Decision Accuracy In IRT Models
Door: Verstralen, H.H.F.M. Verhelst, N.D. | 07-01-1991 A disadvantage of the application of the Rasch Model (RM) or the Partial Credit Model (PCM) is that one may be forced to omit the best discriminating items from an item bank to attain acceptable model fit.A solution to this problem is offered by the family of One Parameter Logistic Models (OPLM). OPLM opens the possibility to model differently discriminating items on one latent scale, without sacrificing sufficient statistics and conditional maximum likelihood estimation. The preservation of these valuable properties of the RM
and the PCM is achieved by avoiding to estimate discrimination indices by treating them as known integer constants. Although a dedicated least squares algorithm helps the user to quickly find appropriate values for the discrimination indices, he is in principle burdened with the responsibility for them.
Kunnen we je helpen?
Stel je vraag via onze kanalen of kijk in de veelgestelde vragen.
Voor scholen: Vergeet niet om het brinnummer bij de hand te hebben en/of in de mail te vermelden, zodat we jouw vraag sneller kunnen behandelen!