CAROLINA PICTURE VOCABULARY TEST: VALIDATION WITH HEARING-IMPAIRED STUDENTS '

Summary.-When mean Carolina Picture Vocabulary Test standard scores were compared with WISC-R Performance Scale IQs significant differences were observed. The Carolina test scores were also correlated with both Performance IQs and the Performance Scale subtest scores. Most correlations were minimal as the two tests were independent; only WISC-R Picture Arrangement scores correlated significantly with Carolina scores. These outcomes coupled with the technical limitations of rhe Carolina raise serious questions regarding its utility lor hearing-impaired children. The Carolina Picture Vocabulary Test (Layton & Holmes, 1985) was developed to measure the receptive sign vocabulary of hearing-impaired children whose primary mode of communication is manual signing. Adrninistration was designed specifically for the hearing-impaired, since adaptation of other receptive language measures has not been successful for this population. The test was normed on a national standardization sample of 767 children between the ages of 2% and 16 yr. Typical characteristics of the sample included prelanguage deafness, 80 dB + hearing threshold in the better ear, an IQ of 80 to 100 points, parents with normal hearing, and manual signing as the primary means of communication (Layton 8: Holmes, 1985). The vocabulary items from which the Carolina test was developed evolved from vocabulary lists for deaf children (Silverman-Dresner & Guilfoyle, 1972) and lists of signed words in Signing Exact English (Gustason, Pfetzing, & Zawolkow, 1972). The efforts of the test authors were laudatory as there have been few recent attempts to measure the receptive vocabulary of deaf children. In the studies that have been reported, Hedger (1977) indicated that performance on three different oral receptive vocabulary measures by orally trained deaf children was better when written words were included. Forde (1977) tested hearing-impaired students with Forms A and B of the Peabody Picture Vocabulary Test (PPVT) in an effort to develop norms for this group. This study compared Carolina test scores with both WISC-R Performance Scale IQs and Performance Scale subtest scores. Since there is