ABSTRACT
Reduced response rate and inflated response bias are the two evils usually linked with direct surveys relating to sensitive items of inquiry. Gambling habits, addiction to drugs and other intoxicants, alcoholism, proneness to tax evasion, drunken driving, history of past involvement in crimes, and homosexuality are some of the characteristics about which people usually do not wish to divulge information. An investigator of such issues of social importance is reluctant to ask the sampled subjects direct questions with respect to such sensitive, stigmatizing, and perhaps incriminating matters. Two disadvantages associated with RR techniques are mentioned by Emrich. First, it is often difficult to implement them, as they require special skills on the part of investigators and more time and trouble in explaining to the subject how RR works. Second, the respondents may not be inclined to give truthful replies when they see that an RR scheme has in-built provisions to accommodate contaminated responses that permit the respondents to lie.
