Creativity Assessment

Divergent Production Tests

In the 1950s, J. P. Guilford developed his divergent production tests to measure the divergent production content in his Structure of Intellect. In the 1960s, E. P. Torrance developed The Torrance Tests of Creative Thinking, with subtests with names such as Ask and Guess, Unusual Questions, Product Improvement, Unusual Uses, and Just Suppose. The tests were scored for Guilford's divergent production aspects, such as fluency, flexibility, elaboration, transformation, and originality. The results of these subtests were factored into something called a Creativity Index, similar to a quotient, or a composite score. Present-day users of the tests recommend using individual profiles and dimensions. The implication became that the higher the score, the more potentially creative the child was. Other researchers besides Guilford and Torrance included Jacob Getzels, Philip Jackson, Nathan Kogan, Mark Runco, and Michael Wallach. Journals such as the Journal of Creative Behavior and the Creativity Research Journal have published many international studies about the use of divergent production assessment instruments.

Evaluating Creativity Assessment Instruments

Validity is best described as the test's truthfulness. The three types of validity are (1) content/construct validity, (2) criterion validity, and (3) concurrent validity. Construct validity is limited by the lack of a universal definition for creativity and the complexity of what creativity is. Is there a separate creativity construct, creativity ability? Criterion validity refers to whether or not test scores predict performance later on in life or in other areas. Criterion validity is also called predictive validity. Concurrent validity shows that a certain test is highly correlated with previously validated measures.

Most creativity tests seem to be based on the original divergent production tests that were validated by Guilford back in the 1950s. What is called creativity is again the question, for many tests show strong concurrent validity when related to other tests that measure divergent production, which may have little relationship to real-life creativity. Another assumption in creativity assessment has been that creative people have above-average IQs. This is called the threshold theory. Do creative people have to be highly intelligent? Average IQs for people in various domains differ. Validity studies of creativity tests have shown that although concurrent validity is often adequate, criterion validity and construct validity are not.

Reliability is the consistency of the instrument. Adequate reliability does not imply that the test is valid. The scoring of creativity tests is difficult, because the scoring requires subjective judgment and scorers must be trained. The person using creativity testing must consider three interrelated types of reliability: (1) stability, (2) equivalence, and (3) internal consistency. The reliability of tests can be increased by administering the tests in a standardized way, by using objective scoring measures, by having item difficulties that are equal (is listing unusual uses for a ball easier than listing unusual uses for a bat?), by having the test measure only one aspect of creativity, and by increasing the number of items on the test. Administering the tests in a standardized way and scoring them objectively is particularly difficult for divergent production tests.

...