Data, Data, Everywhere: Quantifying Software Developers' Privacy Attitudes

摘要

Understanding developers' attitudes towards handling personal data is vital in order to understand whether the software they create handles their users' privacy fairly. We present the results of a study adapting an existing user-focused privacy concern scale to a software development context and running it with a sample of 123 software developers, in order to validate it and develop a model for measuring the extent to which a software developer is (dis)favorable to ensuring their users' privacy. The developed scale exceeds thresholds for internal reliability (α ＞ .8), composite reliability (CR ＞ .8), and convergent validity (AVE ＞ .6). Our findings identified a model consisting of three factors that allows for understanding of developers' attitudes, including: (1) informed consent, (2) data minimization, and (3) data monetization. Through analysis of results from the scale's deployment, we further discuss mismatches between developers' attitudes and their self-perceived extent of properly handling their users' privacy, and the importance of understanding developers' attitudes towards data monetization.