| Abstrakt: |
A recent AACN consensus statement (Guilmette et al. in The Clinical Neuropsychologist, 34(3), 437–453, 2020) endorsed the indeterminate/Borderline range as the third alternative for characterizing the results of performance validity assessment. This study was designed to empirically evaluate the clinical/forensic utility of a three-way classification system (Pass/Borderline/Fail) for the outcome of multivariate models of performance validity tests (MV-PVTs). Retrospective archival data from 100 patients with traumatic brain injury were collected from an academic medical center. Five MV-PVTs were developed, consisting of 5 through 13 embedded PVTs, using two different aggregation methods and two established free-standing PVTs as criterion measures. The indeterminate/Borderline range emerged as a third outcome of PVTs that is psychometrically distinct from both clearly valid and clearly invalid profiles. The interpretation of the indeterminate/Borderline range changes as a function of the number of PVTs interpreted, although the distance between Pass and Fail is well-preserved. The two aggregation methods produced converging results. Still, a strategic method variance in combining multiple PVTs into a single-number summary of performance validity can help disambiguate scores in the indeterminate/Borderline range. Results support the AACN's consensus statement that the indeterminate/Borderline range is a legitimate third outcome of PVTs. Formally endorsing its use can help improve overall classification accuracy in clinical/forensic settings, the agreement between independent experts, and operationalizing known error rate in the context of Daubert standards. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink