Big Data Meets Survey Science

13Effects of Incentives in Smartphone Data Collection

Georg‐Christoph Haas^1,2, Frauke Kreuter^1,2,3, Florian Keusch^2,3, Mark Trappmann^1,4, and Sebastian Bähr¹

¹Institute for Employment Research, Nuremberg, Germany

²Joint Program in Survey Methodology, University of Maryland, College Park, MD, USA

³Department of Sociology, School of Social Sciences, University of Mannheim, Germany

⁴University of Bamberg, Bamberg, Germany

13.1 Introduction

Smartphone sensor data enable researchers to analyze phenomena that cannot be investigated with survey data alone (e.g. Sugie 2018). However, smartphone data may include very sensitive information, e.g. on geolocation or app usage, which users may perceive as too private to share with researchers. To date, very little research has systematically examined participation in studies that collect passive smartphone sensor data. To our knowledge, no study so far has examined whether the knowledge about the effectiveness of incentives in surveys also holds for smartphone sensor data collection. Common incentive amounts and incentive schemes traditionally used in surveys may not be sufficient to motivate participation in a study that collects data passively from smartphones, given that individuals may perceive sensor data as more valuable than survey data. However, since participation in passive data collection requires less effort from participants, burden (measured in time spent on data collection) is much reduced compared to regular surveys. Therefore, ...