Reference : A test-length correction to the estimation of extreme proficiency levels
Scientific journals : Article
Physical, chemical, mathematical & earth Sciences : Mathematics
A test-length correction to the estimation of extreme proficiency levels
[fr] Une correction de l'estimation des niveaux d'habileté extrêmes selon la longueur du test
Magis, David mailto [Université de Liège - ULiège > Département de mathématique > Statistique mathématique >]
Béland, Sébastien mailto [Université du Québec à Montréal > Education et pédagogie > > >]
Raîche, Gilles mailto [Université du Québec à Montréal > Education et pédagogie > > >]
Applied Psychological Measurement
SAGE Publications
Yes (verified by ORBi)
[en] Item response theory ; Rasch model ; Proficiency level
[en] In this paper, the estimation of extremely large or extremely small proficiency levels, given
the item parameters of a logistic item response model, is investigated. On the one hand, the
estimation of proficiency levels by maximum likelihood (ML), despite being asymptotically
unbiased, may yield infinite estimates. On the other hand, with an appropriate prior
distribution, the Bayesian approach of maximum a posteriori (MAP) yields finite estimates,
but it suffers from severe estimation bias at the extremes of the proficiency scale. In a first
step, we propose a simple correction to the MAP estimator in order to reduce this estimation
bias. The correction factor is determined through a simulation study and depends only on the
length of the test. In a second step, some additional simulations highlight that the corrected
estimator behaves like the ML estimator and outperforms the standard MAP method for
extremely small or extremely large abilities. Although based on the Rasch model, the method
could be adapted to other logistic item response models.

File(s) associated to this reference

Fulltext file(s):

Restricted access
APM-09-07-092-R3.pdffinal version accepted for publicationAuthor preprint1.14 MBRequest copy

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.