One way ANOVA: post-hoc comparison “adding letters to graph”

Question

I am working on a project for one of my classes and cannot figure out how to obtain the letters for treatment comparisons after using the tukeyHSD post-hoc test.

Is there a simple way to do this?

Here is my code:

Boulder study

rm(list=ls())
getwd()
setwd('/Volumes/9163980664/GIS/Boulder_Field/CSV_files')
list.files()
bodega <- read.table('biotic_stats.csv', sep=',', h=T)
head(bodega[,1:5])
length(bodega$Face[bodega$Face=='E'])
length(bodega$Face[bodega$Face=='N'])
length(bodega$Face[bodega$Face=='S'])
length(bodega$Face[bodega$Face=='W'])

One way ANOVA

boulder_face <- bodega$Face
proportion_rockweed <- bodega$Pelvetiopsis_prop_cover_

Barplot

setwd('/Volumes/9163980664/GIS/Boulder_Field/R_code_graphs_stats')
ylab = 'Rockweed Proportion'
xlab = 'Face of Boulder'
graphics.off(); quartz(width=4, height=4); par(mai=c(1.2,1.2,.3,.5), cex=1.4)
plot(proportion_rockweed ~ boulder_face, data=bodega, col="grey", ylab=ylab, xlab=xlab)
dev.print(png, file = "Bodega_Rockweed_boxplot.png", wid = 480, height = 480)

ANOVA

bodega.anova <- aov(proportion_rockweed ~ boulder_face, data = bodega)
summary(bodega.anova)
TukeyHSD(bodega.anova)
help(aov)

here is what i have tried

TukeyHSD(bodega.anova)
quartz(width=4, height=4); par(mai=c(1.2,1.2,1,1), cex=1)
plot(TukeyHSD(bodega.aov))
bodega.anova <- aov(proportion_rockweed ~ boulder_face, data = bodega)

library(multcompView)
multcompLetters(extract_p(TukeyHSD(bodega.anova)))

Answer 1

I just saw this. You can use the lsmeans package and get them easily

library(lsmeans)
cld(bodega.anova, "boulder_face")

This requires that the multcompView package be installed. The Tukey method is the default, but you can specify other adjustments using an adjust argument.