Skip to main content
Thesis

Cross-validation in a non-litigating clinical population of Mittenberg's WAIS-III regression formula for identifying malingered head trauma

3 October 2005

Abstract

Analyzing patterns of neuropsychological test performance is one specific technique used by clinicians to detect probable malingering and implausible test performance. The use of empirically derived "cut-off' scores and discriminant validity formulas has also been applied as a strategy for analyzing patterns on commonly administered tests included in neuropsychological test batteries. Mittenberg et aI. (2001) conducted a study which examined two approaches to analyzing Wechsler Adult Intelligence Scale-III (W AIS-III) subtest scores for identifying malingering: (1) a discriminant function analysis (DFS) and, (2) based on a theoretical approach, Vocabulary and Digit Span discrepancy (VDS). The current study examined the specificity and generalizability of Mittenberg et al.'s approaches to identifying malingered cognitive performance using the W AIS-III in a non-litigating clinical population for whom the likelihood of malingering was very low. Test data were obtained through chart reviews of 186 adults (90 females, 96 males) who underwent cognitive assessment at a university psychology clinic. The application of the DFS and VDS methods in this study yielded false positive rates as high as 30%, suggesting that these methods likely are not useful in this type of population. Specific characteristics of each group were examined in relation to the strategies used to classify them in an effort to clarify how both methods identified approximately one third of a non-litigating sample as probable malingerers.


Files

Files are restricted to Pacific University. Sign in to view.