NotesFAQContact Us
Search Tips
ERIC Number: ED350355
Record Type: Non-Journal
Publication Date: 1992
Pages: 35
Abstractor: N/A
Reference Count: N/A
An Information-Based Approach To Monitoring Content Validity and Determining the Relative Value of Polytomous and Dichotomous Items.
Yamamoto, Kentaro; Kulick, Edward
Test items are designed to be representative of the subject areas that they measure and to reflect the importance of specific domains or item types within those subject areas. Content validity is achieved by content specification and number of items in each content domain included in the design of the test. However, largely due to the normal attrition of less desirable items during the field test phase, the averaged proportional domain information of the 1990 National Assessment of Educational Progress (NAEP) reading assessment deviated from the original design. The relative value of polytomous and dichotomous items in a 1990 NAEP science assessment booklet was evaluated. The current study included 1,248 students from the NAEP grade 8/age 13 years sample who were administered booklet 20, which contains 3 science blocks. Results show that the polytomous items were only as informative as dichotomous items in regard to both the model-based relative information values as well as the reduction in the variance of posterior proficiency distribution. It is contended that in the future, there should be a greater concern that the reporting score reflects the content validity of the instrument specifications. The components of information should be monitored routinely by the item response models or by the content, and the reduction of the posterior variance should be monitored through additional items. Six tables and four graphs illustrate the analysis. (Author/SLD)
Publication Type: Reports - Evaluative
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers - Assessments and Surveys: National Assessment of Educational Progress