Abstract Details
Activity Number:
|
68
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
|
Sponsor:
|
ENAR
|
Abstract #313794
|
View Presentation
|
Title:
|
Dimensionality Assessment for Polytomous Item Instrument
|
Author(s):
|
Tan Li*+
|
Companies:
|
Florida International University
|
Keywords:
|
Dimensionality ;
IRT ;
Psychometrics ;
Self-report assessment
|
Abstract:
|
In the field of public health, many research areas, such as HIV preventions, health care decisions, and clinical trials, frequently depend on individual self reported assessments. Item response theory (IRT) modelling offers a powerful tool for evaluating instruments both at the item and scale level. Unidimensionality is the main assumption underlying the widely used IRT models. It assumes that the psychometric instrument only measures a single latent trait. Practically, the violation indicates that it may be biased to report a single score for an instrument. The proposed conditional-covariance-based nonparametric dimensionality assessment method in this paper, Poly-NEWDIM (PND), in conjunct with the proposed conditional covariance based subset selection method, HCPND, demonstrated closer to nominal Type I error control and better power than previous methods in the previous simulation studies. The purpose of this paper is to evaluate the proposed method under more complex data structures, such as fan structures, and more dimensions (3 or 4), as well as a real measurement in public health area.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.