Influence of privacy priming and security framing on mobile app selection. Issue 78 (September 2018)

• Record Type: Journal Article
• Authors: Chong, Isis; Ge, Huangyi; Li, Ninghui; Proctor, Robert W.
• Is Part Of: Computers & security. Issue 78(2018)
• Page Start: 143
• Page End: 154
• Publication Date: 2018-09
• Subjects: App security -- Decision making -- Mobile devices -- Safety index -- Privacy priming; Computer security -- Periodicals; Electronic data processing departments -- Security measures -- Periodicals; 005.805
• DOI: 10.1016/j.cose.2018.06.005
• Languages: English
• ISSNs: 0167-4048

Abstract

Highlights: Experiments studied the impact of privacy-related priming material and risk/safety indexes on a simulated app-selection task. Users relied on user ratings to guide their app selections, but even a single prime item increased the weighting of safety. Real-world uses include presenting users with brief, privacy-related prompts before searching for apps on the app store. Results provide additional support for including safety indexes in mobile app stores. Abstract: Mobile apps have the potential to request access to private information. Given the far-reaching negative consequences that misuse of this kind of information can lead to, interventions that may encourage safer app downloading behaviors need to be investigated. Participants were recruited through a crowdsourcing service to participate in two separate experiments designed to determine the impact of presenting safety-related priming material before selecting mobile apps. Participants completed a simulated app selection task where apps were given overall safety or risk rankings. In Experiment 1 participants were assigned to a (1) control, (2) self-relevant priming, or (3) factual priming condition. In the latter conditions participants were presented eight privacy priming items prior to the app selection task, differing in whether they were to rate their disagreement/agreement with privacy statements or to read and demonstrate understanding of eight mobile app permissions. Experiment 2 was similar to Experiment 1 but with only a single priming item presented. The obtained results showed that although the participants relied heavily on user ratings to guide their app selections, both the self-relevant and factual priming items induced the participants to take safety more into account. Further, we showed that priming can be accomplished with a single item, which holds promise for real-world applications.