Not the best way to learn, but entertaining. ( )

I found this book extremely useful as I helped a friend through an advanced statistics class. I am a very visual learner, and tend to draw concepts to help me understand, but having the drawings done and ready to share during my tutoring session saved time. I think that there was actually a second advantage, as the cartoons were tried and I didn't have to spend time thinking of good examples. ( )

This is the statistics reference book I keep with me at work. It's still the best, simplest guide I've ever found. ( )

The book has simple cartoon with basic explanation of statistics concepts. The book didn't teach me anything new, but helped me refresh/reinforce my understanding. I have to admit that this is not for absolute beginners, but if you are willing to to spend some time amusing yourself with stats, you'll enjoy this book. ( )

This was my introduction to Gonick's "comic" textbooks. As a stats text, it can claim something others can't: it's entertaining. This trait doesn't interfere with the book's ability to teach complicated topics; in fact, the humor allows it to do so with ease. The gags, illustrations, and characters help to break up the hard ideas so they're easier to understand. Many texts jump into the equations without doing enough to explain the basics plainly, but this book invites everyone to understand.

The best parts of this book are the chapters on distributions, sampling, and confidence intervals. The visualization of the binomial distribution being squished by DeMoivre into the normal distribution not only provides intuition on the normal approximation, but it leaves a vivid image of the relationship of the distributions. The part on sampling will imprint the 1/sqrt(n) factor in your mind and verse you in sampling issues and strategies. Finally, the analogy of firing an arrow into a target will make you see confidence intervals in an intuitive light.

My only criticisms are that I wish this book had a sequel that dives into the additional topics presented in the conclusion. Resampling, multivariate analysis, and principle components would be good candidates to bring down to Earth. Also, a chapter on Chi square statistics would have been great to round out the development of proportion tests. Anyone interested in learning how statistics is used should read this book. It won't make you an expert or provide you with problem sets, but the concepts will shine through. ( )