The relationship between Raven's coloured progressive matrices and the Weschsler Intelligence Scale for Children for a group of mildly mentally retarded Mexican-American children

The purpose of this study was to investigate the degree of relationship between the Wechsler Intelligence Scale for Children (WISC) and the Raven's Coloured Progressive Matrices (CPM) (1956) for a group of mildly mentally retarded children of Mexican-American descent. The subjects vzere thirty-two male and female children of Mexican-American descent in two special education classes for the mildly mentally retarded in the Houston Independent School District, The subjects consisted of fifteen males and seventeen females ranging in chronological age from 120 months to 156 months with a mean of 139.97 months and a standard deviation of 10,78 months. Their WISC Full Scale IQs ranged from 51 to 85 with a mean of 68,28 and a standard deviation of 7.39. The subjects vzere screened, by the school nurse and physician, for any physical defects, such as sensory or neurological impairments. They were from homogeneous lower socio-economic backgrounds, and they were bilingual. The WISC and CPM were individually administered and scored by the examiner. The thirty-two subjects were randomly placed in two groups of sixteen; for one group the WISC was administered first and for the other group the CPM was administered first. Each test was given in different sessions with twenty-four hours between the first and second session for each subject. All WISC subtests, six Verbal and six Performance, were given and prorated. Mean male and female CPM raw scores were tested (null hypothesis) for any significant sex difference and no significant difference was found. A t-ratio of 1.77 was obtained. Spearman rank-order correlation coefficients were computed between raw scores for the CPM and WISC subtests. Partial correlation coefficients were computed between the CPM raw scores and the WISC IQs: Verbal, Performance, and Full Scale. Partial correlation coefficients were used to hold the chronological age factor constant. Spearman rank-order correlation coefficients between raw scores for the CPM and WISC subtests yielded correlations significant at the .01 level of confidence for the following: Information, .46; Comprehension, .48; Similarities, .45; Vocabulary, .68; Picture Arrangement, .44; and Block Design, .62; correlations were significant at the .05 level of confidence for the following: Arithmetic, .32; Digit Span, .42; Object Assembly, .38; and Mazes, .36; correlations with Picture Completion and Coding were -.02 and -.05, respectively. Except for Block Design .62, the CPM correlated more highly with WISC verbal and language subtests than with the WISC nonverbal and nonlanguage subtests. Partial correlation coefficients between CPM raw scores and WISC IQs Verbal, Performance, and Full Scale were .45, significant at the .01 level of confidence? .39, significant at the .05 level of confidence; .46, significant at the .01 level of confidence, respectively. Although the correlations in this study were moderate, further research with the CPM might prove heuristic when small groups of homogeneous subjects, explicitly defined and carefully selected, are used. Future research might go in the direction of the CPM as a predictor of language potential and indicator of visual perceptual disorders. Possible groups with language and conununication disorders, such as the deaf and aphasic, might be investigated.