The social desirability bias is a major response set that is possibly active when data are collected in empirical social studies with interviews, psychometric tests, or questionnaires in particular. This tendency interferes with the “true values” of the subjects’ traits or states that are to be assessed and puts a systematic bias onto the measured values. People react in a socially desirable way if they do not answer in a way or choose the alternative that best reflects the intensity or strength of the state or trait that is to be measured. Instead, subjects answer in a way that they assume to be expected by their peer group or the interviewer/ experimenter or approved by the general public. For example, a person who sympathizes with a radical (right-wing) political party and is asked in a questionnaire to state his or her beliefs concerning migration or foreign persons will probably give moderate answers despite holding a more radical view. By that, the person intends not to stand out from other subjects or to the experimenter as being deviant or radical.
A stronger bias toward social desirability is to be expected if the questions are closely related to social norms and value systems and if the subject is aware of that relationship. The amount to which a data collection method is subject to a social desirability bias can be estimated by assessing data for one group of subjects under normal conditions, and comparing with data from another group, where the subjects are additionally instructed to present themselves as much as possible in a socially desirable way.
In early social research the assumption prevailed that the tendency to appear in a good light to others is a general one and applies to all humans in the same way. It has since been found, however, that the social desirability tendency is influenced by factors like the age, gender, and social stratum of the subjects participating in the studies. Furthermore, there are correlations between the social desirability tendency and characteristics of the test situation as well as the topic of the questionnaire. In addition, regional and time-dependent influences on the social desirability tendency can also be observed.
Thorough item construction as well as carefully planned instructions can reduce the effects of social desirability. Questions and answering alternatives should be formulated in a way that reveals as little as possible about test intentions and keeps the personal state or trait to be measured hidden from the subject. Furthermore, groups of alternative answers can be composed that all have a similar low or high social desirability. In that way, subjects are constrained to choose an alternative independently of social desirability (i.e., forced choice). In order to assess the strength of socially desirable responding, scales have been developed that can help to either control the social desirability tendency or even correct test scores (e.g., the Marlowe–Crowne Social Desirability Scale, MCSDS). To do so, a number of items are put together where an honest answer and an answer oriented toward social desirability diverge. For example, if a subject is negating questions like Social Exchange
“Have you ever been dishonest?” more frequently than other persons, this indicates a reaction tendency toward social desirability. The subjects’ answers can indicate whether they have filled out the questionnaire in an honest and true manner or whether they have been dishonest, which could account for biased test scores.
Finally, a tendency toward social desirability can also be reduced by providing appropriate instructions. Subjects should explicitly be asked to give honest answers and not to answer in a way that is biased with respect to social norms. During the test procedure a relaxed atmosphere is of advantage, where subjects do not feel themselves provoked to present themselves in a socially desirable way and to leave a good impression. Subjects should be convinced that it is in their interest to present themselves as accurately as possible and not in terms of what they assume to be considered preferable.
A laborious but effective method for controlling social desirability is the bogus pipeline procedure. Subjects are confronted with a set-up for the measurement of physiological indicators (such as skin conductance or heart rate). They are informed that this procedure allows a person’s attitudes to be recorded with great reliability and validity. Filling out the questionnaire only serves to give an idea of how precisely a person knows his or her own attitudes and is able to communicate them. This method has been demonstrated to be relatively resistant against faking, social desirability, and other response sets.
- Bortz, J. (1984). Lehrbuch der empirischen Forschung: Für Sozialwissenschaftler [Textbook of empirical research: For social scientists]. Berlin: Springer.
- Crowne, D. P., & Marlowe, D. (1991). Marlowe Crowne social desirability scale. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (eds.), Measures of personality and social psychological attitudes. San Diego, CA: Academic Press, pp. 27–31.
- Crowne, D. P., & Marlowe, D. (1994). Social desirability bias. In G. C. Bruner & P. J. Hensel (eds.), Marketing scales handbook: A compilation of multi-item measures, vol. 1. Chicago, IL: American Marketing Association, pp. 568–571.
- Ones, D. S., Viswesvaran, C., & Reiss, A. D. (1996). Role of social desirability in personality testing for personnel selection: The red herring. Journal of Applied Psychology, 81(6), 660–670.