Increasing the Acceptance of Smartphone-Based Data Collection.

Abstract

To study human behavior, social scientists are increasingly collecting data from mobile apps and sensors embedded in smartphones. A major challenge of studies implemented on general population samples, however, is that participation rates are rather low. While previous research has started to investigate the factors affecting individuals' decision to participate in such studies, less is known about features of the study design which are under the researcher's control and can increase the acceptance of smartphone-based data collection methods. Guided by the Technology Acceptance Model, we varied study characteristics in a vignette experiment to examine their effect on individuals' willingness to download a research app on their smartphone. Data were collected from 1,876 members of the NORC AmeriSpeak Panel, a probability-based panel of the general population aged 18+ in the United States. Respondents were randomly assigned to eight vignettes and, after each vignette, were asked to rate their willingness to participate in the described hypothetical study. The results show that individuals are more willing to participate in smartphone-based studies where they have some control over the data collection process, by having the option either to temporarily switch off the data collection or to review the data before submission. Furthermore, they are more willing to participate in research to which they are invited via postal letter rather than receiving a postal letter plus a phone call from an interviewer who walks them through the app installation. Finally, unconditional incentives increase their willingness to engage with smartphone-based data collection over conditional incentives.