NotesFAQContact Us
Collection
Advanced
Search Tips
Back to results
ERIC Number: ED386492
Record Type: Non-Journal
Publication Date: 1993-Oct
Pages: 31
Abstractor: N/A
ISBN: N/A
ISSN: N/A
EISSN: N/A
Available Date: N/A
Estimating the Consistency and Accuracy of Classifications Based on Test Scores.
Livingston, Samuel A.; Lewis, Charles
This paper presents a method for estimating the accuracy and consistency of classifications based on test scores. The scores can be produced by any scoring method, including the formation of a weighted composite. The estimates use data from a single form. The reliability of the score is used to estimate its effective test length in terms of discrete items. The true-score distribution is estimated by fitting a four-parameter beta model. The conditional distribution of scores on an alternate form, given the true score, is estimated from a binomial distribution based on the estimated effective test length. The agreement between classifications on two alternate forms is estimated by assuming conditional independence, given the true score. An evaluation of the method showed that the estimates of the percent of test-takers correctly classified and the percent consistently classified were within one percentage point of the actual values in most cases. Although the estimated effective test length and the estimates of the conditional standard error of measurement are sensitive to changes in the specified minimum and maximum possible scores, the estimates of the decision accuracy and decision consistency statistics are not. (Contains seven tables and seven references.) (Author)
Publication Type: Reports - Evaluative
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: Educational Testing Service, Princeton, NJ.
Grant or Contract Numbers: N/A
Author Affiliations: N/A