||The ability of the Personality Assessment Inventory to detect positive impression management in undergraduate students under two different assessment contexts was investigated. The possibility that the two groups would present different socially desirable responses was based upon research that suggests individuals self presentations may be either moralistically biased or egoistically biased. The two biases result in presentation styles that focus on either a "saint like" approach or a "super hero" like quality. The current study tested the hypothesis that attempts to fake a good impression on the Personality Assessment Inventory (PAI; Morey, 1991) would differ for respondents operating in a custody seeking context versus a job seeking context. One hundred undergraduate students were randomly assigned to one of two faking scenarios: job seeking, which was expected to elicit an egoistic bias, and custody seeking, which was expected to elicit a moralistic bias. The two biases were also assessed and compared between groups using the Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1988). The results indicated that the two groups did not significantly differ in their PAI profiles or BIDR scores. These null findings also indicate that the performance of the PAI validity scales is not dependent on assessment context.