Knowledge that Transforms

An industrial validation of a measure of creative potential, the Remote Associates Test (RAT), found the correlation between RAT scores and supervisory ratings of creativity for 31 physicists to be +.13, with 6 of the 10 high-rated scientists having very low RAT scores. These scientists do not speak English as their native language although their linguistic fluency was no different from that of the other scientists. The proportion of nonEnglishspeaking scientists in this sample similar to that reported for eminent mathematicians; the RAT may be limited as a measure of creative potential in this occupational group. This note reports an attempt at industrial validation for a new measure of creative potential, the Remote Associates Test (RAT). Central to many definitions of creativity the achievement of original and worthwhile ideas. By defining original as statistically infrequent and worthwhile as meeting certain parameters of meaning, Mednick (1962) has developed a measure of the ability to demonstrate unusual and satisfactory associations. In each item of this test, three apparently unrelated words are given and the subject (5) required to name a fourth word which is related to all three; for example, balloon, soda, and Dad have pop as an associate. Mednick reports that performance on the RAT predicts the judged creativity of psychology and design students, and of architects; research currently under way with a number of other groups. This note reports, however, an instance in which RAT scores did not predict rated creativity; further inquiry into the results suggests a limitation of the RAT for certain occupational groups. The 5s were 31 research scientists (physicists, physical chemists, mathematical physicists) in a large industrial space sciences laboratory. Each scientist was rated by his immediate supervisor on a scale of creativity derived from Taylor's measure (1962). The rho between RAT scores and rated creativity