EFFECTS OF RESTRICTION OF RANGE AND TEST UNRELIABILITY ON CORRELATION BETWEEN MEASURES OF INTELLIGENCE AND CREATIVE THINKING*

Abstract

S ummary . The Lorge‐Thorndike Intelligence Test, Level 3, Form A, and the Minnesota Tests of Creative Thinking (Torrance, 1962) were given to groups of 461 and 827 fifth‐grade American children. Subjects were divided into four groups, according to I.Q. range: below 90, 91 to 110, 111 to 130, and above 130. Linear correlations between I.Q. and creativity score for the whole population and for the sub‐groups gave few significant coefficients. Corrections for explicit selection and for unreliability of the creativity measures indicated that the ‘true’ correlation might be as high as 88. There was consistent decrease in size of the correlation as the level of I.Q. of the sub‐groups became higher, this lending support to the idea that, beyond a certain minimum level of intelligence, being more intelligent, does not guarantee a corresponding increase in the creativity. The results do not, however, support the view that creativity is an entity independent of other facets of human intellect.