Using a randomized response research design to eliminate non-response and response biases in business research

A serious distortion of business survey research is that frequently the survey includes questions that the respondent views as sensitive or intimate. The respondent may refuse to answer or may give an inaccurate answer. This refusal to answer or tendency to falsify the answer can seriously distort the research findings and invalidate the interferences made from the survey. This paper discusses a sampling procedure that has recently been developed and used to reduce dishonest responses or nonreponses to sensitive questions. The procedure is called Randomized Response. As an example, it is used to ask sensitive questions of individuals who are marketing majors at the University of Hawaii concerning their behavior and attitudes toward racial discrimination, shoplifting, test cheating, and commodity hoarding.