Abstract: The studies on workplace deviance behavior are criticized for over-reliance on self-report methodology. Two samples of self-reports and projective self-reports of workplace deviant behavior are tested using SEM multiple-group analysis. The result demonstrates that projective self-reports is able to deal with the under-reporting of self-repots to a great extent, and self-reports and projective self-reports are both useful for assessing deviant behavior in organizations. Moreover, it is proved that the selection and recruitment with reward strategy are able to directly affect workplace deviant behavior.

Key words: workplace deviant behavior, self-report, SEM multiple-group analysis