Mobile data donations: Assessing self-report accuracy and sample biases with the iOS Screen Time function

Abstract

With digital communication increasingly shifting to mobile devices, communication research needs to explore ways to retrieve, process, and analyze digital trace data on people’s most personal devices. This study presents a new methodological approach, mobile data donations, in which smartphone usage data is collected unobtrusively with the help of mobile log data. The iOS Screen Time function is used as a test case for gathering log data with the help of screenshots. The study investigates the feasibility of the method, sample biases, and accuracy of smartphone usage self-reports on a general population sample of Dutch citizens ( n=404). Importantly, it explores how mobile data donations can be used as add-ons or substitutes for conventional media exposure measures. Results indicate that (a) users’ privacy concerns and technical skills are crucial factors for the willingness to donate mobile log data and (b) there is a strong tendency for underreporting of smartphone usage duration and frequency.

A systematic review and meta-analysis of discrepancies between logged and self-reported digital media use

“…Our findings, however, indicate only a modest association between self-reports and usage logs, leading us to conclude that self-report measures of media use may not be a valid stand-in for more objective measures. Notwithstanding the potential biases affecting log-data 5,35,39,40 , if these measures are taken to be a valid reflection of actual usage 5,21,32,41,85 , our findings raise important concerns about the validity of findings and conclusions across many areas of the social sciences in which self-reported media use is a central outcome or explanatory variable.…”

“…These data suggest that self-reported and logged measures, rather than simply serving as different ways to measure media use, may in fact capture distinct constructs 31,38 . Log-based techniques, although they are not without their own biases 5,35,39,40 , provide a more direct, and likely more accurate measure of media use than self-report 5,21,32,41 . As such, there exists a need to systematically assess whether self-reported media use is an accurate indicator of actual usage patterns.…”

The Accuracy and Validity of Self-Reported Social Media Use Measures Among Adolescents

“…We found slightly more accurate and valid estimates for adolescents who only used social media via their phones, but these estimates were not strikingly different for adolescents who used both the phone-and web-based applications. Future studies should extend our study by comparing different sources of objective digital trace data with each other (e.g., iOS screen time data or social media archival data donations) (Boeschoten, Van Driel, Oberski, & Pouwels, 2020;Ohme, Araujo, de Vreese, & Piotrowski, 2020).…”

Effects of depression, anxiety and screen use on adolescent substance use

“…Data are self-reported which may raise concerns regarding accuracy. Regarding screen time, research has found individuals tend to underreport their smartphone usage and factors such as short-term exposure, inconsistent usage, and multi-tasking can affect accurate recall of time spent on a given device ( Ohme et al, 2020 , Vanden Abeele et al, 2013 ) Research on self-report for substance use has been mixed. While studies have found that up to a quarter of respondents’ self-reports are not corroborated by urinalysis, other studies have found that accuracy of self-report data can be enhanced by implementing specific methodology ( Williams and Nowatzki, 2005 ).…”