In my recent blog post about the Apple logo experiment, I made the claim that “divergent thinking” tests of creativity were not reliable. In response, I received a collegial but politely critical email from Bonnie Cramond, Director of the Torrance Center at the University of Georgia. The center is named for legendary creativity researcher Paul Torrance, who is primarily known today as the developer of the Torrance Tests of Creative Thinking, or TTCT. As I wrote in my 2012 book Explaining Creativity, the TTCT is the most thoroughly developed and most widely used test of creativity, and it’s largely based on divergent thinking tasks. After reviewing many research articles about the TTCT, I concluded that it does not meet the levels of reliability and validity that psychologists expect from tests of human abilities. For example, there are several studies that fail to find a relationship between TTCT scores and real-world creativity–suggesting that whatever it is measuring, the TTCT might not be a “valid” measure of creativity.
Professor Cramond rejects this conclusion. And she can point to different studies that find a predictive relationship between TTCT scores and real-world creativity. Regarding reliability, in our email exchange, she argued that you can manipulate the outcome of any test by modifying the instructions–even IQ tests. As an example, she referred to the famous “Mozart effect” study (from 1993) that found that when college students listen to Mozart, and take a test of spatio-temporal reasoning a few minutes afterward, their scores go up. (The test that was used was one of the components of the Stanford-Binet IQ test, but it was not the complete IQ test.)
It would be wonderful if educators and schools had a good measure of creative potential. Professor Cramond and I agree that the TTCT is the best one we have. Have you used the TTCT? What is your experience? Are you a fan, or are you skeptical?