Activity Number:
|
346
- Advances in Diagnostics and Reproducibility Research
|
Type:
|
Topic-Contributed
|
Date/Time:
|
Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
|
Sponsor:
|
Social Statistics Section
|
Abstract #317196
|
|
Title:
|
Rank Correlation Inferences for Clustered Data with Small Sample Size
|
Author(s):
|
Sally Hunsberger* and Joanna Shih
|
Companies:
|
NIAID and National Cancer Institute
|
Keywords:
|
Chi square test;
Kendall's tau;
Spearman rank correlation;
U-statistic;
Within-cluster resampling
|
Abstract:
|
This paper develops methods to test for associations with clustered data using a U-Statistic approach with a second order approximation to the variance of the parameter estimate for the test statistic. The method applies to data where clusters have the same number of measurements for each variable or where one of the variables may be measured once per cluster while the other variable may be measured multiple times. Shih et. al. (2017) use the U-Statistic approach but only consider a first order approximation. The first order approximation has inflated significance level in scenarios with small sample sizes. We derive the test statistics using the second order approximations aiming to improve the type I error rates. We evaluate the performance of the test statistics through simulation with small sample sizes. The tests that are presented are for clustered versions of: the Spearman rank correlation and Kendall's tau for continuous data or ordinal data and for alternative measures of Kendall's tau that allow for ties in the data. We also develop a clustered version of Pearsons chi^2 test for data with two or more categories. The methods are all available in an R package.
|
Authors who are presenting talks have a * after their name.