Estimation of a Sensitive Proportion by Warner's Randomized

							                                                                                                             Page
                                      Contents                                                              1 of 1

       Print

                     Estimation of a Sensitive Proportion by Warner's Randomized
                              Response Data Through Inverse Sampling
               Mausumi Bose; Arijit Chaudhuri; Kajal Dihidar
               Indian Statistical Institute, 203 B.T. Road, Kolkata 700108, India.
               E-mail: mausumi.bose@gmail.com
     STCPMs




               Abstract: It is difficult to obtain trustworthy data on stigmatizing matters like habits of tax evasion, drunken
Sessions




               driving, etc., through direct queries. To overcome this difficulty, Warner introduced randomized response
               techniques to estimate the proportion of people bearing such a stigmatizing or sensitive characteristic in a
               given community. For simplicity in estimation he restricted to SRSWR and since then, several researchers
               have extended and applied this technique in various ways. All these results involve sampling with a pre-fixed
               number of draws. In this paper we employ inverse sampling with equal probabilities with replacement and
               show certain advantages in estimation using randomized response data by Warner's device gathered through
               such a simple inverse sampling scheme.

               Key words and phrases: Distinct units, Randomized response device, Stigmatizing characteristic.


           T