A Comparison of Errors in Web Surveys Completed Through PC and Mobile Devices

It was observed that some respondents already try to complete web surveys via mobile devices, even when this is unintended. However, we can expect an effect of the device used to complete the survey on the answers and their quality. Indeed, the devices vary at several levels: size of the screen, kind of keyboard, place and conditions (presence of other persons etc) in which the respondents use them, and so on. Therefore, in order to reduce measurement errors and increase comparability of the data collected through different devices in web surveys, it is necessary to study more deeply the differences in errors that appear when respondents answer through PC or mobile devices. In this presentation, we will report the results of an experiment conducted in Spain with the online fieldwork company Netquest, in which the same respondents participated twice to the same survey, once by PC and once by smartphone (optimized or not optimized smartphone versions). In our experiment we also considered two control groups: the respondents of these groups participated to both surveys using the same device.