Numerical Validation of Tukey's Criterion for Clinical Trials and Sequential Testing
Abstract
A basic problem in conducting either clinical or sequential trials is to determine which or when statistical significance for a predetermined level of alpha has occurred. The consequences of not using Tukey's criterion for k nonoverlapping comparisons are developed. The use of this criterion might be too stringent, however, and an alternative statistic is given.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1984
- Accession Number
- ADP003853
Entities
People
- C. R. Leake
Organizations
- Center for Army Analysis