Textbooks on Statistics ?

January 18, 2006 7:35 AM

Introductory college-level textbook of Statistics ? I've got an half decent understanding of calculus I and some practice studying functions (stationery points, flexes, limits etc) , calculating finite and indefinite integrals and some background theory.

Browsing trought older questions I've found this course but I would like to have more choices. I don't mind 1000-pages long books (on the contrary I like them) so as long as it's well written and is not just a pile of notions piled to make the book look "Important"

posted by elpapacito to education (9 comments total)

I can highly recommend Introduction to the Practice of Statistics. It's well-written, has lots of examples, and explains the practice of the theory in both software packages (SPSS, Minitab, Excel, etc.) and by hand through formulas. I just sold my copy on half.com, or else I'd offer a Mefite discount to you.
posted by charmston at 7:58 AM on January 18, 2006

I should also note that the class I took this for was entitled "Statistics 1", was a 200-level class, and has the following course catalog description:

"Introductory applied statistics: sampling, descriptive (exploratory) statistics, probability, sampling distributions, estimation and hypothesis testing, non-parametrics, simple and multiple linear regression, introduction to analysis of variance; use of statistical packages."
posted by charmston at 8:04 AM on January 18, 2006

I took a Elementary Probability & Statistics course a couple of years ago, and I still have this book: Statistics, 3rd Ed (David Freedman).

Here's the course description: An introductory course to serve students of any field who want to apply statistical inference. Descriptive statistics, elementary probability, estimation, hypothesis testing and small samples.

I'd certainly be willing to sell the book if you're interested. It's big and thick, but I thought it was well written, and I did great in the class (A), even though I'm bad at the maths. My email is in my profile.
posted by misterioso at 10:04 AM on January 18, 2006

I've been on both ends of Wonnacott and Wonnacott's Introductory Statistics. It's a decent enough book, with (I think) good examples, a fair number of worked-through problems, and a good, reasonable attitude about not treating some things (eg, 5% significance) as holy writ.

I would avoid a book that relies on a software approach, especially if it relies on proprietary and costly (SPSS), little-used (Minitab), or often-inappropriate (Excel) software. Better to learn the concepts and then apply them to whatever software you want. If you really want to go that way, there are several books that link to R (or S, R is a free workalike of S), which would be far less limiting in the medium or long term.
posted by ROU_Xenophobe at 10:42 AM on January 18, 2006

I second misterioso's suggestion of the Freedman, Pisani, and Purves 'Statistics' (the third edition is still the latest, I think). I'm currently taking an intro to statistics course, and the book is much more informative than the lecturer. Also, based on a quick google, it seems to be one of the most popular intro stats texts out there.
posted by penchant at 10:56 AM on January 18, 2006

I was pleased with Mathematical Statistics with Applications. We used it for Statistical Models and Methods I, which has this description:

This course presents basic ideas of probability theory and statistics, and is recommended for students throughout the natural and social sciences who want a broad background in statistical methodology and exposure to probability models and the statistical concepts underlying the methodology. Probability is developed for the purpose of modeling outcomes of random phenomena. Random variables and their expectations are studied; including means and variances of linear combinations, and an introduction to conditional expectation. Binomial, Poisson, normal and other standard probability distributions are considered. Some probability models are studied mathematically and others via simulation on a computer. Sampling distributions and related statistical methods are explored mathematically, studied via simulation and illustrated on data. Statistical methods for describing data and making inferences based on samples from populations are presented. Methods include, but are not limited to, inference for means and variances for one- and two-sample problems, correlation and simple linear regression. Graphical and numerical data description are used for exploration, communication of results, and comparing mathematical consequences of probability models and data. Mathematics is employed to the level of univariate calculus and is less demanding than that required by STAT 24400.

The back of the book reads, "the most widely used mathematical statistics text at the top 200 schools in the United States," if that means anything to you.

In subsequent years I've found that I've had to consult other more specialized texts in order to solve the problems that have come up at work, but this definitely does a good job with the basics.
posted by epugachev at 10:57 AM on January 18, 2006

Big ol' shoutout for Elementary Statistics, by Spence, Cotton, Underwood, and Duncan. It's clean and simple, and doesn't waste a load of time discussing proofs. Going to lecture after reading seemed like a waste of time, it was that good. I loved this book.
posted by moira at 12:18 PM on January 18, 2006

Rou_Xeno: You figured me out, I prefer some good generalization that I can later apply to any package then a push-button-understand-nothing approach.

there are several books that link to R (or S, R is a free workalike of S)

I lost you here :o ! R, S ?

Thanks to everybody for their recommendations so far !
posted by elpapacito at 1:29 PM on January 18, 2006

I lost you here :o ! R, S ?

S is a statistics data analysis language that is now sold as S-PLUS. Really expensive, IIRC.

R is the GNU clone of S. Very powerful and it seems to have an active user committee that has extended the language with a good many packages.
posted by epugachev at 10:05 AM on January 19, 2006