| The purpose of this study was to develop statistically stable stepwise multiple regression models that can be used by clinicians to predict the probability for completion of a consistent MMPI-2 profile based on an adult forensic outpatient litigant's WAIS-R FSIQ, WRAT-R reading level, and years of completed education before the MMPI-2 is administered.;Using correlational analysis for the complete forensic sample, significant negative correlations were found between the MMPI-2 criterion variables F, FB, TRIN, and F + FB + /F - FB/ and the WARS-R FSIQ, WRAT-R reading level, and years in education. Significant negative correlations were found between VRIN and VRIN + /F - FB/ and the WAIS-R FSIQ and WRAT-R reading level. Finally, a significant negative correlations was found between /F - F B/ and WAIS-R FSIQ.;Next, these forensic litigants were divided into odd and even samples based on odd and even identification numbers in the forensic data set. An independent sample West was used to compare mean T-Scores for the fast four MMPI-2 criterion variables and raw score for the last three MMPI-2 criterion variables. All seven comparisons between the odd and even forensic samples were non significant.;Then, stepwise multiple regression analyses for each MMPI-2 criterion variable F, FB, TRIN, /F - FB/, VRIN + /F - FB/, and F + FB + /F - FB/ regressed on the WAIS-R FSIQ, WRAT-R reading level, and years in education produced significant multiple correlation coefficients for all models except VRIN for forensic litigants in the odd sample. Finally, cross-validation was conducted for all significant MMPI-2 criterion variable models derived from the odd forensic sample and significant multiple correlation coefficients were found for all the models except /F - FB/.;Clinical applications are discussed that will allow clinicians to determine if adult forensic outpatient litigants can endorse MMPI-2 items consistently if their WAIS-R FSIQ scores and years of completed education are known before the test is administered. |
