Deafness and WISC-III item difficulty: Invariance and fit

Journal/Book: J Sch Psychol. 1997; 35: The Boulevard, Langford Lane, Kidlington, Oxford, England OX5 1GB. Pergamon-Elsevier Science Ltd. 299-314.

Abstract: The following WISC-III subtests were hypothesized to contain differentially difficult items for deaf and hearing children: Picture Completion, Information, Similarities, Vocabulary, Comprehension. The WISC-III was translated into sign language and administered to 110 severely and profoundly deaf children(1). Rasch item difficulty estimates obtained for the sample of deaf children were compared to: (a) a sample of 110 nonreferred hearing children who were similar to the deaf children in terms of mean age and Performance IQ and (b) the WISC-III standardization sample (N = 2,200). Differential Item Functioning (DIF) was detected using the standardized difference in difficulty and anchor item methods. Numerous items consistently exhibited: (a) DIF both against and in favor of the deaf children, or (b) poor anchor item fit statistics. In addition, items did not retain their order of difficulty. Thus, the items appeared to function differently for deaf children.

Note: Article Maller SJ, Univ S Florida, Dept Educ Measurement & Res, 4202 E Fowler Ave, Fao 100U, Tampa,FL 33620 USA

Keyword(s): HEARING

Zurück | Weiter