Personality testing for autistic individuals: Potential issues for selection

Employment selection is a significant contributor to poorer employment outcomes for autistic individuals. However, the impact of conventional employment selection tools on autistic applicants is poorly understood, including the preemployment personality test. Drawing upon the realistic accuracy model and the HEXACO model of personality, we investigated contributors to measurement bias (i.e., noninvariance) and adverse impact (i.e., mean differences) against autistic individuals in self-report personality measures, which preemployment personality tests typically use. Participants in a multinational sample (N = 49,912) completed one of several language versions of the HEXACO-100. We then created a matched subsample (N = 2,898) equally comprised of formally diagnosed autistic individuals, self-identified autistic individuals, and allistic (i.e., nonautistic) individuals matched across several individual differences. Our findings suggested that preemployment personality tests may disadvantage both formally diagnosed and self-identified autistic individuals, not due to measurement noninvariance, but because of substantial mean trait differences likely leading to adverse impact, particularly with the traits of extraversion and conscientiousness. We largely replicated these findings across subsamples by regions of residence and a supplemental sample from the same platform, which also linked extraversion and conscientiousness with employment status. While we are not calling for a stoppage in the use of these tests, there is a need to evaluate how they are used in selection contexts to ensure fairness for autistic individuals. (PsycInfo Database Record (c) 2026 APA, all rights reserved).