App Privacy on App Platforms: Developer Side Trade-offs, Monetization Strategies and Third-Party Services

2021 | International Conference on Information Systems | Citations: 0

Authors: Tripathi, Shagun; Kyriakou, Harris

Abstract: App platforms have been plagued with concerns for users' data privacy, with pri ... Expand

Abstract: App platforms have been plagued with concerns for users' data privacy, with prior research pointing to two key user side trade-offs: privacy vs. utility and privacy vs. price. Motivated by literature on platform network effects and data effects, we shift our focus to the developers' side, where developers' choices play a key role in the overall privacy of an app. We focus on two main factors: choice of monetization strategies and use of thirdparty services (software development kits, abbrev. SDKs). Using data on 121,808 apps on Google Play, Android platform's app market, we analyze the relationship between apps' monetization strategies and developer's requests for privacy sensitive information. Our results a) confirm prior research showing a negative relationship with request for privacy related information when apps are paid and b) show that compared to ad-based monetization, developers request for more privacy sensitive data when they monetize their apps with 'in-app' strategies (allowing future transactions to take place within the app).. Additionally, , we also investigate patterns underlying developer's choice of thirdparty services i.e., typical versus atypical choices in the deployment of third-party services. Using unsupervised text clustering on 518 SDKs as seen in our data, we find that greater deviation from typified SDK choices is strongly and positively associated with increased requests for privacy sensitive data. Taken together, our results suggest that the mechanisms driving a developer's requests for privacy sensitive requests point to a third functionality vs. privacy trade-off on the app developers' side in app economies Collapse

Topics: mobile application privacy mobile application market digital distribution Android

Methods: cluster analysis cross sectional research k-means clustering machine learning computational algorithm