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/2142/23452
Description
Title
Some issues in item response theory
Author(s)
Wu, Hongsheng
Issue Date
1996
Doctoral Committee Chair(s)
Stout, William F.
Department of Study
Statistics
Discipline
Statistics
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Statistics
Language
eng
Abstract
"The issues studied in this thesis are: the assumption of local independence (LI) in item response theory (IRT) models in the first three chapters, and differential item functioning (DIF) in IRT in the last chapter. A new term ""conditional trivariance"" for three items is defined, which is a measurement of joint local dependence among item triples. This is extended to any J items. Relevant theorems are stated and proved, which provide necessary and sufficient conditions involving conditional co-variances, trivariances, etc. for local independence to hold. These theorems lead us to a method to test LI that goes beyond conditional covariance exploration. We realize this procedure by using a kernel smoothing technique. Asymptotic normality of the test statistic is given and proved, which in turn justifies our procedure. Simulation studies show the procedure works well for the case of large examinee sample sizes. Some Graduate Record Examination (GRE) verbal test data is analyzed."
Chapter 4 deals DIF. A two-stage kernel smoothed DIF detecting method is proposed. Two-stage method is obtained by the amalgamation both Douglas, J., Stout, W., & DiBello, L.'s kernel smoothed DIF detection method and Shealy & Stout's linear regression correction used in SIBTEST. It is compared with Douglas, et al.'s one-stage method. The factors influencing the performance of the procedures are studied. A best combination of these factors is recommended.
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.