Assessing the Validity of Self-Reported Data on Smartphone Usage: Comparing Subjective and Objective Data

Nowadays smartphones are the main source of access to online platforms and services and the key driver of internet proliferation both in Russia and abroad. Expanding the functionality of smartphones and registering statistics on their usage creates a new and important source of digital data on user behavior for researchers. However obtaining this type of data requires the development of new methodological approaches and solving a number of technological, legal and ethical issues. A significant step in this process is a comparative assessment of survey data based on retrospective self-reports that are traditionally used for researching online user activity with objective data on behavior based on digital traces of online activity. This article presents the results of an experiment on cross-validation of self-reported data on smartphone usage and digital behavioral data (N = 80). The data for the experiment were collected within the methodological approach referred to as “data donation”, which means that the respondents share with the researchers the data that their mobile devices have already collected. The results of the experiment show a moderate positive correlation between retrospective self-report data and digital behavioral data on mobile phone usage. There is no statistically significant time window effect for estimating the volume of smartphone usage over a short (day) and longer period (current week). It indicates a comparative stable validity of self-reporting and, consequently, its reliability-reproducibility within a week. Thus, when there is no possibility to collect and integrate digital behavioral data into the study, the use of retrospective self-reports may be warranted.