|
Activity Number:
|
36
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #304921 |
|
Title:
|
Calculating Standard Error with Parametric Bootstrapping
|
|
Author(s):
|
Anli Lin*+
|
|
Companies:
|
Pearson
|
|
Address:
|
4615 Briley Elm, San Antonio, TX, 78247,
|
|
Keywords:
|
Standard Errors ; Parametric Bootstrapping ; IRT Model
|
|
Abstract:
|
There are two major methods to calculate standard error: analytic and bootstrapping. With analytic method, either the formula needs to be found to calculate standard error or a derivation of the formula needs to be made. Major disadvantages are inaccurate and not consistent across different processes. In this study, parametric and non-parameter bootstrapping to calculate standard errors were evaluated. Following results are presented: a) relationship between standard errors and sample sizes; b) comparison of the methods between parametric bootstrapping and nonparametric bootstrapping; c) comparison of the conditional standard errors of measurement for rounding and un-rounding scores.
|
- 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 2009 program |
