Literaturnachweis - Detailanzeige
Institution | Civil Service Commission, Washington, DC. Personnel Research and Development Center. |
---|---|
Titel | Computers and Testing: Steps Toward the Inevitable Conquest. Professional Series 76-1. |
Quelle | (1976), (45 Seiten)
PDF als Volltext |
Dokumenttyp | gedruckt; online; Monographie |
Schlagwörter | Adaptive Testing; Bayesian Statistics; Computer Oriented Programs; Computer Programs; Guessing (Tests); Item Analysis; Item Banks; Latent Trait Theory; Mathematical Models; Multiple Choice Tests; Probability; Response Style (Tests); Scoring; Simulation; Test Validity Computerprogramm; Computer program; Computer programme; Erraten; Itemanalyse; Latent-Trait-Modell; Mathematical model; Mathematisches Modell; Multiple choice examinations; Multiple-choice tests, Multiple-choice examinations; Multiple-Choice-Verfahren; Wahrscheinlichkeitsrechnung; Wahrscheinlichkeitstheorie; Antwortverhalten; Bewertung; Simulation program; Simulationsprogramm; Testvalidität |
Abstract | This pamphlet reprints three papers and an invited discussion of them, read at a Division 5 Symposium at the 1975 American Psychological Association Convention. The first paper describes a Bayesian tailored testing process and shows how it demonstrates the importance of using test items with high discrimination, low guessing probability, and a uniform (rectangular) distribution of difficulty. The second paper describes a relatively easy, inexpensive, and reliable method of estimating these three item parameters. The procedure assumes that the underlying ability is normally distributed. Data were examined to determine the reasonableness of the assumption. The third paper reports the results of two simulation studies evaluating the effectiveness of the parameterization technique, showing that the procedure works well under "standard" conditions (at least 2000 cases and 100 items) and is adequate with 1000 cases and 60 items. The discussion paper points out that accurate estimations of item parameters are essential to successful tailored testing; compares the Bayesian technique with parameterization based on maximum likelihood estimates; and concludes that there is now no apparent obstacle to successful operational tailored testing. (Author) |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |