|
Activity Number:
|
374
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #309402 |
|
Title:
|
Determining Sample Sizes for Estimating and Comparing Item Parameters from the Rasch Model
|
|
Author(s):
|
Linda Young*+
|
|
Companies:
|
University of Florida
|
|
Address:
|
Dept of Statistics, Gainesville, FL, 32611-0339,
|
|
Keywords:
|
power ; sample size ; Rasch model
|
|
Abstract:
|
In the Rasch Model, the probability of a specified response (e.g. right vs. wrong) is modeled as a function of person and item parameters. With the VA, this model is frequently used to analyze results from functional status and quality of life surveys. Yet, sample size requirements for assessing an item's difficulty or for comparing the difficulty of two items are based on only rough "rules-of-thumb" that are often neither empirically nor theoretically based. The purpose of this study is to determine the sample sizes needed to estimate a parameter with the desired precision or to compare the difficulty of two items with specified power. The sample size requirements over a range of item difficulty and distribution of person parameters are determined. Sample sizes can be more precisely determined than the "rules-of-thumb" currently prevalent in the literature.
|
- 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 2007 program |
