engleski

Usefulness of item response latencies in separating self-enhancement from impression management

An important problem in personnel selection is susceptibility of personality questionnaires to impression management (i.e., faking). Faking introduces unwanted variance in personality scores that may cause incorrect and unfair selection decision. The main trouble with the faking problem is that it is still impossible to identify reliably the applicants who faked their responses. Social desirability scales that are most often used for this purpose have been shown to measure a mixture of unconscious self-enhancement and conscious impression management (Paulhus, 1984, 2002). In the time of computerized testing, measurement of item response latencies represents one of the most promising approaches to the detection of faked responses. However, before the response latencies are used for this purpose it should be checked whether they differentiate honest responding, self-enhancement and impression management on personality items. We collected the data on a sample of students (N = 229) who completed the electronic version of a five-factor personality questionnaire (IPIP-100) under honest responding and in two simulated selection situations. In addition to the participants’ self-reports on the questionnaire items, we registered the item response latencies and collected peer ratings on the same items. Based on the comparison between the self-ratings and the other-ratings we were able to identify honest responses, and unconsciously self-enhanced responses. Consciously faked responses were defined based on the comparison between self- reports under honest responding and those obtained in the simulated selections. Using several analytical approaches, we compared the latency times for the honest, the self- enhanced and the faked responses. The latency times for the self-enhanced responses did not differ from the latency times for the honest responses. This was in accordance with our expectations that self-enhancement represents an automatic, unconscious process. At the same time, the latency times for the faked responses were significantly longer than the latency times for the honest responses and the responses that indicated self-enhancement. These results supported our expectations that faking represents a deliberate, conscious form of socially desirable responding, and that item response latencies can be used to identify responses that underwent impression management.

socially desirable responding; self-enhancement; impression management; response latency

nije evidentirano