Teaching Statistics: A Bag of Tricks

by Andrew Gelman, Deborah Nolan

This volume takes a positive spin on the field of statistics. Statistics is seen by students as difficult and boring, however, the authors of this book have eliminated that theory. Teaching A Bag Of Tricks , brings together a complete set of examples, demonstrations and projects that not only will increase class participation but will help to eliminate any negative feelings toward the area of statistics.

Genres: Education, Mathematics, Teaching, Nonfiction

320 pages, Paperback

First published January 1, 2002

Reviews

[Judy · Rating 5 out of 5]

I'm almost done reading this book. It is exceptional. It's not just informative, but the author's style and enthusiasm makes it very easy to read (I found it hard to put down). I am, of course, very interested in the topic, but there are so few books out there on this subject, and to find a really good one is amazing. This book is full of creative ideas for how to teach specific topics, how to engage the students, various class exercises and projects, etc. A great resource, and an excellent example of how a book like this should be written. We need more like this!

[Michael Emond · Rating 5 out of 5]

I am very impressed with this book. I have been teaching statistics for many years at a University level and felt the need to shake up my class a bit and this book was filled to the brim with in class participation demonstrations and assignment ideas. It is very readable and the demonstrations are all creative and well designed. You can tell this book was written by people who have DONE these demonstrations and fine tuned them and have feedback suggestions for you so you can avoid the pitfalls they encountered when first trying them out. There are multiple demonstrations for almost every aspect of an Introductory Statistics class.

So my take home message is - this is a must read book for anyone starting to teach statistics or who wants to try something new to spark your class's interest.

But I will comment on certain (personal) things that weren't as strong:
There are sometimes newspaper articles or maps that need to be used for a demonstration, it would be helpful to have a web-link to them or at least have better reproductions in the book so I could photocopy them. Most of them are so badly photocopied in the book they can't be used.
There are certain topics covered in this book I don't cover - that's not a big deal but worth noting. It was interesting to see how much they focused on logarithmic transformations, which is a topic I have never covered.
I cover hypothesis testing and consider it the main focus for the second half of my course but it is given a lot less time in this book - again, not a problem but worth noting.
These demonstrations are (for the most part) designed for SMALL classes (ideally 30 but up to 50) and my classes are 100+ so a lot of demonstrations would not work.
Some of the demonstrations are now outdated - they seemed to have been designed in the 90's but now many wouldn't work (their one that relies on phone books comes to mind).
At the end they do a very detailed run down of their 26 lecture plan. This is admirable but looking at their time allotment blew my mind. They run up 4 full class-participation demonstrations (plus lecturing) in a 75 minute class. Good for them but impossible for most of us.

These critiques do not take away from the fact that this book is an amazing resource for any teacher of intro statistics.

[Dio Mavroyannis · Rating 2 out of 5]

I was intrigued by the topic, I have read a few articles in education journals and I had learned very funky ways to present statistics concepts, I hoped I would get more of those kinds of tricks. Instead, the book went through too many concepts, too quickly. Though it seems to be fairly complete with going through the majority of things you would go through it is NOT thorough at all. Too much focus on completeness and not enough on original tricks.

[Jerzy · Rating 4 out of 5]

I haven't read the whole thing, just looked through for exercises I could do with my Experimental Design students. The exercises here didn't help me, since they focus on intro-to-stats course topics that were review for my students... but whenever I do teach stats 101, I will return to the ideas in this book.

I particularly liked the "experiment that looks like a survey" (which also illustrates anchoring bias).