Online Program Home
My Program

Abstract Details

Activity Number: 308
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 8:30 AM to 10:20 AM
Sponsor: Health Policy Statistics Section
Abstract #321364 View Presentation
Title: Real-Time Modeling of Variation in Longitudinal Momentary Self-Report Data
Author(s): Trent L. Lalonde* and Elysia Clemens
Companies: University of Northern Colorado and University of Northern Colorado
Keywords: longitudinal ; dispersion modeling ; real-time ; momentary assessment

In recent years attention has been paid to ecological momentary assessment (EMA) data. In such studies, data are collected from subjects in-the-moment, instead of as summary recall self-report. One common method for collecting such data is through short surveys administered via smartphones. When such data are collected repeatedly over a short period of time, the interest is typically not in assessing an intervention, but to gain perspective on the tendencies of a specific population. Another goal of short-term longitudinal data collection is to measure variation in subjects' responses. Dispersion in longitudinal data is often considered nuisance parameters in the analysis. However, important information about individuals behaviors can be obtained by modeling variation using short-term longitudinal momentary data. In this study methods of modeling variation are explored, from simple visuals and descriptive statistics to joint models of both the mean and dispersion. The models have been implemented through a web-based interface to allow for real-time analysis of such data. As an example, self-reported behavior measures are used to describe and make inferences about subject variation.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association