Try TukeyHSD(anova). TukeyHSD is implemented in the base package, i.e., ready to use with the base installation of R. Also, the TukeyHSD for the interaction term should have a colon ":" not a "*". The "which" argument in TukeyHSD() must be identical to the name of the coefficient in the summary table.

See the example below:

Thesis<-sample(c(1:3),300,replace=T)

Days<-sampe(c(1:2),300,replace=T)

Days<-sample(c(1:2),300,replace=T)

e=rnorm(300)

y<-Thesis+Days+Thesis*Days+e

my.anova<-aov(y~factor(Thesis)*factor(Days))

summary(my.anova)

TukeyHSD(my.anova)

TukeyHSD(my.anova,which="factor(Thesis)*factor(Days)") #fails

TukeyHSD(my.anova,which="factor(Thesis):factor(Days)") #works

HTH,

Daniel

Roberto wrote

Hi,

I need to determinate HSD value from a matrix like that

Thesis Days A1

Cx 0 66.07

Cx 0 60.24

Cx 0 42.86

Tw 0 66.07

Tw 0 60.24

Tw 0 42.86

Aa 0 66.07

Aa 0 60.24

Aa 0 42.86

Qe 0 66.07

Qe 0 60.24

Qe 0 42.86

Cx 56 310.86

Cx 56 223.17

Cx 56 186.77

Tw 56 149.42

Tw 56 127.75

Tw 56 138.59

Aa 56 130.24

Aa 56 214.83

Aa 56 137.95

Qe 56 186.64

Qe 56 189.09

Qe 56 187.87

this is my script

require(agricolae)

file <- "A1.txt"

data_table <- read.table(file, header=T, row.names=NULL)

attach(data_table)

anova <- aov(A1 ~ factor(Thesis)*factor(Days))

detach(data_table)

first of all, I need to know if my script has conceptual errors

secondly, I need to calculate HSD between Thesis:Days means

I try with Tukey.test command from agricolae package, but it do not work

I can use only these commands

Tukey.test(anova, "Thesis")

Tukey.test(anova, "Days")

but not

Tukey.test(anova, "Thesis*Days")

how can I do?

my apologies for the banal questions, but I'm taking the first steps in statistical computing

Thanks,

Roberto