Chandraketu Singh
Decision Sciences Area, Jaipuria Institute of Management Lucknow, Lucknow, 226010, Uttar Pradesh, India.
Vijay Lakshmi Singh
OB & HR Management Area, Jaipuria Institute of Management Lucknow, Lucknow, 226010, Uttar Pradesh, India.
Sanjay K. Singh
Economics & Business Environment Area, Indian Institute of Management Lucknow, Lucknow, 226013, Uttar Pradesh, India.
DOI https://doi.org/10.33889/IJMEMS.2025.10.6.090
Abstract
Sensitive topics such as workplace misconduct, corporate fraud, unethical decision-making, diversity and inclusion challenges, and leadership accountability often present significant obstacles in survey research. Respondents may hesitate to disclose truthful information due to fear of negative consequences, damage to their professional reputation, or concerns about confidentiality. Such apprehensions frequently lead to non-response or biased responses, compromising data quality. To address these challenges, this study introduces two scrambled randomized response techniques specifically designed to enhance the accuracy of estimates for population means involving quantitative sensitive variables. These methods are designed to enhance respondent confidentiality while ensuring greater accuracy in the collected data. Empirical studies have been conducted to validate theoretical findings and assess the efficiency of the techniques. The results demonstrate that the proposed techniques offer superior efficiency and stronger privacy protection compared to the existing method. The findings highlight the practical significance of these methods for researchers and practitioners working with sensitive survey topics and encourage for their adoption in future large-scale data collection initiatives.
Keywords- Sensitive characteristics, Poisson distribution, Privacy protection, Empirical study.
Citation
Singh, C., Singh, V. L. & K. Singh, S. (2025). New Randomization Technique to Estimate the Population Mean of Quantitative Sensitive Variable. International Journal of Mathematical, Engineering and Management Sciences, 10(6), 1953-1966. https://doi.org/10.33889/IJMEMS.2025.10.6.090.
