ERIC Number: ED205604
Record Type: RIE
Publication Date: 1980-Nov
Comparison of Program Effects: The Use of Mastery Scores.
Yeh, Jennie P.; Moy, Raymond
The setting of a cut-off score on a mastery test usually involves a consideration of one or more of the following elements: (1) the distribution of observed test scores; (2) the type of mastery criterion used; (3) the level of acceptable risks of mis-classification; (4) the loss of functions of mis-classifications; and (5) the distribution of true scores. In order to illustrate several procedures for setting cut-off scores, and how various considerations may change the cut-off score value, a data set was obtained consisting of 99 foreign engineering graduate students' test scores on a sample of 87 items from the UCLA English as a Second Language proficiency test, their grade point average, the number of university courses failed, and percentile scores from the Graduate Record Examination. The methods examined include a norm-referenced vs. a theoretical criterion; cut-off scores based on acceptable risks of mis-classification; Huynh's optimal decision rule model; Wilcox's optimal cut-off score based on observed scores and an external criterion; and Wilcox's method for approximating true score distribution. (Author/BW)
Publication Type: Reports - Evaluative
Education Level: N/A
Sponsor: National Inst. of Education (ED), Washington, DC.
Authoring Institution: California Univ., Los Angeles. Center for the Study of Evaluation.