ABSTRACT
The practice of requiring test takers to complete a test within a specific time period dates to Yerkes’s administration of IQ tests to army recruits in 1917. With thousands of recruits to test and rooms large enough to hold only fifty at a time, he had no choice but to use a time limit. These and other compromises he made to establish a baseline scale for intelligence testing have undermined the legitimacy of timed tests ever since (Gould 1981). Previously, students were given whatever period of time they needed to complete tests or until it was apparent to the proctor that the student was unable to finish. This older approach to the duration of tests survives today in master’s and doctoral written comprehensive exams and in the tradition of assigning take-home essays as exams. Many experienced instructors believe that if the students know the material, they ought to be able to complete the test in the allotted time. They argue that timed tests measure students’ grasp of course content because the instructors design the test for those time limits. These assertions miss the point that timed tests unnecessarily stress students. After searching for the evidence, I have not found a double-blind test result that timed tests evaluate control over the course material better than untimed ones.
