Abstract
The randomized response (RR) technique pioneered by Warner, S.L. (1965)[Randomised response: a survey technique for eliminating evasive answer bias. J. Amer. Statist. Assoc.60, 63–69] is a useful tool in estimating the proportion of persons in a community bearing sensitive or socially disapproved characteristics. Mangat, N.S. & Singh, R. (1990)[An alternative rendomized response procedure. Biometrika77, 439–442] proposed a modification of Warner's procedure by using two RR techniques. Presented here is a generalized two-stage RR procedure and derivation of the condition under which the proposed procedure produces a more precise estimator of the population parameter. A comparative study on the performance of this two-stage procedure and conventional RR techniques, assuming that the respondents' jeopardy level in this proposed method remains the same as that offered by the traditional RR procedures, is also reported. In addition, a numerical example compares the efficiency of the proposed method with the traditional RR procedures.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
