It is seldom a good idea to use the _significance_ of a difference as a surrogate for its magnitude. The reason is that the significance varies with many irrelevant aspects of the analysis, such as the model and the sample size. If you have other measures of group similarity, then there are many hierarchical clustering methods that might meet your needs. Try ?hclust to get started.
Ben Fairbank -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of bob sandefur Sent: Tuesday, April 05, 2005 11:39 AM To: [email protected] Subject: [R] Dendrogram for a type unbalanced ANOVA Hi- I have about 20 groups for which I know the mean, variance, and number of points per group. Is here an R function where I can plot (3 group example) something like | |----- 2 | |-----------| | | |----- 1 | | |------| | |----------------- 3 | | 0 25 50 75 100 ie 1 and 2 are different at 75% level of confidence 1 2 combined are different from 3 at 25% level of confidence If this plot is a silly idea can someone point me to a reference for anything similar? Thanx Robert (Bob) L. Sandefur PE Senior Geostatistician / Reserve Analyst CAM 200 Union Suite G-13 Lakewood, Co 80228 [EMAIL PROTECTED] 303 472-3240 (cell) <-best choice 303 716-1617 ext 14 ______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html ______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
