NotesFAQContact Us
Search Tips
ERIC Number: ED454240
Record Type: Non-Journal
Publication Date: 2001-Apr
Pages: 15
Abstractor: N/A
Reference Count: N/A
The Classification Accuracy of Shortened versus Full Length Tests with Number Correct Scoring.
Schulz, E. Matthew; Wang, Lin
In this study, items were drawn from a full-length test of 30 items in order to construct shorter tests for the purpose of making accurate pass/fail classifications with regard to a specific criterion point on the latent ability metric. A three-item parameter Item Response Theory (IRT) framework was used. The criterion point on the latent ability metric corresponded to a criterion domain true score (80% correct) established by an expert panel. The shorter tests were compared to the full-length test in terms of classification accuracy. Number correct (NC) scoring was used. It was found that the classification accuracy of shorter tests met or even exceeded that of the full-length test. Results suggest that, in general, a test targeted on a specific level of ability can be about half the length of a test designed to classify examinees with regard to several (five) levels of ability, without compromising classification accuracy. For lower levels of ability, where guessing at difficult items on the test contributes more measurement error than information, tests can be shortened even more. These conclusions are limited to tests in which pass/fail decisions are based on a number correct score. (Author/SLD)
Publication Type: Reports - Evaluative; Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A