How to Optimize Online Mixed-Device Surveys: The Effects of a Messenger Survey, Answer Scales, Devices and Personal Characteristics

Abstract

The goal of this research was to determine the best way to present mixed-device surveys. We investigate the effect of survey method (messenger versus regular survey), answer scale, device used, and personal characteristics such as gender, age and education on break-off rate, substantive answers, completion time and respondents’ evaluation of the survey. Our research does not suggest that a messenger survey affects mixed-device surveys positively. Further research is necessary to investigate how to optimally present mixed-device surveys in order to increase participation and data quality.