Applied Statistics
Preface
This book is not for statisticians. It is for the large majority of other users of statistics. You need some overview of what methods there are to begin with. You probably don’t have time to learn all the basics first and build your knowledge from the ground up. That is why I want to take a different, hopefully more pragmatic approach in this book.
Motivation
Classical statistics textbooks typically start with some introductory math, then go from basic probability theory to the normal distribution and the central limit theorem, before moving on to univariate tests (\(t\)-test, \(\chi^2\)-test, \(F\)-test, non-parametric alternatives), and perhaps ending somewhere around ANOVA.1 This makes sense if you’re reading the book as an aspiring statistician—you need to understand each method at its core, and that requires some logical buildup of methodology. I did not make this book as a critique of this style of teaching. In fact, this is how I got into the field of statistics myself.
However, while teaching statistics to life scientists, it has always bothered me that we do not get to work with the interesting, actually useful models till near the end of the course. That is what I want to do different in this book. You need to spend most of the time during your course working with flexible models. From the get go, we will start thinking in probability distributions and try to decide which might be suitable for our problem. You may not understand what a generalized linear model really is in the beginning, but you will already be able to use it. How it actually works is what I want the course to end with, because much as mathematicians may lament it, it simply isn’t as relevant to the average user of statistics.
For example, openstax free Introductory Statistics.↩︎