Year: 2025
Type: conference-paper
Abstract: As software development gradually operates under the scrutiny of privacy regulations such as GDPR and CCPA, developers are required not only to implement robust privacy practices but also to communicate these practices to users. Privacy labels emerge as a potential solution to help users understand these data practices. The major app platforms require developers to disclose their apps’ data practices through standardized privacy labels. In this paper, we investigate user perceptions of privacy labels through a comprehensive analysis of online discussions and a structured user study. First, we analyze 2,880 Reddit comments related to privacy labels, using a chain-of-thought LLM classifier to isolate 548 privacy-label–specific comments and BERTopic clustering to surface five core user concerns about privacy labels: Transparency, Trust in self-reporting, Messaging-app practices, App Store permissions, and Data-category accuracy. To complement our Reddit analysis, we conduct a user study with 50 participants. The study reveals that 76% of the participants indicated that privacy labels influence their app download decisions and the amount of data practice in the privacy label is the most significant factor.
Authors: Jingwen Yan, Mohammed Aldeen, Jalil Harris, Kellen Grossenbacher, Song Liao, Feng Luo, Long Cheng
Institutions: Clemson University, Texas Tech University
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
Awards: Collaborative Research: SAI-R: Integrative Cyberinfrastructure for Enhancing and Accelerating Online Abuse Research, Collaborative Research: CISE Core Small: HCC: Towards Accessible and Safe Human-VPA Interaction in Smart Home Environments, Collaborative Research: SaTC: EDU: Developing Curriculum, Hands-on Labs, and Research Platform to Engage Students in Privacy Compliance Education, CAREER: Ensuring Privacy, Inclusiveness, and Policy Compliance in the Era of Voice Personal Assistants
OpenAlex ID: W4416307500