|
Activity Number:
|
154
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #309753 |
|
Title:
|
Repeated Classification as a Cost-Effective Sample Design To Test Association When There Are Random Misclassification Errors
|
|
Author(s):
|
Nathan Tintle*+
|
|
Companies:
|
Hope College
|
|
Address:
|
27 Graves Place, Holland, MI, 49423,
|
|
Keywords:
|
misclassification error ; genotyping error ; replicate sample ; repeat sample ; experimental design
|
|
Abstract:
|
Misclassification errors are well-known to cause significant bias and loss of power in related hypothesis tests. While random misclassification errors have received a fair bit of attention in the literature, repeated classification to address these errors has not been as fully explored. Recently, a probabilistic model for repeated classification was proposed, along with a method for incorporating repeated classification data into a traditional chi-squared test of association between two categorical variables. We present a method for comparing a traditional sampling design (all units classified once) with repeated classification (some portion of the sample is classified multiple times) assuming a fixed budget for sampling and classification costs. We find that repeated classification can be cost-effective (i.e. have increased power) to test association in some 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 2007 program |
