Employee Recruitment: Identifying Response Distortion on the Personality Measure

Abstract

The aim of this study was to identify individuals who fake their response on personality assessments in the context of employee recruitment. In the study experiment, participants were randomly divided into two groups. The first group was instructed to complete a measure (BFI-44) honestly while the second group was instructed to complete the inventory as if they were job applicants participating in an employee recruitment. It was hoped that the second group would be induced to give fake responses. Cluster analysis and latent class analysis for a two-class constrained model was applied to and fitted to the data. The correlation between actual group (honest vs. faking) and predicted group obtained from analysis result (normal vs. unique) was moderate: values for the phi correlation using a 2x2 crosstab table yielded values from 0.02 to 0.37. The percentage of participants detected as giving fake responses ranged from 37 to 69 %.