ESTIMATING THE OPTIMUM NUMBER OF OPTIONS PER ITEM USING AN INCREMENTAL OPTION PARADIGM

Previous studies have demonstrated the efficacy of using three-option multiple-choice items. In these studies three- and four-option items w ere constructed from preexisting item analysis data obtained from five -option items. In this study, a two-option test was constructed, and o ptions were systematically added to this test by using a taxonomy of i tem writing rules to guide the process of distractor development. Nons ignificant differences (p greater-than-or-equal-to .05) were found amo ng the reliability coefficients, the reliability estimates for the thr ee-, four-, and five-option formats being all on the same order of mag nitude. These findings continue to provide evidence for the efficacy o f the three-option item.