Personal and Situational Determinants of Personality Responses: A Partial Reanalysis and Reinterpretation of the Schmit et al. (1995) Data

Using generalizability theory, a partial reanalysis and reinterpretation of the Schmit et al. (1995) data was conducted with undergraduate students who were asked to respond to both noncontextualized and work-specific versions of the NEO Five-Factor Inventory (Costa & McCrae, 1989) either honestly (N = 100) or as if they were applying for a job (N = 100). Results indicated that the person variance was larger in the applicant compared to the general instructions condition, that the person × situation interaction accounted for significant variance across all scales included in the study (Neuroticism, Extraversion, Agreeableness, and Conscientiousness), that these person × situation effects were attenuated in the applicant instructions condition as compared to the general instructions condition, and that responses to the Neuroticism scale were most affected by this interaction. Implications for the use of personality constructs in personnel selection include a shift from concerns of faking to a concern for proper contextualization of personality measures with the goal of maximizing prediction of work-related behavior.