Privacy Dashboards: The Impact of the Type of Personal Data and User Control on Trust and Perceived Risk

Abstract

Website owners often collect personal data to, among other things, advertise more efficiently and to analyze and increase sales. They can inform users in various ways about what data they collect and how they process it. This study focuses on the use of privacy dashboards, which are increasingly present on websites, but not very well studied yet. In this study, making use of an experimental webshop and various interviews, we investigate which elements of privacy dashboards increase customers' trust and reduce the perceived privacy risks. The results indicate that the presence of derived data, such as average values or profile classifications, had a more negative impact on trust and perceived privacy risk than inferred data, such as interest probabilities - and both categories had a larger impact than provided and observed, unprocessed user data. Further, it emerges that a greater degree of control leads to a somewhat greater trust. The study confirms the benefits of a privacy dashboard in addition to a privacy policy and provides guidelines on the level of abstraction on which user data should be presented.