Development and validation of the malingering discriminant function index for the MMPI-2

The predictive capacity of the MMPI-2 (Butcher et al., 2001) "fake-bad" validity scales (17, F-B, and F-P) is diminished when respondents have knowledge (i.e., coached) about the operating characteristics of these scales. In this investigation, we endeavored to develop a MMPI-2 fake-bad validity index that would be less vulnerable to validity-scale knowledge. Applying discriminant function procedures, we derived a set of weighted Clinical and Content scales that reliably distinguished large samples of validity-scale coached undergraduate research participants instructed to feign mental illness (n = 534) from psychiatric patient samples (n = 590). We subsequently validated this Malingering Discriminant Function Index (M-DFI) in independent samples of research participants (n = 230) and patients (n = 300) and showed relatively less attenuation in predictive capacity compared with F, F-B, and F-P across uncoached and validity-scale coached feigning conditions.