| Sumario: | A study was conducted to investigate three computer-based credentialing test designs—the Linear Parallel-Forms Test, multistage (MST), and the Computer Adaptive Test (CAT)—under common examination conditions. Findings indicated that item bank size and item quality had a practically significant influence on decision consistency and accuracy; and that the choice of test design was not a factor in the results, even in almost ideal situations. It was concluded that more time and resources should be committed to expanding the size and quality of item banks; and that designs that individualize an exam administration, such as MST and CAT, may not be helpful when the primary role of the examination is to make pass-fail decisions and when conditions suit using parallel forms with a target information function that can be centered on the passing score.
|
|---|