Rank Tests in Weather Modification Experiments.
Abstract
Rank tests are frequently used in the comparison of seeded and unseeded responses or of control and target responses in weather modification experiments. One of the more frequently used tests is the sum of squared ranks test, which under certain questionable assumptions, has been shown to have a higher asymptotic efficiency relative to the more conventional Wilcoxon test, that is the sum of (first power) ranks test. This paper investigates the pertinence to reality of the underlying assumption in the analysis of weather modification experiments by rank methods. It is shown that the squared rank test is not uniformly more efficient than the Wilcoxon test.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 1977
- Accession Number
- ADA042051
Entities
People
- Morgan A. Hanson
Organizations
- Florida State University