Social desirability dimensionality: one or two continua?

Abstract

BackgroundThe objective of this paper is to examine the structure of social desirability. This variable is important in self-report question-naires. Theoretical analyses indicate the existence of at least two dimensions of this construct, but there is no unambiguous em-pirical evidence for such a structure of social desirability. The present research aimed to determine the factor structure of the social desirability construct and the number of its dimensions.Participants and procedureThe inductive research strategy was applied and a new pool of questionnaire items was generated. This approach allows for detachment from the existing social desirability assessment instruments. This study focuses on the concepts of social desirability that examinees have, rather than on the content of existing items measuring social desirability. The verification of the social de-sirability structure was performed in three stages. At each stage, the number of items was gradually reduced. The successive phases of the study involved three different samples, respectively, consisting of 657, 415, and 1,709 examinees.ResultsInitially, at the first stage of the study, two dimensions were detected, but in the two subsequent stages, the second dimension represented acquiescence. The procedure of partialling the mean out of the correlation matrix was applied to eliminate acquies-cence. This dimension is related, as expected, to agreeableness and conscientiousness. It also appeared that the social desirability structure can be confounded by acquiescence in longer questionnaires.ConclusionsIt is concluded that every item reflects both social desirability and acquiescence, albeit to different extents. A one-dimensional construct of social desirability was finally obtained.