This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/69694
Description
Title
Identification of Best Modeled Test Items
Author(s)
Davey, Timothy C.
Issue Date
1987
Department of Study
Psychology
Discipline
Psychology
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Psychology, Psychometrics
Abstract
Latent trait models have tremendous power for solving practical measurement problems. This power is realized, however, only when the models adequately characterize observations. Accordingly, a large number of procedures have been proposed to evaluate model goodness of fit. None of these procedures, though, is entirely satisfactory. Some are theoretically or methodologically flawed, others are limited in power or scope. A new approach to the question of model fit was therefore developed. This approach differs from existing procedures in several respects, and thus avoids many common pitfalls. For example, while most available methods are designed to recognize misspecified item response functions, test multidimensionality, or unmodeled dependencies among items, the new approach is designed to be sensitive to all of these. Further, whereas most methods attempt to identify and eliminate poorly modeled items, the new approach attempts to discover subsets of items that are well modeled. The advantage is that while available methods will eventually reject every item as sample size increases, the well modeled subset identified by the new approach should remain relatively invariant.
The performance of this new procedure was assessed through its application to both simulated and real data sets. A high level of sensitivity to a variety of modeling errors was demonstrated.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
