Social Desirability Bias: Sex, Drugs & Rocking Surveys
Posted by Vovici Blog on Wed, Jan 13, 2010
Social desirability bias is the tendency of respondents to provide the politically correct answer rather than the candid answer. Respondents often give the answer that they think will make them seem as mainstream as possible in their views, from an innate desire to be seen favorably by others.
Among the many topics that prompt respondents to alter their answers:
- Health initiatives - Respondents exaggerate their frequency of exercise and their compliance with medical regimens.
- Voting behavior - Respondents exaggerate their intent to vote.
- Illegal behavior - Respondents underreport drug usage and criminal history.
- Sexual behavior - Respondents deny, sanitize or mainstream aspects of their sexual lives.
- Bigotry - Respondents downplay any prejudices.
- Salary - Poor respondents overstate income; rich respondents understate it.
Telephone surveys with human interviewers (CATI) seem to show the highest rate of social desirability bias, followed by face-to-face surveys (see "The Survey Response Process in Telephone and Face-to-Face Surveys: Differences in Respondent Satisficing and Social Desirability Response Bias"). The best mode of research is self-administered questionnaires, whether done on paper, on telephone through IVR (Interactive Voice Response) or on computer. In fact, federal medical researchers conducting face-to-face interviews use laptops to administer those sections of the survey most prone to social desirability bias (e.g., sexual practices, illicit drug use). In one comparison of a CATI survey, IVR survey and web survey, respondents to the web survey were the most likely to report socially undesirable information about themselves and were the least likely to falsely deny behavior (see "Social Desirability Bias In CATI, IVR, and Web Surveys: The Effects of Mode and Question Sensitivity").
These results closely match respondent perceptions. The Mars Food/Colmar Brunton research that Ray Poynter of The Future Place presented at the ESOMAR Online Research 2009 conference asked research participants about how frequently they could be honest in different modes of research: 71% said they could always be honest in online surveys, 59% said they could in focus groups, 57% in MROCs, 54% in face-to-face interviews and 49% in telephone surveys.
When you are concerned that the sensitive nature of your questions might distort the answers you receive, online surveys are your best bet to minimize social desirability bias.