and research design attributes and next use the output from this model to address potential selection bias in a subsequent model estimating compliance rates. We find that smartphone ownership is a major predictor of enrollment and that offering participants a loaner smartphone is insufficient to overcome this barrier. Ultimately, our findings suggest that while there are some potential design strategies that prospective researchers could implement to improve enrollment and compliance, there will likely always be some people whose attributes decrease their probability of enrollment and compliance regardless of intervention.