The Effect of Persuasive Design on the Adoption of Exposure Notification Apps: A Case Study of COVID Alert

Background: The COVID-19 pandemic began in the first quarter of 2020 necessitated the imposition of public health restrictions and the shutting down of the global economy. To slow down the spread of the coronavirus, governments worldwide rolled out nationwide contact tracing apps (CTAs) to notify people that may have been exposed to COVID-19. The emergence of new variants of COVID-19, which can cause breakthrough infections, necessitate the continued use of CTAs. However, the uptake of these apps has been low and slow worldwide. Some experts have argued that the low adoption rate of CTAs can be attributed to their minimalist design and lack of motivational features, trust- and privacy-related issues aside. However, there is little to no research to show that the incorporation of persuasive principles in the design of CTAs has the potential of increasing their effectiveness and adoption. Method: To bridge the gap in the extant literature, we conducted an empirical study on Amazon Mechanical Turk to investigate the effect of persuasive design in CTAs using the Government of Canada’s exposure notification app (“COVID Alert”) as a case study. Our study is based on 204 participants (comprising adopters and non-adopters) resident in Canada and two app designs: persuasive and control.Findings: Regarding the willingness to download the COVID Alert app, our three-way analysis of variance (ANOVA) shows there is an interaction between adoption status and app design. Among adopters, there is no significant difference between the persuasive and the control design. However, among non-adopters, there is an effect of app design (p < 0·001), with participants being more likely to download the app using the persuasive design (M = 5·37) than the control design (M = 4·57). Similarly, regarding the intention to report COVID-19 diagnosis, there is an interaction between adoption status and app design. Among non-adopters, there is no significant difference between the persuasive and the control design. However, among adopters, there is an effect of app design (p < 0·01), with participants being more likely to report their diagnosis using the persuasive design (M = 6·00) than the control design (M = 5·03). Interpretation: The results of our study show that non-adopters are more likely to download the persuasive version of a CTA (equipped with self-monitoring) than the control version. Moreover, adopters are more likely to report their COVID-19 diagnosis using the persuasive version of a CTA (equipped with social-learning) than the control version. Overall, the study shows that that CTAs are more likely to be effective if equipped with persuasive features.Funding Information: Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant (RGPIN-2017-05310) and Cybersecurity and Privacy Institute, University of Waterloo.Declaration of Interests: None to declare. Ethics Approval Statement: Our survey was approved by the Research Ethics Board of our university.