In This Chapter
Comparing two samples
Comparing more than two samples
Testing relations between categorical variables
Working with models
It’s one thing to describe your data and plot a few graphs, but if you want to draw conclusions from these graphs, people expect a bit more proof. This is where the data analysts chime in and start pouring p-values generously over reports and papers. These p-values summarize the conclusions of statistical tests, basically indicating how likely it is that the result you see is purely due to chance. The story is a bit more complex — but for that you need to take a look at a statistics handbook like Statistics For Dummies, 2nd Edition, by Deborah J. Rumsey, PhD (Wiley).
R really shines when you need some serious statistical number crunching. Statistics is the alpha and omega of this language, but why have we waited until Chapter 15 to cover some of that? There are two very good reasons ...