A very good book. I chose it because it seemed to be complete (and French!). In addition, R is a free statistical software, this book helps to understand how to choose his statistical tests while learning to use the software R. For those who find it tedious, it is also possible to perform the proposed scripts directly in the book regardless of why or how.
The only downside I could find in this book for non-parametric statistical tests. Best known as the Kruskal Wallis ANOVA with or without correction are there, but I was very surprised, for example, does not find the Friedman test ...? This is still a non-parametric test used enough so it's a shame.