The display of multiple-choice item locations in ascending difficulty, next to the written-response thresholds, helps to characterize the scale in terms of what increasing proficiency “looks like” in the pool of test-takers. For example, if a
committee were considering 500 as a cut point between performance levels, it could note that 500 is a point at which items like 34, 43, 45, and 48 are expected to be chosen correctly about 50% of the time, a harder item like 37 is expected to be chosen correctly about 30%, and easier items like 2 are expected to be chosen correctly 80% of the time. The set of multiple-choice items, sorted so they are in order of ascending difficulty, is available to the committee so that the members can relate these probabilities to their understanding of the items. The committee could also note that a student at that point (i.e., 500), would be equally
likely to score a 2 or a 3 on the first written-response item (40% each) and more likely to score a 2 than a 3 on the second (56% vs. 26%). Examples of student work at these levels would be available to the committee for consideration of the interpretation of these scores. Committee members can examine the responses of selected examinees to both the multiple-choice and written-response items, chart their locations on the map, and judge their levels.
The display of multiple-choice item locations in ascending difficulty, next to the written-response thresholds, helps to characterize the scale in terms of what increasing proficiency “looks like” in the pool of test-takers. For example, if acommittee were considering 500 as a cut point between performance levels, it could note that 500 is a point at which items like 34, 43, 45, and 48 are expected to be chosen correctly about 50% of the time, a harder item like 37 is expected to be chosen correctly about 30%, and easier items like 2 are expected to be chosen correctly 80% of the time. The set of multiple-choice items, sorted so they are in order of ascending difficulty, is available to the committee so that the members can relate these probabilities to their understanding of the items. The committee could also note that a student at that point (i.e., 500), would be equallylikely to score a 2 or a 3 on the first written-response item (40% each) and more likely to score a 2 than a 3 on the second (56% vs. 26%). Examples of student work at these levels would be available to the committee for consideration of the interpretation of these scores. Committee members can examine the responses of selected examinees to both the multiple-choice and written-response items, chart their locations on the map, and judge their levels.
การแปล กรุณารอสักครู่..