SUMMARY. The ' creativity ' as an entity independent of intelligence and aptitudes closely related to intelligence (Cureton, 1964 ;Thorndike, 1963 ;Wodtke, 1964). It is quite clear that although the problem of separating creativity and intelligence has been raised frequently (Barron, 1963 ;Getzels and Jackson, 1962 ; Guilford, 1950 ;McNemar, 1964 ;Mednick and Mednick, 1965 ;Torrance, 1959 ; Wdlach and Kogan, 1964 ; Yamamoto, , 1966, no adequate resolution has yet occurred. The persistence of the problem is in part due to the wide variety of creativity measures, often poorly validated and with low or unknown reliabilities, and in part to the conceptual confusion and variety subsumed under the rubric ' creativity ' (cf. Cureton, 1964 ;Ginsburg, 1964 ;Wodtke, 1964). Moreover, as McNemar (1 964) has pointed out, the current fad among most creativity investigators is either to discount the role of intelligence or to assume that i t is not importantly involved in their performance measures of creativity.McNemar argued that there should be a triangular scatter diagram between ' I honest to goodness criterion measures . . . of creativity " and IQ, where having " a high IQ is not a guarantee of being creative ;having alowIQ meanscreativity is impossible " (1964, p. 879). The purpose of this paper is to report a direct application of McNemar's challenge and to discuss the ramifications of the results.11,-METHOD.
McNemar, 1964 ;Form I of Mednick's (1962) Remote Associates Test (RAT) was selected as our measure of ' creative potential ' (a more accurate term than ' creativity,' for reasons mentioned below), and ACT-Verbal scores were used as a measure of scholastic atptitude. Both tests were administered to 292 entering freshmen a t the University of Nevada (Reno) in September, 1965, as part of the entrance examination program. The ACT-Verbal scores are assumed t o reflect verbal IQ, which was of direct interest in this study because of the verbal nature of the RAT. This assumption appears justified both by the strong, positive relationship between ACT and IQ scores and by the similar results obtained when ACER Verbal IQ and RAT scores were correlated in another population.
I33