Psychometrist University of Illinois at Chicago Melrose Park, Illinois, United States
Objective: Many individual embedded performance validity tests (PVTs) have been independently cross-validated as validity indices among various clinical populations; however, research examining the additional predictive value from combining multiple embedded PVTs is more limited. The study's purpose was to examine the accuracy of eight embedded PVTs for predicting pass/failure performance on a well-validated freestanding PVT, the Advanced Clinical Solutions: Word Choice Test (WCT).
Method: The study consisted of 225 adults referred for outpatient neuropsychological evaluation that included the following embedded PVTs: Rey Auditory Verbal Learning Test (RAVLT) Forced Choice (FC), RAVLT Effort Score, Brief Visual Memory Test-Revised Recognition Discrimination (BVMT-RD), Digit Span Age Corrected Scaled Score, Reliable Digit Span, Stroop Color and Word Test Word Reading T-Score, Letter Fluency T-Score, and Trail Making Test Part A T-Score. The sample was 56% female and racially diverse (40% Black, 35% White, 17% Hispanic, 5% Asian, and 2% other ethnicities) with a mean age of 45.96 (SD=16.12) and mean education of 13.96 (SD=2.73) years.
Results: A forward stepwise logistic regression was performed to predict WCT pass/fail using the embedded PVTs as predictors. The model was achieved in three steps (p <.005); Step 1: BVMT-RD (Classification Accuracy=90.1%; R2 =.45), Step 2: BVMT-RD + RAVLT-FC (Classification Accuracy=90.6%; R2 =.51). Step 3: BVMT-RD + RAVLT-FC + TMT-A (Classification Accuracy=92.1%; R2 =.55). All steps had >95% accuracy in correctly classifying WCT pass group.
Conclusion: RAVLT-FC, BVMT-RD, and TMT-A combined reliably predict WCT pass/fail group. Verbal and attention based embedded PVTs were found to have the most robust predictive capacity of WCT validity status.
