Abstract
We use a within-person research design to compare response distortion on personality test scores obtained in a simulated selection context with scores obtained in a condition without motivation to distort. Even after accounting for measurement error, rank order changes result in a significant number of individuals being selected based on their selection context scores who would not be selected based on their nonmotivated scores. Yet, many individuals who distort do not elevate their scores enough to move to the top of the distribution, which alleviates some concern about combining personality measures with top-down selection procedures that adopt a low selection ratio. Common correction methods failed to accurately detect individuals who distort, and the correction practice of partialling lie scale scores from personality traits inaccurately eliminated individuals who did not truly distort.