The present study attempted to develop a quantitative model using the WISC-R that could be used to predict those students most likely to be successful in gifted education programs. The study was conducted in two phases using two groups of gifted children. In phase one, 120 elementary students randomly chosen from a pullout program in a suburban school served as the subject pool. The subjects had varying degrees of success in the program. Phase one subject data were used to develop a quantitative model; phase two subjects were used for predictive purposes. Specific results in phase one yielded six maximally discriminant WISC-R subtests. These subtests were then used to predict at well above chance levels (87.8%) those students who were known to have been either marginally or highly successful in the program (n = 41). The present study goes one step beyond previous research by using these maximally discriminant subtests to predict program performance of gifted children.
A great deal of time, effort, and money is spent on the process of proper identification and education of gifted children (Oglesby & Gallagher, 1983). Debate has been intense in the field regarding these topics, and despite this often long and costly process, methods of evaluation and identification are still under attack (Kirschenbaum, 1983;Renzulli, 1976). This is in part a response to the purportedly high number of false positives and false negatives that are found in the selection pool (Harrington, 1983) because researchers are still attempting to determine the specific characteristics of gifted children and how to identify them.
According to Mueller, Matheson, and Short (1983), the Wechsler Intelligence Scale for Children-Revised (WISC-R; Wechsler, 1974) is currently the single most popular and well-researched psychometric instrument for the assessment of intellectual functioning of individual children in both clinical and academic settings. Research has begun utilizing WISC-R subtest patterns with different populations, including the gifted, to investigate the cognitive profiles that emerge for them (Mueller, Matheson, & Short, 1983).
A number of researchers have attempted to develop WISC-R short forms for the process of identifying gifted groups. General findings in this area have consistently shown that the Vocabulary and Block Design subtests have been the most valuable in predicting WISC-R Full Scale IQ among the gifted (Dirks, Wessels, Quarfoth, & Quenon, 1980; Elman, Blixt, & Sawicki, 1981). The findings, however, have not been applied directly to the selection process to predict successful candidates for gifted programs. Due to these factors, many gifted programs contain a significant number of children who are Requests for reprints should be sent to Richard Lustberg, 117-01 Park Lane South, Kew Gardens, New York 11418.