A method of determining the minimum number of questions which should be included in a multiple-choice test is described. This method ensures a desired probability of correctly ranking any given pair of examinees who know answers to different proportions of questions in the test.
References
1.
HsuL.Determination of the risk of mis-ranking a pair of examinees on a multiple-choice test. Perceptual and Motor Skills, 1978, 46, 1265–1266.
2.
WeitzmanR. A.Ideal multiple-choice items. Journal of the American Statistical Association, 1970, 65, 71–89.
