Innovative Approaches to Sensitive Question Surveys: A Hybrid Binary Model Using Paillier Encryption and Randomized Response Techniques

In sensitive question surveys, respondent privacy is a major concern that can influence both participation rates and data accuracy. The key concern is the tendency of the respondents to lie or to not respond to the question of sensitive nature due to a fear of judgment of the surveyor.

Traditional methods, such as the Randomized Response Technique (RRT), help preserve privacy but may compromise estimation accuracy. On the other hand, Paillier encryption—a widely used asymmetric encryption protocol—offers high security and perfect accuracy but can be computationally expensive and still relies on the integrity of the decryption key holder.

In this webinar, we introduce a novel hybrid model combining the Paillier encryption scheme with Warner's Randomized Response Technique (RRT) as an attempt to enhance both privacy and efficiency in the process of data collection for sensitive binary response questions.

Participants will gain insights into:

• The fundamental attributes of the Paillier encryption protocol and its application in secure data collection.
• The fundamentals of RRT for mitigating social desirability bias in sensitive questions.
• The structure and benefits of a hybrid Paillier-RRT model.
• Simulation results illustrating the model's performance in terms of Mean Squared Error (MSE).
• A discussion on the concept of overall privacy under the hybrid Paillier-RRT model.

This webinar is ideal for statisticians, data scientists, and researchers interested in advancing secure and accurate methods for surveying sensitive topics.