Advancing the Science & Practice of Ecological Momentary Assessment
Arthur Stone (Univ Southern California)
Joshua M. Smyth (site PI, Penn State)
NIH Reporter listing ClinicalTrials.gov registration coming within 21 days of first participant being enrolled
This proposal is submitted in response to the program announcement PAR-16-260, “Methodology and Measurement in the Social and Behavioral Sciences,” with the specific goal of improving measurement in real-time data capture of experiences. Ecological Momentary Assessment (EMA) is a family of techniques for repeatedly capturing self-reports while people go about their everyday lives, and is particularly useful for assessing subjective states. It aims to achieve accurate and granular measurement of symptoms, well-being, affect, beliefs, and the social and physical environment over time in natural contexts. Although EMA methods are widely used in behavioral, social, and medical research, there are critical gaps in our understanding that must be resolved for EMA methods to be optimally understood and applied. This proposal examines five issues to provide necessary conceptual information about EMA and practical tools for conducting rigorous, high-quality, and replicable EMA studies; it employs several novel approaches for achieving these goals. The main issues to be investigated include: understanding how respondents interpret and answer momentary questions by conducting real-time, qualitative interviews; determining how various EMA protocol factors (e.g., length of assessment, number of days of assessments) impact selection bias of respondents and influence momentary compliance, and the development of a tool for efficiently evaluating the impact of particular designs; testing a model for a new generation of EMA measurement tools using modern psychometric principles (Item Response Theory), and evaluating the possibility of computerized adaptive tests for EMA; examining a core feature of EMA – that EMA reports are less biased than recall reports – by comparing both types of self-reports with an objective standard; and, finally, experimentally examining factors that have been hypothesized to bias EMA reporting, based on the idea that the act of self-monitoring and reporting may influence the nature of reports (i.e., measurement reactivity).