[R] SNK-test

Jim Rogers jrogers at cantatapharm.com
Thu Sep 4 16:22:27 CEST 2003

Please note Student-Newman-Keuls is NOT a recommended multiple comparison procedure.

In the language of Hsu (1996), Student-Newman-Keuls is not even a "confident inequalities" method. In other words, it does not control the probability of making at least one incorrect assertion of inequality (which is the least you should expect of any multiple comparisons procedure). 

Tukey's HSD does protect the desired error rate, and gives you confidence intervals as well (not just assertions of inequality). 


Hsu (1996). Multiple Comparisons, Theory and Methods. Chapman Hall.

Jörg Peter Baresel wrote:
>  How can I perform a Student-Newman-Keuls-Test for multiple 
> comparision of means in R? (I did not manage to find any specific 
> function in the libraries)
> Jörg Peter Baresel
> Technische Universität München
> Institut für Ackerbau und INformatik im Pflanzenbau
> D-85354 Freising
> Lange Point 51

James A. Rogers, Ph.D. <rogers at cantatapharm.com>
Statistical Scientist
Cantata Pharmaceuticals
300 Technology Square, 5th floor
Cambridge, MA  02139
617.225.9009 x312
Fax 617.225.9010

More information about the R-help mailing list