17

I bought this book:

How to Measure Anything: Finding the Value of Intangibles in Business

and

Head First Data Analysis: A Learner's Guide to Big Numbers, Statistics, and Good Decisions

What other books would you recommend?

Justin Bozonier
  • 1,119
  • 2
  • 11
  • 24
  • 7
    Related: http://stats.stackexchange.com/questions/421/what-book-would-you-recommend-for-non-statistician – Shane Jul 26 '10 at 19:40
  • 1
    You tagged this as bayesian and machine-learning. What kind of data analysis are you interested in? – Shane Jul 26 '10 at 19:46
  • Shane: Honestly, I don't really know yet. What kind of data analysis is there? Should I be posing this as yet another question for the site? – Justin Bozonier Jul 26 '10 at 19:49
  • There are many kinds. :) Those are just two specific areas, so it seems odd for tags on such a general question. – Shane Jul 26 '10 at 19:54
  • @Justin If you don't know it yet, this question is too broad and vague. A nice thing to learn what a field contains, is to look on the tags of this sites and see what questions match with it. Also wikipedia and the books you bought can give you and idea of what you want (although the books maybe not name the field wherein they operate, more on the practical overall part) – Peter Smit Jul 26 '10 at 19:54
  • @Peter Smit: Thanks I'll give that a shot. @Shane: Thanks as well... I was trying to give some idea of context with my tags. Beware the noobs! – Justin Bozonier Jul 26 '10 at 20:10
  • @robin There are only 7 users able to vote for closing... so it can take a small while – Peter Smit Jul 26 '10 at 20:17
  • @robin @peter: It isn't clear to me that it's an exact duplicate either. Are you both referring to the one I linked above or another one? – Shane Jul 26 '10 at 21:23
  • @shane I agree that it is not exact duplicate, but then it is too broad or not enough and the title is quite similar. In addition, my view is that someone asking for a book on stat.stack could at least say for what purpose and for what level. – robin girard Jul 26 '10 at 21:29
  • @robin: I agree 100% with that. – Shane Jul 26 '10 at 21:33

12 Answers12

7

This book is dynamite: George E. P. Box, Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building

It starts from zero knowledge of Statistics but it doesn't insult the reader's intelligence. It's incredibly practical but with no loss of rigour; in fact, it underscores the danger of ignoring underlying assumptions (which are often false in real life) of common tests.

It's out of print but it's very easy to find a copy. Follow the link for a few options.

Carlos Accioly
  • 4,715
  • 4
  • 25
  • 28
  • Thank you, Carlos, for the recommendation. It is indeed a great read, despite (or perhaps because of) its age. I'm especially impressed that the authors (Box, Hunter, & Hunter) appeal to *permutation* distributions, rather than arguing for normality, as the "ultimate" justification for the classical tests (t, F, etc.). – whuber Sep 10 '10 at 19:50
  • Looks like an excellent book! Going to buy a copy right now. =) Thanks for this, Carlos. – Graeme Walsh May 27 '13 at 02:49
  • 2
    Book was first published in 1978, revised in 2005 and is currently in print. – Nick Cox May 27 '13 at 08:03
6

I didn't find How To Measure Anything, nor Head First, particularly good.

Statistics In Plain English (Urdan) is a good starter book.

Once you finish that, Multivariate Data Analysis (Joseph Hair et al.) is fantastic.

Good luck!

Neil McGuigan
  • 9,292
  • 13
  • 54
  • 62
5

I am a big fan of Statistical Models - Theory and Practice by David Friedman. It succeeds remarkably well to introduce and motivate the different concepts of statistical modeling through concrete, and historically important problems (cholera in London, Yule on the causes of poverty, Political repression in the McCarty era..).

Friedman illustrates the principles of modeling, and the pitfalls. In some sense, the discussion shows how to think about the critical issues and is honest about the connection between the statistical models and the real world phenomena.

Georgi
  • 101
  • 3
4

The classic "orange horror" remains an excellent introduction: Exploratory Data Analysis by John Tukey.

http://www.amazon.com/Exploratory-Data-Analysis-John-Tukey/dp/0201076160

Wesley Burr
  • 756
  • 3
  • 10
3

Statistics as Principled Argument by Abelson is a good side book to learning statistics, particularly if your substantive field is in the social sciences. It won't teach you how to do analysis, but it will teach you about statistical thinking.

I reviewed this book here

Peter Flom
  • 94,055
  • 35
  • 143
  • 276
3

The best intro in my eyes is the following one:

David Howell - Statistical Methods for Psychology

It is the BEST in making statistical concepts understandable for non mathematicians so that they get the math afterwards! Unfortunately it is updated every year and, hence, pricey.

Henrik
  • 13,314
  • 9
  • 63
  • 123
3

My favourite book on Statistics is is David William's Weighing the Odds. Davison's Statistical Models is good too.

seancarmody
  • 788
  • 6
  • 8
2

You might find useful this one: The Elements of Statistical Learning: Data Mining, Inference, and Prediction

UPDATE #1:

This book might be useful as well: O'Reilly: Statistics in a Nutshell

niko
  • 1,261
  • 3
  • 15
  • 18
1
  1. Wilcox, Rand R. - BASIC STATISTICS - Understanding Conventional Methods and Modern Insights, Oxford University Press, 2009
  2. Hoff, Peter D. - A First Course in Bayesian Statistical Methods, Springer, 2009

  3. Dalgaard, Peter - Introductory Statistics with R, Second Edition, Springer, 2008

also take a glance at this link, though it's R-specific, there are plenty of books that can guide you through basic statistical techniques.

aL3xa
  • 2,113
  • 3
  • 23
  • 27
1

As a biologist, I found the Sokal and Rohlf text to be quite readable, despite its voluminous-ness. It's not so great as a quick reference, but does walk one through statistical theory.

R. R. Sokal and F. J. Rohlf, Biometry the principles and practice of statistics in biological research, Third. (New York: W.J. Freeman and Company, 1995).

Duke of Lizards
  • 391
  • 1
  • 5
1

An old favourite of mine as an introduction to biostatistics is Armitage & Berry's (& now Matthew's):

Statistical Methods in Medical Research

Thylacoleo
  • 4,829
  • 5
  • 24
  • 32
1

Agresti & Finlay's Statistical Methods for the Social Sciences is quite good, though I'd like to believe there is a good open source alternative. Is it wrong to use an amazon affiliate link here?

Michael Bishop
  • 2,171
  • 3
  • 21
  • 31