Leveraging randomized response frameworks to enhance sensitive data acquisition: implications for data regarding...

Area of Science:

• Public Health Research
• Survey Methodology
• Data Privacy

Background:

• Traditional surveys struggle with sensitive data due to respondent bias and privacy concerns.
• Social desirability and over-estimation are prevalent in sensitive public health data collection.
• Accurate quantification of sensitive numerical variables is challenging while maintaining respondent anonymity.

Purpose of the Study:

• To introduce and evaluate multi-stage randomized response models (MRDRMs) for sensitive data collection.
• To enhance privacy protection and estimation efficiency in surveys.
• To address challenges in obtaining precise estimates for sensitive quantitative variables.

Main Methods:

• Development of two-stage and three-stage MRDRMs.
• Utilizing two randomized response mechanisms for improved privacy and efficiency.
• Theoretical analysis via Monte Carlo simulations.
• Empirical validation through a cross-sectional survey in Punjab, Pakistan.

Main Results:

• MRDRMs significantly improve the precision and relative efficiency of estimates compared to conventional methods.
• Demonstrated substantial improvement in estimation accuracy and reduced variance.
• MRDRM-I and MRDRM-II proved effective for sensitive data collection, enhancing privacy and response validity.

Conclusions:

• MRDRMs offer a robust framework for collecting and evaluating sensitive data in public health.
• These methods strengthen data reliability, reduce participant psychological strain, and support evidence-based policy.
• Proposed models are applicable to public health surveillance, epidemiological studies, and social science research.