Abstract
Two versions of the Nedelsky procedure for setting minimum passing scores are compared. Two groups of judges, one using each version, set pass ing scores for a classroom test Comparisons of the resulting sets of passing scores are made on the ba sis of (1) the raw distributions of passing scores, (2) the consistency of pass-fail decisions between the two versions, (3) the consistency of pass-fail deci sions between each version and the passing score established by the test designer, and (4) the mean pairwise agreement between judges across groups. The two versions of the procedure are found to pro duce essentially equivalent results. In addition, a significant relationship is observed between the passing score set by a judge and that judge's level of achievement in the content area of the test.
Get full access to this article
View all access options for this article.
