The purpose of the present investigation was to examine the predictive validity of Kaufman's short form version of the McCarthy Scales of Children's Abilities for samples of English-speaking and Spanish-speaking Mexican-American children. Comparisons of the observed correlations between the McCarthy General Cognitive Index (GCI) and Kaufman's estimated GCI with academic achievement (as measured by the Comprehensive Tests of Basic Skills) showed that the conventional McCarthy and Kaufman's short form predicted achievement about equally well. Implication of this finding was discussed in light of supportive evidence for the validity of the Kaufman short form and in the context of screening of culturally diverse children.
Abbreviated tests that have been developed from established intelligence tests have been present since the 1937 Stanford-Binet. However, the current interest in these shortened versions, often referred to as "short forms," has been revitalized by recent legislation emphasizing the need to detect young children who are likely to experience problems in school learning (Kaufman, 1977). Given the need for more rapid mass screening of young children who may have potential learning problems, short forms that are reliable, valid, and can be administered in a relatively short period of time would be valuable tools for practitioners in school psychology. Clearly, research in the development of sound short forms is needed.
An example of one such line of research has been the development of a short form from the McCarthy Scales of Children's Abilities (MSCA) (McCarthy, 1972), a well known and established cognitive instrument. Currently, three different McCarthy short forms exist (Kaufman, 1977; McCarthy, 1978; Taylor, Slocumb, & O'Neil, 1979). In a recent review of the MSCA, which included a review of these three short forms, Kaufman (1982) has contended there is some evidence that the Kaufman version (Kaufman, 1977) may hold the most promise as an effective short form screening tool. In contrast, the McCarthy Screening Test (McCarthy, 1978) has been noted as having poor reliability and stability and as being a poor predictor for identifying learning disabled children referred by their teachers (Kaufman, 1982). The third McCarthy short form, developed by Taylor, Slocumb, and O'Neill (l979), has been criticized as a viable short form because it was developed from data from a very small and nonrepresentative sample, used an inappropriate regression analysis, and contains subtests that are not representative of the different scales of the MSCA (Kaufman, 1982).
Although there have been numerous recent studies of the Kaufman short form version of the McCarthy, virtually no research has been done utilizing samples of minority children. In the case of Mexican-American children, only one study exists (Valencia & Rankin, 1983). This investigation examined the concurrent validity and reliability of the Kaufman short form with a large sample of preschool and kindergarten English-and Requests for reprints should be sent to