Putting applicant faking effects on personality tests into context

Abstract

Purpose Estimates of the effects of faking on personality scores typically represent the difference from one sample mean to another sample mean in terms of standard deviations. While this is technically accurate, it does put faking effects into the context of the individuals actually engaging in faking behavior. The purpose of this paper is to address this deficiency. Design/methodology/approach This paper provides a mathematical proof and a computational simulation manipulating faking effect size, prevalence of faking, and the size of the applicant pool. Findings The paper illustrates that reported effects of faking are underestimates of the amount of faking that individual test takers are engaging in. Results provide researchers and practitioners with more accurate estimates of how to interpret faking effects sizes. Practical implications To understand the impact of faking on personality testing, it is important to consider both faking effect sizes as well as the prevalence of faking. Originality/value Researchers and practitioners do not often consider the real implications of faking effect sizes. The current paper presents those results in a new light.