Stable inconsistency: a study of response inconsistency over time

Personality measures are currently a popular method for selection in the business world, despite issues such as poor predictive ability and the potential for output manipulation by participants. Another issue with personality testing that is often overlooked is that individuals sometimes respond differently on test items that otherwise measure the same traits. This phenomenon has been called Response Inconsistency (Reddock, Biderman, & Nguyen, 2011). The focus of this study is to attempt to show the phenomenon as a measurable trait that is stable over time. The study administered two different Big Five Inventories to participants taken from a local university. The two inventories were administered at different dates. Inconsistency measured as the mean standard deviation of responses within domains was computed and then correlated across the two time periods to determine reliability of the inconsistency measure. High correlations supported the hypothesis that response inconsistency is stable across time