The Statistical Sleuth: A Course in Methods of Data Analysis

by Fred L. Ramsey, Daniel W. Schafer

STATISTICAL SLEUTH is an innovative treatment of general statistical methods, taking full advantage of the computer, both as a computational and an analytical tool. The material is independent of any specific software package. In "The American Statistician" (February 2000, Vol. 54, No. 1), George Cobb commented, "What is new and different about Ramsey and Schafer's book, what makes it a 'larger contribution,' is that it gives much more prominence to modeling and interpretation of the sort that goes beyond the routine patterns." His students did "substantially better" on term papers based on the analysis of data. In the book, the focus is on a serious analysis of real case studies; on strategies and tools of modern statistical data analysis; on the interplay of statistics and scientific learning; and on the communication of results. With interesting examples, real data, and a variety of exercise types (conceptual, computational, and data problems), the authors get students excited about statistics.

Genres: Textbooks, School, Reference

768 pages, Hardcover

First published January 1, 1990

Reviews

[Suzanne · Rating 1 out of 5]

Absolute garbage but seems to be the book of choice for beginning statistics courses for scientists. The authors' attempt to simplify statistics is abysmal (come on, a train analogy to explain PCA). They leave out the math, fundamental to understanding this subject, and replace it with pathetic analogies and examples. One gets a precursory, if that, understanding of the subject from reading the text, but the material is so slick in presentation and overly simplified that one must read it multiple times to get all out of it that's necessary to actually conduct an analysis. For a good general statistics book, buy Biometry, or if you want something in a similar style, buy Ellison's A Primer of Ecological Statistics (he actually seems to understand that people reading statistics texts aren't generally morons).

[Jerzy · Rating 4 out of 5]

I didn't read the whole thing, just skimmed it while looking for datasets to use in homework problems for my Experimental Design course. But it looks to be well-written, with good explanations and really nice use of extended case studies that tie different topics together. I might try it as a textbook next time.

[Gi · Rating 5 out of 5]

A very intuitive look into data analysis. Refreshing!

[Mike Martos · Rating 3 out of 5]

Difficult to read.