Circumventing Social Desirability Response Bias using Randomized Response Techniques
by
Sat Gupta
Department of Math/Stats & Psychology University of Southern Maine
Coauthors: Bill Thornton

Randomized response techniques are important tools in circumventing response bias in personal interview surveys involving sensitive questions. In this paper we discuss the use of several of these techniques in a campus survey involving students enrolled in psychology and statistics courses. The purpose of the survey was to examine a variety of social issues such as smoking, drug use, alcohol use etc. among students. The results are compared with bench-mark results based on an anonymous/confidential survey, and also with results based on the Bogus Pipeline (BPL) technique which has been used by psychologists to elicit more accurate responses to personal questions of sensitive nature. BPL technique requires attaching electrodes to the fingers of the respondent's non- preferred hand and wiring to an automated data acquisition computer system. Through a series of test questions the respondent is made to believe that the system can detect the truthfulness of a person’s response. Preliminary results indicate that some of the randomized response techniques are at least as effective as the BPL technique while being less intrusive.

Date received: September 26, 2002

Copyright © 2002 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cais-27.