The effects of faking on the relationship between cognitive ability and conscientiousness: A cautionary note

It is well-established that selection decisions can be improved using multiple non-redundant assessments. Two such assessments are cognitive ability and conscientiousness. Though meta-analytic findings demonstrate little or no relationship between cognitive ability and conscientiousness, faking research suggests the two variables are related when test-takers are motivated to fake. We extend this logic by showing that incremental validity for conscientiousness declines when respondents fake, due to enhanced collinearity between conscientiousness and cognitive ability. Three studies, employing within-subjects designs and utilizing three different faking conditions, reveal a consistent increase in collinearity between conscientiousness and cognitive ability when respondents are motivated to fake, leading to reduced incremental validity of conscientiousness beyond cognitive ability in predicting the performance. Implications for selection systems are discussed.