ERIC Number: ED210309
Record Type: RIE
Publication Date: 1981
A Bayesian Method for Assigning Course Grades.
Powers, James E.
The use of Bayesian methodology to assign grades in classroom situations is presented. Assigning a grade is viewed from a criterion, as opposed to norm, referenced perspective. Criteria include mastery of some proportion, determined by the teacher, of the subject matter covered in a course. Different levels of mastery are deemed possible and, therefore, the assignment of different grades to different students may be appropriate. A prior probability distribution is assigned to the level of mastery, in terms of proportions, of each student at the outset of the course. After the first test is administered in the course, a likelihood distribution is obtained for each student. This distribution reflects the probability that the observed score would have been obtained, given each previously specified level of mastery. A posterior probability distribution on level of mastery is obtained for each student by combining the prior and likelihood distributions through the application of Bayes formula. For all subsequent tests, the posterior from the previous test is taken as the prior for the current test. A complete example illustrating the necessary calculations is included as well as a brief description of the Bayesian method. (Author/DWH)
Publication Type: Speeches/Meeting Papers; Reports - Evaluative
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the New England Educational Research Organization (Lenox, MA, 1981).