In a classic experiment carried out from 1918 to 1934, growth of apple trees of six different rootstocks were compared on four measures of size.

`data(RootStock)`

A data frame with 48 observations on the following 5 variables.

`rootstock`

a factor with levels

`1`

`2`

`3`

`4`

`5`

`6`

`girth4`

a numeric vector: trunk girth at 4 years (mm x 100)

`ext4`

a numeric vector: extension growth at 4 years (m)

`girth15`

a numeric vector: trunk girth at 15 years (mm x 100)

`weight15`

a numeric vector: weight of tree above ground at 15 years (lb x 1000)

This is a balanced, one-way MANOVA design, with n=8 trees for each rootstock.

Rencher, A. C. (1995). *Methods of Multivariate Analysis*.
New York: Wiley, Table 6.2

# NOT RUN { data(RootStock) ## maybe str(RootStock) ; plot(RootStock) ... root.mod <- lm(cbind(girth4, ext4, girth15, weight15) ~ rootstock, data=RootStock) Anova(root.mod) pairs(root.mod) # test two orthogonal contrasts among the rootstocks hyp <- matrix(c(2,-1,-1,-1,-1,2, 1, 0,0,0,0,-1), 2, 6, byrow=TRUE) linearHypothesis(root.mod, hyp) heplot(root.mod, hypotheses=list(Contrasts=hyp, C1=hyp[1,], C2=hyp[2,])) heplot1d(root.mod, hypotheses=list(Contrasts=hyp, C1=hyp[1,], C2=hyp[2,])) # }