Journal of Psychoeducational Assessment

This study presents results of confirmatory factor analyses of the recently published Fourth Edition of the Stanford-Binet Intelligence Scale. Four possible models of the scale's factor structure were tested with confirmatory factor analysis via LISREL at five age levels and for the standardiza tion sample as a whole. Results of these analyses show that although the four-factor model presented by the scale's authors fits the general standardiza tion data fairly well, it did not fit similarly well with data from several age groups. Overall, however, models based on the authors' theoretical structure were usually significantly better fitting to empirical data than were models with fewer fac tors. Results are discussed in terms of their im plications for interpretation and use of the Stanford-Binet, Fourth Edition with young children.

Published in 1985, the Detroit Tests of Learning Aptitude-2, (DTLA-2) is used widely in schools with students who are experiencing learning prob lems. This study investigated the relationship of the DTLA-2 to the WISC-R with a sample of referred students. Correlation coefficients between the three subscales of the WISC-R and the overall score of the DTLA-2 were high; however, a sig nificant difference between means for those scores also was found. Significant differences between DTLA-2 composite scores for attention and motor skills also were found with this group, a finding that the authors interpret as accurately describ ing the sample. Practitioners are cautioned against using the DTLA-2 overall score as the sole criterion for special education placement until fur ther research can be conducted. While additional studies also are needed to establish the reliability and validity of the composites, the DTLA-2 ap pears promising as an instrument for assessing in traindividual differences.

This study investigated the usefulness of the MAT-SF and the DAP: A Quantitative Scoring System as screening instruments. In this study, 44 regular and 33 LD classroom fourth- and fifth- grade students were group administered the screening tests and individually administered the Kaufman Test of Educational Achievement. The results found both groups to score within the average range on both screening tests and the LD group to score significantly lower on the achieve ment than on the two screening tests. The MAT-SF was found to correlate significantly with all areas of achievement for the normal group. The DAP did not correlate significantly with any areas for the normal group; it correlated significantly with reading, but not math, for the learning- disabled students. The results indicated that the MAT-SF may have more utility as a general screening instrument than does the DAP.

The Hiskey-Nebraska (H-NTLA), K-ABC Nonverbal Scale (NS), Leiter, and the WISC-R Performance Scale (using Anderson and Sisco deaf norms), were administered to 31 hearing-impaired (HI) children enrolled in self-contained, hearing- handicapped programs in a public school setting. Testing spanned a 3-month interval. The group mean score on the K-ABC NV (M = 94.10) was significantly lower than the Leiter mean score (M = 100.71) or the WISC-R PIQ mean score (M = 103.00), as demonstrated by significant dependent t-test analyses (p < .01, p < .001, respectively). All other t-test results, including the K-ABC NV to the H-NTLA (M = 99.21), were nonsignificant. The Pearson product-moment cor relations among the four instruments were all significant and ranged from .57 (K-ABC NV with H-NTLA, p < .01) to .78 (K-ABC NV with Leiter, p < .0001). While results showed the K-ABC NV to have ample construct validity, it appeared to underestimate the intellectual skills of the HI population. Separate deaf norms on the K-ABC NV would be necessary for the recom mended general use of this instrument.

Unlike its predecessors, the Fourth Edition of the Stanford-Binet Intelligence Scale (Thorndike, Hagen, & Sattler, 1986a) includes subtest and area standard scores as well as an overall composite standard score. Kaufman's (1979) popular method of profile analysis of intelligence test performance includes examination of differences between scores obtained within a total test profile. These score differences may be interpreted in two ways: the statistical significance of the difference and the ab normality of uniqueness of the difference. Tables are presented to aid in precise interpretation of these score differences at four age levels: 2 to 5 years, 6 to 10 years, 11 to 14 years, and 15 to 18 years. Implications of these data for test interpreta tion are discussed.

This study examined the in- and out-of-school adaptive behavior of 154 randomly selected Black, Hispanic and White 7-year-old children. Six mea sures of adaptive behavior were used, two ad ministered to each of three sources of information (parents, teachers, and children themselves)— Adaptive Behavior Inventory for Children; Mer cer's Teacher Questionnaire; Children's Adaptive Behavior Scale; and Behavior Rating Profile- Parent, Teacher, and Child. Overall, there was a consistent pattern of within-source correlations among adaptive behavior ratings higher than between-source correlations for the parent and teacher measures. That pattern was particularly clear for the minority samples, an indication of the relative independence of children's adaptive be haviors in home and school settings and the need to obtain information from both sources. The two children's measures were not correlated with each other and were not related strongly to other measures of adaptive behavior from other informa tion sources. Examination of intercorrelations among adaptive behavior measures and between adaptive behavior and indices of ability and school achievement provided some evidence of domains that were being addressed by these measures, with indications of both convergent and divergent validity for the various measures. Finally, multivariate analyses resulted in gender and ethnic group differences, and examination of correlation matrices for each ethnic group suggested possible differential meaning of adaptive behavior scores for the three groups.

To interpret a battery of tests, it is important to know how various measures are related among themselves. This article investigated the relation ships among the WISC-R (Wechsler, 1974), Matrix Analogies Test-Expanded Form (Naglieri, 1985a), and Peabody Picture Vocabulary Test- Revised (Dunn & Dunn, 1981) with a referred school-age population. Specifically, the significance of the divergent validity of the verbal and nonverbal measures was tested to determine the extent to which practitioners should expect such results. Additionally, the differences between specific pairs of subtests (e.g., PIQ vs. MAT-EF) required for significance when one is interpreting results for an individual subject were computed. Implications for the use and interpretation of these measures as part of a test battery are presented.

Stability of K-ABC performance over a 2-year in terval with 25 nonhandicapped preschool children was examined. Global scale stability coefficients (corrected for restriction in range) for Time 1 testing (age 4) and Time 3 testing (age 6) were: MPC = .77; SEQ = .73; SIM = .76; ACH = .80. Subtest stability coefficients ranged from .53 to .75 for the Time 1/Time 3 comparisons. Repeated measures analyses of variance revealed significant increases in mean SIM scores from age 4 and 5 to age 6, and a significant decrease in mean ACH score from age 4 to ages 5 and 6. Overall, results indicated a high level of stability for both global scale scores and subtest scores. Less stability was indicated for profile patterns (SIM > SEQ, SIM < SEQ SIM = SEQ) and subtest strengths and weaknesses with this sample of non handicapped preschool children tested at ages 4, 5, and 6.