Suche

Wo soll gesucht werden?
Erweiterte Literatursuche

Ariadne Pfad:

Inhalt

Literaturnachweis - Detailanzeige

 
Autor/inXiao, Beiling
TitelDichotomous Search Strategies for Computerized Adaptive Testing.
Quelle(1990), (21 Seiten)Verfügbarkeit 
Spracheenglisch
Dokumenttypgedruckt; Monographie
SchlagwörterAbility Identification; Adaptive Testing; Computer Assisted Testing; Equations (Mathematics); Estimation (Mathematics); Mathematical Models; Maximum Likelihood Statistics; Monte Carlo Methods; Robustness (Statistics); Scoring; Search Strategies; Simulation; Test Construction; Test Items; Testing Problems
AbstractDichotomous search strategies (DSSs) for computerized adaptive testing are similar to golden section search strategies (GSSSs). Each middle point of successive search regions is a testing point. After each item is administered, the subject's obtained score is compared with the expected score at successive testing points. If the subject's obtained score does not exceed a confidence interval of an expected score at a testing point, the subject's current ability estimate is assumed to be equal to that of the testing point. Otherwise, the upper or lower half of the search region is discarded and the process is continued until the test-taker's current ability estimate is determined and the next item is selected. Monte Carlo studies with 3,300 subjects in 33 ability levels using one-parameter and three-parameter models to compare the efficiency and accuracy of DSSs, GSSSs, and maximum likelihood estimate strategies (MLESs) indicated that all three measured well in the one-parameter situation, but that DSSs and GSSSs were more accurate and efficient than were MLESs for the three-parameter model. Both DSSs and GSSSs were more robust against guessing and simpler to operate than were MLESs. Twelve graphs illustrate the study. (Author/SLD)
Erfasst vonERIC (Education Resources Information Center), Washington, DC
Literaturbeschaffung und Bestandsnachweise in Bibliotheken prüfen
 

Standortunabhängige Dienste
Da keine ISBN zur Verfügung steht, konnte leider kein (weiterer) URL generiert werden.
Bitte rufen Sie die Eingabemaske des Karlsruher Virtuellen Katalogs (KVK) auf
Dort haben Sie die Möglichkeit, in zahlreichen Bibliothekskatalogen selbst zu recherchieren.
Tipps zum Auffinden elektronischer Volltexte im Video-Tutorial

Trefferlisten Einstellungen

Permalink als QR-Code

Permalink als QR-Code

Inhalt auf sozialen Plattformen teilen (nur vorhanden, wenn Javascript eingeschaltet ist)

Teile diese Seite: