Overt honesty measures have been shown to be multidimensional. The multidimensionality of those tests rests, in part, on the different kinds of item styles that make up the measures such as items referring to admissions of dishonesty, temptations to be dishonest, and rationalizations about dishonest behavior. The purpose of the study presented here was to determine the extent to which these item styles measure the same underlying construct by examining their intercorrelations and their differential ability to predict actual dishonest behaviors. The results suggest that the different item styles found in overt honesty measures may not be measuring the same construct.