NotesFAQContact Us
Search Tips
ERIC Number: ED399294
Record Type: Non-Journal
Publication Date: 1992-Apr
Pages: 6
Abstractor: N/A
Reference Count: N/A
A Compromise Grading Model for Classroom Tests.
Johanson, George A.
Most educational measurement texts distinguish between norm-referenced (NR), or relative, methods of assigning letter grades to objective test scores, and criterion-referenced (CR), or absolute, methods. Both NR and CR approaches have serious limitations in typical classroom situations, and neither approach, in its pure form, may be entirely suitable. An alternative method is proposed and illustrated with scores from 57 secondary school students taking a 26-item objectively scored test. The approach involved using a smoothed or fitted cumulative distribution and a ratio of standard errors to fix the slope of the line through the ideal cut-points. This is a modification of the method of C. H. Beuk (1984). The rationale for this type of compromise is that it acknowledges the sample status of both the set of test items and the group of examinees and shares sampling error equally between NR and CR methods. The algorithm has been programmed in PASCAL for the microcomputer. A structured grading method of this sort would allow teachers of multiple sections or those within the same department to give somewhat comparable grades to their students if they used agreed-on NR standards and individual CR standards. This compromise would be especially useful when an entirely new test is used or an unfamiliar group of students is encountered. (Contains 1 table, 2 figures, and 10 references.) (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A