Asking sensitive questions: The effects of type of referent and frequency wording in counterbiasing methods

Respondents typically underreport socially undesirable behaviors and overreport socially desirable behaviors because of a social desirability bias. This not only leads to biased data, but also has important implications for health-care marketers interested in encouraging people to perform in more socially desirable ways. This article investigates the effects of providing counterbiasing information on reducing this social desirability bias. Study 1 investigates whether information that a socially desirable behavior is performed less frequently than expected reduces the extent of overreporting. In Study 2, alternative methods of presenting information about the target behavior are studied. The data demonstrate that counterbiasing information is more effective when the referent is a population base rate versus an individual, and is moderated by the wording of the frequency, such that the counterbiasing information is more effective when the frequency is worded as an actual number versus a percent. Implications for questionnaire design and health-care marketing are discussed. (C) 1996 John Wiley & Sons, Inc