Effects of applicant response distortion on the relationship between personality trait scores and cognitive ability

Abstract

This study examined differences in the psychometric scale properties of scores from a personality inventory as a function of cognitive ability when administered in low-stakes (incumbent employees; N = 1108) vs. high-stakes (job applicants; N = 79,339) testing contexts. Based on the idea that applicant response distortion increases trait score means and covariance and that cognitive ability is related to successful distortion, mean scores and loadings on a bifactor model were compared across groups that differed in motivation and ability to distort responses on the personality inventory. Results indicated that mean trait scores and loadings on a common factor were higher for the applicants than the incumbents. In addition, there was a positive linear trend relating cognitive ability to increases in scale scores and common factor loadings in the applicant sample; the observed changes in structure are a noted departure from what has often been found with regard to the differentiation of personality by intelligence hypothesis. Implications are discussed in terms of use of personality inventories in personnel selection contexts.