Social desirability bias
From Wikipedia, the free encyclopedia
Social desirability bias is the inclination to present oneself in a manner that will be viewed favorably by others. Being by nature social creatures, all people are inclined to seek some degree of social acceptance, and as with other psychological terms, "social desirability" is by no means an epithet.
Social desirability in its extreme however, can cause difficulties in research, particularly in psychological and medical research. When participants (subjects) in research provide "socially desirable" answers, results can often be confounded.
One example of how social desirability can affect research is in the hypothetical example of the researcher who wants to study sexual behaviour. The researcher may administer a questionnaire that asks the subject, "Do you currently masturbate? If so, how often?" Although it could be argued that no one would have trouble answering such a question for the benefit of science, it has been repeatedly proven otherwise.
A partial, by no means comprehensive, list of some areas that are sensitive to participants' interpretations of social desirability follows.
- Sexual behaviour and fantasies
- Personal income and earnings
- Feelings of self-worth and/or powerlessness
- Excretory functions
- Compliance with medicinal dosing schedules
- Religion
- Patriotism
- Intellectual achievements
- Physical symptomatology
- Acts of real or imagined physical violence
- Indicators of "kindness" or "benevolence"
- Illegal acts
When social desirability cannot be guarded against in research, often the researcher will resort to a scale that measures socially desirable responding, with the assumption that if a participant answers in a socially desirable manner on that scale, they are in all likelihood answering similarly throughout the study. One example of a test that measures socially-desirable responding is the Marlowe-Crowne Scale. Depending upon the goals of the research, respondents that engage in significant amounts of socially-desirable responding are discarded from statistical consideration; midrange scorers on a scale of socially-desirable responding may or may not be included in statistical consideration at the researcher's discretion, or their answers may be recalibrated commensurate with their perceived degree of skew, depending upon the measures involved, the goals of the study, and the robustness of the measures used.
