Fred H. Wallbrown

A factor analysis of reading attitudes along with measures of reading achievement and scholastic aptitude

The construct validity of the Survey of Reading Attitudes was investigated through a principal-factor analysis of the 88 items comprising the scale, along with measures of scholastic aptitude and reading skills (vocabulary and comprehension). The sample included 600 students selected so that 200 (100 boys and 100 girls) were obtained from each of the three intermediate grades. The Survey was administered orally to class-size groups, but copies of the booklet were also available to students. The factor structure showed a reading achievement factor along with the following eight dimensions of reading attitude: (1) expressed reading difficulty, (2) reading as direct reinforcement, (3) reading as enjoyment, (4) alternative learning modes, (5) silent vs. oral reading, (6) reading anxiety, (7) reading group, and (8) comics.

The Structure of Intelligence for Black Children: A Hierarchical Approach

A hierarchical factor solution (Wherry & Wherry, Note 1) was obtained on WISC-R subtest intercorrelations for 150 black children and adolescents referred to a community agency for psychological assessment. Factorization was controlled by specifying the extraction of two primary factors and allowing for one higher-order factor. The factors thus obtained not only provided a parsimonious account of the common variance in the WISC-R subtests, but also indicated an ability arrangement congruent with Vernon’s (1950) structural paradigm. The ability hierarchy included a relatively weak general g factor defined by a pattern of positive loadings from all WISC-R subtests. At the primary level, there was almost complete bifurcation between the verbal and performance subtests, with the former defining a verbal-educational (v:ed) factor and the latter defining a spatial-perceptual (k:m) factor. These findings suggest that the basic ability dimensions are the same for referred black children as they are for referred whites and normals in the standardization sample.

Comparison of the WISC-R and PPVT for a group of mentally retarded students

This study compared the WISC-R and PPVT scores for a group of 65 students classified as mentally retarded, ranging in age from 7.5 to 14.5. Pearson product moments (r) and correlated t values were employed in the analysis of data. The evidence from the present study indicated that for this sample of mildly retarded children and youth, the PPVT IQ score is significantly higher than the Full Scale IQ from the WISC-R. The current findings indicate that the same general pattern exists for the WISC-R and PPVT as for the WISC and PPVT scores.

HIERARCHICAL FACTOR STRUCTURE OF THE WISC-R FOR REFERRED CHILDREN AND ADOLESCENTS

A hierarchical factor solution was obtained for intercorrelations among the 10 regular WISC-R subtests for 169 children and adolescents referred to a community agency for assessment. This sample was predominantly rural but quite heterogeneous with respect to age, ability, and type of referral problem. The findings indicated an ability hierarchy composed of a strong general (g) factor and two sub-general factors corresponding to the spatial-perceptual (k:m) and verbal-educational (v:ed) parameters of Vernons (1950) structural paradigm.

Analysis of Cognitive Abilities for Mentally Retarded Children on the WISC-R

This study investigated the cognitive abilities of 238 retarded children and youth on the WISC-R. The sample consisted of 238 children (135 boys and 103 girls) ranging in ages from 5 yr., 7 mo. to 16 yr., 7 mo., with a mean age of 10 yr., 9 mo. The variation in subtest scores was analyzed by running an analysis of variance for correlated data for the sample’s subtest scores. Individual differences between subtest means were analyzed by Newman-Keuls test for simple effects. Students’ t and Fisher’s F ratios were used to analyze the subtest scale scores of males and females. On the WISC-R Verbal Scale, the boys scored slightly higher than the girls on each of the subtests. The girls excelled on the Coding. Greatest variation in subtest scores occurred in the Object Assembly, while the lowest variation occurred in Picture Arrangement. Evidence from the present study of mentally retarded children’s performance on the WISC-R indicated that the relative strength and/or weakness of this sample is not restricted to either the Verbal or Performance area. Differences among the WISC-R subtests for retarded subjects seem to be as great within the Verbal-Performance areas as they are between them.

The Stability of Four Kinds of Perceptual Errors on the Bender-Gestalt

The test-retest reliability of the Bender-Gestalt Test was investigated for 157 second graders from two middle-class schools. The test-retest interval ranged from 11 days through 15 days. Testing was conducted in accordance with the Koppitz procedure, but errors of distortion, rotation, integration, and preservation were determined separately. The total time required to reproduce the designs were also recorded. The reliability estimate for the Koppitz score was generally satisfactory, but the estimates obtained for the four separate error categories and total working time were substantially smaller.

The validity of two clinical tests of visual-motor perception.

The study investigated the relative efficiency of the Bender and MPD as assessors of achievement-related errors in visual-motor perception. Clinical experience with these two tests suggests that beyond first grade the MPD is more sensitive than the Bender for purposes of measuring deficits in visual-motor perception that interfere with effective classroom learning. The sample was composed of 153 third-grade children from two upper-middle-class elementary schools in a surburban school system in central Ohio. For three of the four achievement criteria, the results were clearly congruent with the hypothesis stated above. That is, SpCD errors from the MPD not only showed significantly higher negative rs with the criteria (reading vocabulary, reading comprehension, and mathematics computation) than Koppitz errors from the Bender, but also accounted for a much higher proportion of the variance in these criteria. Thus, the findings suggest that psychologists engaged in the assessment of older children seriously should consider adding the MPD to their assessment battery.

DIMENSIONS OF CLASSROOM BEHAVIOR FOR KINDERGARTEN CHILDREN

The construct validity of the Devereux Elementary School Behavior Rating Scale (Spivack & Swift, 1967) was investigated for a sample of 408 children enrolled in the 15 kindergarten classes of a suburban school system. The 9 teachers completed behavioral ratings for the children in their classes during 1 wk. in May, 1974. A principal-factor solution was obtained on intercorrelations among the 47 behaviors included in the Devereux scale and the factors thus obtained were rotated to Varimax criterion. The results were generally positive in that 9 of the 11 behavioral dimensions described by Spivack and Swift (1967) were evident in the factor structure. Yet, there were enough differences to suggest the possibility of modifying the Devereux score categories somewhat for use with suburban kindergarten children. For example, the three individual items did not remain distinct and two dimensions did not emerge as factors. Also, the items loading the 9 factors were not always exactly the same as those which the authors specified for the corresponding behavioral dimensions.

Correlations between WISC-R subtests and verbal, performance, and full scale IQ scores for minority group children

The relationship between WISC-R subtest scores and Verbal, Performance, and Full Scale IQs was investigated for a sample comprised of rural Appalachian children. The sample was comprised of both white and black children who might be described as culturally different by virtue of low family income and residence in the mountainous areas of Virginia and North Carolina. Data analysis consisted of computing product-moment correlations (rs) between each of the ten subtests and the Verbal, Performance, and Full Scale IQs. Correlations between individual subtests and the FS IQ ranged from.49 through.63, but none of the differences between these rs were large enough to attain significance. Correlations between the verbal subtests and Verbal IQ ranged from.64 for Comprehension up to.78 for Vocabulary. For the performance subtest, the rs ranged from.54 between Coding and the Performance IQ up to.72 for Object Assembly.