Concurrent validity of the Stanford-Binet, 4th Edition: Agreement with the WISC—R in classifying learning disabled children.

Abstract

Concurrent validity of the Stanford-Binet, 4th Edition (SBIV) was studied in 80 learning disabled children assigned to 3 groups according to Verbal and Performance IQ differences on the Wechsler Intelligence Scale for Children-Revised (WISC-R). The groups were designated Auditory-Linguistic (Verbal IQ < Performance IQ), Visual-Spatial (Performance IQ < Verbal IQ), and Mixed (Verbal IQ = Performance IQ). Multivariate analyses of variance indicated highly significant differences between the 3 groups on various SBIV scores. Discriminant analyses using SBIV scores at different levels revealed variable classification rates, with the 12 subtest scores showing the highest rate (75%) in correctly classifying the children into the 3 groups. Correlations between various SBIV scores and WISC-R scores for the total group were generally lower than expected, ranging from low to moderately high. Possible explanations for the discrepancy are discussed as well as potential problems in using the 2 scales as equivalent measures in identifying and classifying learning disabled children.

The Predictive Value of IQ

“…These coefficients range from low to high: for example, for two groups of children with learning disabilities, the correlations between the SBIS and WISC-R ranged from 0.49 (Brown & Morgan, 1991) to .92 (Phelps, Bell, & Scott, 1988). Moreover, the validity coefficients resulting from the studies of gifted children are consistently lower, such as .21 (McCall et al, 1989), .39 (Phelps, 1989), .70 (Hayden et al, 1988), than they are for broader samples, almost certainly in part because of restriction of range.…”

Assessing Intellectual Functioning in Young Adolescents: How do the WISC-IV and SB5 Compare?

“…Previous studies comparing earlier editions of the two tests produced mixed results but, in general, moderate to high correlations were reported, with the majority ranging from .68 to .88 (Carvajal et al, 1993; Greene, Sapp, & Chissom, 1990; Hollinger & Baldwin, 1990; Lavin, 1996; Lukens, 1990; Lukens & Hurrell, 1996; Prewett & Matavich, 1994; Roid, 2003b; Rust & Lindstrom, 1996; Saklofske, Schwean, Yackulic, & Quinn, 1994; Simpson et al, 2002). At times, however, there have been relatively large differences between composite scores on the two tests, with the higher scoring instrument varying across studies (Brown & Morgan, 1991; Lukens & Hurrell, 1996; Minton & Pratt, 2006; Prewett & Matavich, 1994).…”