Year: 1979
Type: report
Abstract: Some of the more popular multiple-comparisons procedures are discussed and compared. Some new nonparametric methods are introduced. One procedure is an analog to the Fisher's least-significant-difference method for the completely randomized design. Some simulation studies indicate this procedure is a reasonable nonparametric method to use. A summary description is given for other nonparametric methods, which may be used with the completely randomized or randomized blocks designs. 3 tables.
DOI: 10.2172/6057803
Authors: W Conover, R Iman
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W58593274
