The Importance of Relying on the Manual-Scoring Error Variance in the WISC-IV Vocabulary Subtest

摘要

Classical test theory assumes that ability level has no effect on measurement error. Newer test theories, however, argue that the precision of a measurement instrument changes as a function of the examinee's true score. Research has shown that administration errors are common in the Wechsler scales and that subtests requiring subjective scoring are the most prone to errors. In this study, graduate students scored fictitious Wechsler Intelligence Scale for Children-Fourth Edition Vocabulary subtests that were constructed to yield actual scaled scores of 4, 10 and 16. Results support the conditional error variance hypothesis: Participants made significantly more scoring errors in the low and high ability conditions than they did in the average ability condition. Implications of these findings with regard to the interpretation of extreme scores are discussed.