Investigating Discrepancies between Predicted and Observed Wechsler Adult Intelligence Scale‐Version IV Full‐Scale Intelligence Quotient Scores in a Non‐Clinical Sample

Abstract

ObjectiveIn neuropsychological assessment estimates of how a client was performing prior to an injury is achieved by using instruments that have been developed to predict premorbid abilities, such as the National Adult Reading Test (NART), the re‐standardised NART (NART‐2), the Wechsler Test of Adult Reading (WTAR), and the Test of Premorbid Functioning (TOPF), and comparing these to scores from tests of current ability, such as the Wechsler Adult Intelligence Scale‐Fourth Edition (WAIS‐IV). These instruments have been criticised, however, due to concerns regarding the accuracy of their predictions.MethodThis study evaluated the ability of the NART, NART‐2, WTAR and TOPF to accurately estimate current full scale intelligence quotient (FSIQ) as measured by the WAIS‐IV in an unimpaired Australian sample.ResultsFor the sample as a whole, the NART significantly over‐estimated FSIQ, whilst the TOPF significantly underestimated FSIQ. In the low average IQ group, scores were over‐estimated by the NART‐2 and WTAR. In the average group, the predicted FSIQ scores produced by each measure were relatively accurate. In the high average group, NART was the best predictor of FSIQ, while the other tests all significantly underestimated IQ.ConclusionsIt was concluded that reading tests may not be the most accurate method for predicting IQ, and that other approaches which include demographic correction the should be further explored.