Suitability of event-based prompts in experience sampling studies focusing on location changes

Abstract

Among others, location changes and activity level are indicators for state changes of patients suffering from affective disorders such as Bipolar disorder, Borderline personality disorder or depression. It is a common means to assess this information via self-report questionnaires. Usually, these are sent out either randomly throughout the day or at fixed points in time. However, this might lead to missing records of location changes. We propose to rely on event-triggers: send out self-report prompts when a location change is automatically detected. We enhanced the ESMAC application by a location change detection event. Then, we created three different study configurations for each trigger type: random, time-based, and event-based. In a three-week within-subject study we let subjects experience each trigger type in randomized order. We found statistically significant differences in favor of the event-triggers in terms of number of prompts, response rate, prompts after detected location changes, and prompts after detected activity changes. We conclude that event-triggers based on a location change detection shall be used as trigger type for experience sampling studies focussing on location or activity changes.