How to improve the efficiency of randomised response designs

Abstract

This paper describes ways to maximise the efficiency of randomised response designs. When randomised response designs become more efficient their value as a tool to study sensitive topics will increase. An overview of the literature shows that when sensitive or incriminating topics are studied, the overall results of randomised response studies are more valid than the results of direct question designs. This positive effect is small, however, and randomised response designs are known to be less efficient than direct question designs, making it necessary to recruit larger samples. In this paper the efficiency of six randomised response methods (Warner's design, both forms of the unrelated question technique, the forced response technique, Moors's design and Mangat's improved model) will be compared relative to direct question designs. Using the right design parameters we can make randomised response studies up to twice as efficient. The forced response method and a special form of the unrelated question design are the most efficient designs until now. © Springer 2005.