A Randomized Comparison of Direct Assessment Techniques

The purpose of this study was to compare, within a single class of chiropractic students, results achieved when administered a multiple-choice and completion exam in which the information tested was identical. The study was carried out in a classroom of 47 students. All students were informed that, within a single testing session, they would be administered two 10-question exams, but not informed about the format of either exam. The course director formulated 10 multiple choice questions, from which 10 completion test items were written, and the tests were randomly distributed. Achievement was significantly higher on the multiple choice exam as compared to the completion exam (mean=8.14+1.37 and 6.91+2.25 respectively; F=8.791; df=1,84; P=0.004). No significant differences were observed in scores based on the order the exams were taken (F=2.105; df=1,84; P=0.151). No interaction effect was observed between the type of exam taken and the order in which it was taken (F=1.106; df=1,84; P=0.296). The results of this study indicate that the student's ability to recognize a learned concept was superior to their ability to recall that same concept. While this study supports the concept that multiple choice tests may prove efficacious for testing basic recognition, completion tests may in fact challenge the student to think and respond at a higher level of learning. With expediency being important in evaluation and feedback, multiple choice items may be used but with stems that require more analysis as compared to basic knowledge.