Introduction

During the last decade, the Internet penetration rate and the ownership of mobile devices, such as smartphones and tablets, have increased continuously. According to the Pew Research Center (2018a, 2018b), this trend applies to most countries, but especially to developing countries. Simultaneously, the number of respondents taking part in web surveys using mobile devices, particularly smartphones, has increased (Revilla, Toninelli, Ochoa, & Loewe, 2016).1 One reason for this phenomenon might be that people have their smartphone with them most of the time.

The rise of smartphones in web survey responding is accompanied by a large body of research investigating survey layout strategies, such as optimized and non-optimized, systematic measurement error, such as break-off rates and item non-response, and paradata, such as response times and device orientation (see, for instance, Buskirk & Andrus, 2012; Couper & Peterson, 2017; de Bruijne & Wijnant, 2013; Hohne, Revilla, & Lenzner, 2018; Mavletova, 2013; Mavletova & Couper, 2013; Revilla & Couper, 2018a, 2018b; Revilla & Ochoa, 2015; Schlosser & Mays, 2018; Wells, Bailey, & Link, 2013). In addition, smartphones allow survey researchers for the passive collection of so- called sensor data—that is, data that are collected via a variety of built-in sensors, such as accelerometers, barometers, compass, Global Positioning System (GPS) trackers, and gyroscopes. Sensor data have the potential to complement survey responses (Hohne & Schlosser, 2019; Toepoel & Lugtig, 2015) by providing information about respondents' physiological states, such as altitude level, motion, geographic orientation, and speed (see Elhoushi, Georgy, Noureldin, & Korenberg, 2017; Harari et al., 2016; Hohne & Schlosser, 2019; Toepoel & Lugtig, 2015). Data from these sensors can be collected either by JavaScript functions implemented in web survey pages or by apps installed on the smartphone.