Abstract:

The reasonable and full use of user data is an important part of the Internet economy. Data collection is inevitably involved with user privacy information and requires user authorization, which is the procedure of privacy disclosure. Although it is well-documented that users’ regulatory focus type is a robust predictor of privacy disclosure intention, few studies have explored the underlying decision-making mechanism. Group-level studies based on objective data are also limited. In order to further explore the effects of regulatory focus and regulatory fit on app users’ privacy disclosure, this project considers privacy disclosure as a decision-making process. Privacy disclosure decisions largely depend on the tradeoff between costs and benefits, which is affected by the user’s regulatory focus. First of all, Study 1 explores the effects of individual regulatory focus types (promotion versus prevention) on app users’ privacy disclosure decision-making preferences through a situational simulation experiment (Experiment 1). Then, the titration experimental paradigm, which has previously been used in decision-making fields, is adopted in Experiment 2 to quantitatively represent the indifference point in the tradeoff between privacy disclosure’s costs and benefits. After that, it is examined whether users with promotion (versus prevention) regulatory focus have significantly higher (versus lower) indifference points. If so, users with a promotion regulatory focus tend to provide personal data for smaller benefits. Users’ privacy disclosure preferences are also influenced by contextual factors such as the information frame. Different from previous studies that examined personal traits and information features separately, in Study 2, the user’s regulatory focus and information frame (gain versus loss) are combined to find matching effects (regulatory fit). Experiment 3 explores the regulatory fit effects in nudging app users’ privacy disclosure. This experiment also tests whether the user’s perceived uncertainty mediates the relationship between regulatory fit and privacy disclosure preference. According to the Elaboration Likelihood Model, the level of message elaboration influences the persuasion effect. Following a high elaboration process, changed attitudes are more likely to guide behavior and persist over time. Thus, Experiment 4 explores whether users adopt a higher (versus lower) elaboration level for regulatory fit (versus nonfit) privacy authorization information. Specifically, the experiment tests whether the participants who receive strong arguments in the matched (versus mismatched) condition tend to show a stronger privacy disclosure preference as well as lower perceived uncertainty. The above individual-level experiments are advantageous in that they control for variables and causal interpretation; however, there are some deficiencies in the sample size and ecological validity. Accordingly, in Study 3, several participants’ regulatory focus questionnaire scores, as well as their Weibo accounts, are first selected. Then, machine learning classification algorithms are used to train a prediction model of the users’ regulatory focus types based on the data obtained from the questionnaire scores and the corresponding Weibo text contents (Experiment 5). Privacy-related behaviors not only occur in the authorization stage but can also be reflected in daily usage data. Experiment 6 evaluates the explicit and implicit privacy disclosure level by analyzing the user’s microblog profile and original texts. A machine learning association algorithm can be used to finally output the association rules and the frequent item sets of users’ regulatory focus types and of their privacy disclosure level. This project helps explain the mechanism of regulatory focus in the process of privacy authorization from the individual level to the group level. From the perspective of decision-making, it provides an integrated explanation of privacy disclosure from the stages of intention to behavior. In addition, the results are expected to have potential application values in modifying the designs of privacy authorization information.

Key words: regulatory focus, regulatory fit, privacy disclosure, message framing, App user