March 2019
Beginner to intermediate
448 pages
13h 14m
English
In general, once we have determined that a factor is significative (at least one of its levels has a mean that's different from the rest), we want to plot the differences between the levels, along with their 95% confidence intervals. This can be achieved via the Tukey Honest Differences function (TukeyHSD in R), as follows:
rbind(TukeyHSD(d)$Colour,TukeyHSD(d)$Font)
The preceding code generates the following output:

This prints all of the mean differences between each pair of factors, along with the lower and upper 95% confidence intervals (this can be changed via the conf.level parameter). Finally, the p-value is printed ...
Read now
Unlock full access