"I'm not as afraid as a woman might be about sharing my exact location:" On the Intersection of Identity and Privacy Concerns in Fitness Tracking

Research Background and Issues

Issues and Challenges

• This study examines how various dimensions of social identity (gender, race, age, LGBTQ* status) influence users' perceptions and practices regarding fitness tracking privacy.
• Despite the growing popularity of fitness tracking devices and applications, which have become a multi-billion-dollar industry, the sensitive health and activity data they collect pose privacy and security risks. For example:
  1. Data may reveal users' home addresses, creating safety concerns.
  2. Third parties, such as employers and insurance companies, may misuse such data.
  3. Previous studies have shown that women and LGBTQ* individuals may face higher risks.

Significance of the Study

• Privacy issues related to fitness tracking data have garnered significant attention, but research on how different social identities shape these privacy perspectives remains scarce.
• Specific groups (e.g., women, racial minorities, LGBTQ* individuals) may face heightened data privacy risks due to cultural and social contexts.
• Against the backdrop of increasing technological adoption and data commercialization, it is crucial to ensure that no user faces greater privacy threats due to their identity.

Motivation and Related Work

• This study integrates literature on fitness tracking privacy issues with research on the impact of social identity on privacy, a combination that has not been systematically explored before.
• It addresses gaps in prior research by linking technical discussions of data privacy with social issues, aiming to guide the development of more inclusive privacy solutions in the future.

Solutions

Methods and Solutions

• The authors designed and conducted an online survey (N=322) to collect respondents' attitudes toward fitness data sharing, risk perceptions, and related risk mitigation strategies.
• Research questions include:
  1. How does identity affect comfort levels in sharing fitness data across different social contexts?
  2. Does identity influence perceptions of data-sharing risks?
  3. Does identity impact strategies for addressing privacy risks?

Innovations

• Incorporating social identity (gender, race, age, sexual orientation) into the study of fitness privacy risks, a novel approach not previously seen in the literature.
• Developing a qualitative framework for categorizing risks and mitigation strategies, providing systematic tools for analyzing user responses.

Implementation Steps and Key Techniques

1. Survey Design:
   • Contextualized through scenario-based prompts (e.g., respondents were asked to imagine themselves having similar fitness tracking habits as a sample user, Sam).
   • Focused on three main areas: comfort with data sharing, potential risks of data sharing, and possible mitigation strategies.
2. Data Collection:
   • Participants were recruited via the Prolific platform, meeting criteria such as engaging in over 150 minutes of exercise per week.
   • Ensured balanced representation across key identity characteristics (gender, race, age, LGBTQ* status).
3. Data Analysis:
   • Applied a Bayesian framework for regression analysis of comfort level data.
   • Conducted qualitative coding of risk data and used chi-square tests to compare significant differences across identities.

Research Findings

Key Findings

1. Comfort Level Analysis:
   • Women and users aged 30–39 were less willing to share fitness data with institutional groups (e.g., employers, advertisers) compared to men and users over 40.
   • 64% of participants indicated that at least one aspect of their identity influenced their comfort level with data sharing.
2. Risk Perception:
   • Participants with different identity characteristics exhibited significant differences in their perceptions of potential privacy risks. For example, Black participants more frequently mentioned "general privacy risks," while older groups were more concerned about specific physical safety risks.
   • Scores for the likelihood and severity of the same risk type varied based on identity characteristics.
3. Risk Mitigation Strategies:
   • Regardless of identity, most users preferred simple privacy settings adjustments to mitigate risks, such as "stopping data sharing with a specific group" or "disabling privacy features."
   • Non-technical offline strategies (e.g., running with a companion or carrying self-defense tools) were widely mentioned, especially in response to physical safety risks.

Comparison with Existing Solutions and Advantages

• This study goes beyond traditional technical discussions by offering a social and psychological perspective, situating privacy issues within specific social contexts.
• The survey clearly demonstrates that users with different identity characteristics have varying privacy perceptions, suggesting that future privacy designs need to be more diverse and inclusive.

Limitations and Future Directions

• Limitations:
  1. The study did not capture the nuanced effects of intersecting identities (e.g., how race and gender jointly influence privacy perceptions).
  2. The sample was drawn from the Prolific platform, which may include participants with higher technical literacy, potentially differing from the general population in privacy concerns.
  3. The study did not explore fitness tracking privacy in non-U.S. cultural contexts.
• Future Directions:
  1. Combine qualitative research to analyze the interactive effects of multiple identities in greater depth.
  2. Expand sample diversity, particularly to include underrepresented identity characteristics (e.g., Latinx individuals).
  3. Investigate personalized privacy technology designs to better meet user needs.

Overall, this study pioneers a new research domain centered on identity-based fitness privacy analysis, paving the way for the development of more inclusive and personalized data privacy solutions.