Abstract #300964

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2002 Program page
JSM 2002 Abstract #300964
Activity Number: 50
Type: Contributed
Date/Time: Sunday, August 11, 2002 : 4:00 PM to 5:50 PM
Sponsor: Section on Statistical Education*
Abstract - #300964
Title: DIF Detection and Estimation Using Measurement Error Models
Author(s): Xin Feng*+ and Zhiliang Ying
Affiliation(s): Columbia University and Columbia University
Address: 618 Math Building, New York, New York, 10027, US
Keywords: Differential item functioning ; compterized adaptive testing ; IRT model ; Rasch model ; logistic regression
Abstract:

When calibrating the item parameters under a specific IRT model, we can use the (online) estimated latent traits as covariates and fit a nonlinear regression model. Because of the use of the estimated, not the true ability, the regression fit has to adjust for the bias caused by the covariate "measurement error." We propose two bias-correction procedures; one is under normal error assumption, which is not unreasonable in testing, and exploits the sufficiency; and the other only requires the second moment conditions of errors. After adjusting for the bias, Wald test approaches its nominal level and can be used for DIF detection and estimation. A Monte Carlo study for both Rasch model and 2PL model illustrates the superiority of the proposals in certain situations.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2002 program

JSM 2002

For information, contact meetings@amstat.org or phone (703) 684-1221.

If you have questions about the Continuing Education program, please contact the Education Department.

Revised March 2002