New Information about Applications of Scale Anchoring.

Autor/inn/en	Isham, Steven P.; Allen, Nancy L.
Titel	New Information about Applications of Scale Anchoring.
Quelle	(1992), (29 Seiten)Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; Monographie
Schlagwörter	Academic Achievement; Criterion Referenced Tests; Elementary School Students; Elementary Secondary Education; Item Response Theory; Multiple Choice Tests; National Norms; National Programs; Rating Scales; Scaling; Scores; Secondary School Students; Student Evaluation; Test Items; Test Reliability; Testing Programs; Trend Analysis; National Assessment of Educational Progress + Suchen Sie Ihr Suchwort? Schulleistung; Item-Response-Theorie; Multiple choice examinations; Multiple-choice tests, Multiple-choice examinations; Multiple-Choice-Verfahren; nicht übertragen; Rating-Skala; Scale construction; Skalenkonstruktion; Sekundarschüler; Schulnote; Studentische Bewertung; Test content; Testaufgabe; Testreliabilität; Trendanalyse
Abstract	As a result of the dual roles of the National Assessment of Educational Progress (NAEP) to measure trends in academic achievement over time and to measure what students know and can do, a scale anchoring procedure was developed. Although the NAEP provides norm-referenced information about student proficiency, the scale anchoring procedure gives criterion-referenced information about the NAEP scales. Two studies compared results of scale anchoring the same items using data for three different grades and compared results of scale anchoring based on subscale proficiencies as opposed to composite proficiencies. In all, 275 items in 5 subscales were given to 19,251 students in grades 4, 8, and 12. Scale anchoring within each grade identifies slightly different sets of items as opposed to those identified by anchoring with all grade levels. However, only two items anchored at different levels for different grades. These items had different item characteristic curves for the different grades and were scaled separately for the different grades when the items were calibrated. Scale anchoring using subscales, as opposed to composite scales, also identifies slightly different sets of items. In both cases, items that anchor under one set of conditions but not another come very close to anchoring under the second set of conditions. There are nine tables of study data and an eight-item list of references. (Author/SLD)
Erfasst von	ERIC (Education Resources Information Center), Washington, DC