Discovering Statistics Using R

Book Preface

HOW TO USE THIS BOOK

When the publishers asked me to write a section on ‘How to use this book’ it was obviously tempting to write ‘Buy a large bottle of Olay anti-wrinkle cream (which you’ll need to fend off the effects of ageing while you read), find a comfy chair, sit down, fold back the front cover, begin reading and stop when you reach the back cover.’ However, I think they wanted something more useful. 

What background knowledge do I need?

In essence, I assume you know nothing about statistics, but I do assume you have some very basic grasp of computers (I won’t be telling you how to switch them on, for example) and maths (although I have included a quick revision of some very basic concepts so I really don’t assume anything).

Do the chapters get more difficult as I go through the book?

In a sense they do (Chapter 16 on MANOVA is more difficult than Chapter 1), but in other ways they don’t (Chapter 15 on non-parametric statistics is arguably less complex than Chapter 14, and Chapter 9 on the t-test is definitely less complex than Chapter 8 on logistic regression). Why have I done this? Well, I’ve ordered the chapters to make statistical sense (to me, at least). Many books teach different tests in isolation and never really give you a grip of the similarities between them; this, I think, creates an unnecessary mystery. Most of the tests in this book are the same thing expressed in slightly different ways. So, I wanted the book to tell this story. To do this I have to do certain things such as explain regression fairly early on because it’s the foundation on which nearly everything else is built.

However, to help you through I’ve coded each section with an icon. These icons are designed to give you an idea of the difficulty of the section. It doesn’t necessarily mean you can skip the sections (but see Smart Alex in the next section), but it will let you know whether a section is at about your level, or whether it’s going to push you. I’ve based the icons on my own teaching so they may not be entirely accurate for everyone (especially as systems vary in different countries!):

1 This means ‘level 1’ and I equate this to first-year undergraduate in the UK. These are sections that everyone should be able to understand.
2 This is the next level and I equate this to second-year undergraduate in the UK. These are topics that I teach my second years and so anyone with a bit of background in statistics should be able to get to grips with them. However, some of these sections will be quite challenging even for second years. These are intermediate sections.

3 This is ‘level 3’ and represents difficult topics. I’d expect third-year (final-year) UK undergraduates and recent postgraduate students to be able to tackle these sections.
4 This is the highest level and represents very difficult topics. I would expect these sections to be very challenging to undergraduates and recent postgraduates, but postgraduates with a reasonable background in research methods shouldn’t find them too much of a problem.