A Psychometric Scale to Measure Individuals' Value of Other People's Privacy (VOPP)

AI Ethics, Fairness & AccountabilityPrivacy by Design & User ControlPrivacy Perception & Decision-MakingPrivacy Policy MakersHCI Researchers

Title of the Paper

A Psychometric Scale to Measure Individuals’ Value of Other People's Privacy (VOPP)

Paper Information

  • Research Area: Privacy Protection, Psychometrics, Technology Adoption
  • Keywords: Privacy, Data Tracking, Scale Development, Others' Data Privacy, Technology Adoption, Behavior Prediction, Psychometrics, Interaction Design, Data Sharing Risks, Privacy-Enhancing Technologies

Research Background and Problem

  • Issues and Challenges: With technological advancements, people increasingly share others' private data unintentionally, such as uploading photos of others on social media or using apps that require access to contact data. This issue of "interdependent privacy" has been under-researched. Furthermore, there is a lack of studies on quantifying users' valuation of others' privacy.
  • Significance: While individuals can typically control their own privacy, privacy protection is often overlooked when it comes to the collection and sharing of others' information. Such behaviors can lead to privacy crises at societal and national levels, such as large-scale facial recognition and misuse of private data.
  • Research Motivation: To establish a psychometric tool for studying the relationship between user behavior and technology adoption, and to support the design of technologies and interventions that protect others' privacy.
  • Related Work: Previous studies have primarily focused on users' concerns and behaviors regarding their own privacy. While some research has addressed issues like anonymization and permission control, there is insufficient exploration of how to quantify and understand users' valuation of others' privacy.

Solution

  • Methodology and Innovations:

    • Proposed a new psychometric construct: "Value of Other People's Privacy" (VOPP).
    • Developed and validated a psychometric scale (VOPP Scale) to quantify the degree to which users prioritize protecting others' privacy.
    • The innovation lies in emphasizing the stable, cross-context construct of "valuing others' privacy," rather than merely assessing privacy attitudes or concerns.
  • Implementation Steps and Techniques:

    1. Construct and Item Development:
      • Defined "value of others' privacy" as "the importance individuals assign to protecting others' information."
      • Synthesized theories, literature review, and expert opinions to generate an initial pool of 87 questionnaire items.
    2. Scale Development:
      • Used statistical analyses (e.g., item-total correlation, factor analysis) to refine items, resulting in a unidimensional scale with 13 final items.
      • Validated the scale's internal consistency and reliability.
    3. Scale Validation:
      • Conducted three empirical studies to validate the scale's internal consistency (Cronbach’s Alpha) and structural validity (factor analysis).
      • Tested convergent and discriminant validity to ensure the scale appropriately correlates with related behaviors and psychological constructs while remaining distinct from unrelated constructs.

Research Outcomes

  • Specific Results:

    1. Scale Development and Validation:
      • Developed a unidimensional psychometric scale (VOPP) with 13 specific items to measure the degree to which users value others' privacy.
      • The scale demonstrated high internal consistency (Cronbach’s Alpha = 0.92), indicating that all items are driven by the latent construct of "valuing others' privacy."
    2. Relationship with Other Psychological Constructs:
      • The VOPP scale showed a significant correlation with "altruistic motivation" (Kendall’s Tau correlation coefficient = 0.21), indicating a link between valuing others' privacy and prosocial motivations.
      • The scale was uncorrelated with "self-interest motivation," further validating its construct validity.
  • Advantages:

    • Compared to existing privacy attitude or concern scales, the VOPP scale is more stable and less influenced by situational factors.
    • The tool can predict technology adoption-related behaviors and provides data support for designing privacy-enhancing technologies (PETS).
    • It enables cross-cultural and cross-context research, revealing differences in privacy behaviors across populations.
  • Experimental and Evaluation Results:

    • Based on a sample of 1,450 participants, the scale demonstrated high predictive power and stability across multiple survey experiments.
    • Results showed that individuals who place a high value on "others' privacy" exhibit more conscientious behaviors in privacy protection (e.g., seeking permission before sharing others' photos).
  • Limitations and Future Directions:

    1. Limitations:
      • Data collection was concentrated in the United States, limiting generalizability to other cultural contexts.
      • The development process did not address systematic differences across specific contexts (e.g., data types).
    2. Future Directions:
      • Extend the scale to cross-cultural research.
      • Explore the impact of different data types and specific scenarios on the VOPP construct.
      • Conduct comparative studies on the interaction between "privacy value" and factors like "risk perception."
      • Investigate how interventions can enhance individuals' valuation of others' privacy, thereby promoting privacy-protective behaviors.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/95802/2023

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3544548.3581496
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2023
emoji_events
Award
No award tagged
group
Authors
4 authors
sell
Subtopics
AI Ethics, Fairness & Accountability, Privacy by Design & User Control, Privacy Perception & Decision-Making
work
Professions
Privacy Policy Makers, HCI Researchers
article
Content Status
Full text indexed
hub
Related Papers
10 related papers