(Full disclosure - this review is part of the Books for Bloggers program)
Think Bayes, by Allen B. Downey, is a great hands-on introduction to using Python in the world of statistical analysis. With plenty of examples and links to source code, if you have a computer, some knowledge of statistics, and how to install Python & packages, and a desire to write code for statistical analysis, this is the book for you.
However, if you're like me and have heard about this thing called Bayesian filtering, and you think that this book will teach you all you need to know, then think again. You certainly don't need an advanced statistics degree, but if the terms PMF and CDF mean nothing to you, then you'll find yourself struggling with the concepts presented in this book. You can easily follow the examples, but there are no attempts at hand holding or explaining other statistics concepts. However, if you are interested in statistics and Bayesian statistics, then this book will provide all the terms that you need to research and really grok. Plus the source is available, so you can easily dig in there and see what's going on and researching further.
I think that overall this book accomplishes what it sets out to do. If you have a decent knowledge of statistics, and you want to quickly be able to write code to crunch your numbers, pick up this book. But if your statistics knowledge is a bit rusty and you were hoping for a refresher, this book may disappoint.
Even so, I'd give this book a 5/5.