Reliability and Validity Evaluation Based on Monte Carlo Simulations in Two-Stage Cluster Sampling on Sensitive Question
In this paper, additive randomized response technique (RRT) model was presented in application to asking questions on sensitive topic in a survey of men who have sex with men (MSM) in Beijing, China. And MSM were selected by two-stage cluster sampling des
- PDF / 230,779 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 68 Downloads / 136 Views
Reliability and Validity Evaluation Based on Monte Carlo Simulations in Two-Stage Cluster Sampling on Sensitive Question Survey Zongda Jin, Hongru Zhu, Qiaoqiao Du, Xiangyu Chen and Ge Gao
Abstract In this paper, additive randomized response technique (RRT) model was presented in application to asking questions on sensitive topic in a survey of men who have sex with men (MSM) in Beijing, China. And MSM were selected by two-stage cluster sampling design. Monte Carlo simulation was successfully applied in the assessment of reliability and validity in additive RRT model for sensitive question investigation in two-stage cluster sampling.
Keywords Monte Carlo simulation Sensitive question Randomized response technique Additive model Two-stage cluster sampling
79.1 Introduction Sampling survey is one of the most common methods of scientific research in numerous fields, including psychology, demography, sociology, medicine, and health [1]. Sampling investigation related to sensitive information is inevitably encountered. Sensitive issues mean that the content of questions themselves can be considered as intrusive and the answers to questions are socially undesirable or unacceptable and the meanings of questions have threat of disclosure. Such as evading income tax, prostitution, HIV-positive, homosexuality, and usage of illegal drugs, that it is difficult to achieve valid information in the area of sensitive topics. If a direct survey research is employed to assess the characteristics of Z. Jin Q. Du X. Chen G. Gao (&) School of Public Health, Medical College of Soochow University, Suzhou 215123, China e-mail: [email protected] H. Zhu Jiangsu Institute of Parasitic Diseases, Wuxi 214064, China
Y. Yang and M. Ma (eds.), Proceedings of the 2nd International Conference on Green Communications and Networks 2012 (GCN 2012): Volume 3, Lecture Notes in Electrical Engineering 225, DOI: 10.1007/978-3-642-35470-0_79, Ó Springer-Verlag Berlin Heidelberg 2013
647
648
Z. Jin et al.
sensitive topic, respondents often decline to take part or reply untruthfully and are willing to give socially desirable responses rather than accurate ones because of privacy concerns. Potential threat to accuracy is both nonresponsive and reporting error [2]. A proposed solution is the RRT designed for the purpose of circumventing social desirability bias (SDB), guaranteeing confidentiality, improving respondent cooperation, and procuring reliable estimates [3]. Sensitive issues can be broadly categorized according to character as either quantitative sensitive questions or qualitative sensitive questions (dichotomy and multichotomy) [4]. The RRT was originally introduced by Warner in 1965 [5]. In Warner method, the respondents are given two logically opposite questions, which are sensitive one and nonsensitive one. A first alternative to Warner’s design is known as Simmons’ model or unrelated question model. In Simmons’ model, the respondents are provided with the opportunity of replying to one of two questions by replacing nons
Data Loading...
