Literaturnachweis - Detailanzeige
Autor/inn/en | Reckase, Mark D.; Ackerman, Terry A. |
---|---|
Titel | Building a Test Using Items That Require More than One Skill to Determine a Correct Answer. |
Quelle | (1986), (38 Seiten)
PDF als Volltext |
Sprache | englisch |
Dokumenttyp | gedruckt; online; Monographie |
Schlagwörter | Achievement Tests; College Entrance Examinations; Estimation (Mathematics); Goodness of Fit; High Schools; Item Analysis; Latent Trait Theory; Mathematical Models; Mathematics Tests; Psychometrics; Research Methodology; Sample Size; Tables (Data); Test Construction; Test Items; ACT Assessment Achievement test; Achievement; Testing; Test; Tests; Leistungsbeurteilung; Leistungsüberprüfung; Leistung; Testdurchführung; Testen; Aufnahmeprüfung; Estimation; Mathematics; Schätzung; High school; Oberschule; Itemanalyse; Latent-Trait-Modell; Mathematical model; Mathematisches Modell; Psychometry; Psychometrie; Research method; Forschungsmethode; Tabelle; Testaufbau; Test content; Testaufgabe; Assessment; Eignungsprüfung; Eignungstest; Hochschulzulassung |
Abstract | This paper demonstrates the relationship between the concept of unidimensionality and direction of an item in a multidimensional space. The basic premise is that if items that measure in the same direction are combined to form a test, that test will meet the item response theory requirements of unidimensionality. This will be true even if the items measuring in the same direction measure more than one psychological construct. A form of the ACT Mathematics Usage Test was analyzed using the multidimensional extension of the two-parameter logistic model to determine the direction for each item using the multidimensional difficulty formulated by Reckase. Using the direction, three unidimensional sets of items were formed and one multidimensional item set. The performance of 1,000 examinees on these items sets was analyzed using LOGIST 5 to determine the fit of the three-parameter logistic model to the data and the relationship of the unidimensional ability estimates and item parameter estimates to the multidimensional counterparts. Overall, the results strongly support the conception of unidimensionality suggested by a common direction in the multidimensional space for a set of items and the use of multidimensional difficulty statistics in forming unidimensional item sets. (PN) |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |