Abstract
A microcomputerized adaptive testing system was developed for placing students into various levels of college mathematics courses. Based on the student's math background, an initial cluster is identified for the testing. Using either a Bayesian or Maximum Likelihood decision, the testing is branched to a lower cluster or higher cluster depending on the student's performance. Based on the final estimate of math ability, appropriate courses are suggested. The feasibility of the system was examined with respect to three criteria: 1) item pool characteristics, 2) item performance, and 3) students' and advisors' reactions. Although the results show room for improvement, the system is promising as a placement instrument.
Get full access to this article
View all access options for this article.