Literaturnachweis - Detailanzeige
| Autor/inn/en | Buhr, Dianne C.; Algina, James |
|---|---|
| Titel | A Comparison of Item Parameter Estimates and Ability Parameter Estimates Obtained by Different Methods Implemented by BILOG. |
| Quelle | (1986), (38 Seiten)
PDF als Volltext kostenfreie Datei |
| Sprache | englisch |
| Dokumenttyp | gedruckt; online; Monographie |
| Schlagwörter | Ability; Bayesian Statistics; Comparative Analysis; Computer Oriented Programs; Correlation; Estimation (Mathematics); Item Analysis; Latent Trait Theory; Mathematical Models; Maximum Likelihood Statistics; Statistical Distributions; Statistical Studies |
| Abstract | The focus of this study is on the estimation procedures implemented in BILOG, a computer program. One purpose is to compare the item parameter estimates produced by various procedures available in BILOG. Four different models are used: the one, two, and three parameter model and a three parameter model with common guessing parameters. The results generally indicate that the various item parameter estimation procedures tend to yield similar results. The major exception concerned the Bayesian and maximum likelihood procedures (MLP) applied to the three parameter model. The MLP has a tendency to produce more extreme estimates than the Bayesian procedure. A second purpose is to compare the ability estimates produced by the available procedures: maximum likelihood, expected a'posteriori, and maximum a'posteriori. The results indicated that: (1) robustification is not a strong effect on the mean or standard deviation of the ability estimates; (2) the mean and variance of the ability estimates are not strongly effected by the type of item parameter estimates used in calculating ability estimates; and (3) the effect of ability estimation procedure is fairly strong on the ability estimates. (PN) |
| Erfasst von | ERIC (Education Resources Information Center), Washington, DC |
Literaturbeschaffung und Bestandsnachweise in Bibliotheken prüfen
Permalink als QR-Code
Inhalt auf sozialen Plattformen teilen (nur vorhanden, wenn Javascript eingeschaltet ist)
Teile diese Seite:
