JSM 2004 - Toronto

Abstract #300346

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

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 2004 Program page



Activity Number: 187
Type: Contributed
Date/Time: Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Computing
Abstract - #300346
Title: Bayesian Multivariate Item Response Theory Model and Its Software BMIRT
Author(s): Lihua Yao*+
Companies: CTB McGraw-Hill
Address: 20 Ryan Ranch Rd., Monterey, CA, 93940,
Keywords: item response theory ; Bayesian ; MCMC ; multidimension ; multigroup
Abstract:

We propose a multivariate item response theory model for educational testing data (MIRT). This model is a generalization of the standard item response theory model. We use a Bayesian approach and MCMC for parameter estimation of this model. The estimation procedure is implemented in a computer program called BMIRT (Bayesian Multivariate Item Response Theory model). BMIRT has several important properties. First, the guessing parameters are estimated based on the data, not fixed in an ad hoc manner prior to fitting the model as in some existing software such as TESTFACT, and MICROFACT. Second, by combining different content areas for concurrent calibration, BMIRT provides more accurate objective scores. Third, BMIRT can determine the dimension loadings of the test items that are in different categories but are correlated, and as a result, more accurate parameter estimation and score reporting. Finally, with BMIRT, multigroup tests can be put on the same scale so that guidance and methods to measure longitudinal growth at the student level can be provided. We use real and simulated data sets to illustrate the applications of the model and BMIRT.


  • 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 2004 program

JSM 2004 For information, contact jsm@amstat.org or phone (888) 231-3473. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2004