Personality testing in personnel selection: Is "socially desirable" responding really undesirable?

There is unequivocal evidence that test-takers are able to deliberately alter their scores on non-cognitive measures if instructed to do so. Based on this result, many scholars of personality have recommended to refrain from using personality tests for personnel selection. In the present paper, potential arguments in favor of, this position are critically examined. It is concluded that (1) these arguments are logically erroneous; (2) empirical evidence showing a negative impact of "social desirability" on criterion-related validity of personality testing for personnel selection is virtually non-existent; (3) the consequences of "faking" or '"impression management" for personnel selection are much too complex to justify labelling them as "response bias", and (4) the validity of personality tests may even benefit from the motivational conditions of a selection situation. An alternative conception of self-presentation in assessment for personnel selection is presented. and the implications of distinct determinants of individual personality test scores (ability to identify social expectations, motivation, to adapt to these, expectations) for validity are discussed.