The detection of faking on the Millon Clinical Multiaxial Inventory (MCMI).

This study investigated the effects of a variety of faking strategies on the Weight Factor correction scores (designed to detect malingering) of the Millon Clinical Multiaxial Inventory, a relatively new personality questionnaire. Subjects (both psychiatric patients [N = 95] and general medical/surgical controls [N = 90]) were asked to take the MCMI according to one of the following instructional sets: traditional faking-good; traditional faking-bad; role faking-positive; role faking-negative; role faking-neutral; and honest. The results indicated that neutral social role faking resulted in no significant differences in weight factor correction from subjects in the honest condition and that directionally role faked profiles did not differ in correction from those in the traditional "best" and "worst" conditions. Finally, only the Weight Factor corrections in the protocols from the fake bad conditions (whether in traditional "worst" or negative social role conditions) differed significantly from those in the honest conditions for both subject groups. Implications for the practicing clinician are discussed.