Several people have offered here positive comments about Rossman's book. I canceled an order and ordered Rossman at the last moment in August, because I had heard so many positive things about it.
I have always disliked four-pound, 800-page, comprehensive, enclyclopedic textbooks, for several reasons. But the principal one probably is that I prefer books that provides only a strong "skeleton," so I can add the flesh myself. For those who feel that way, Rossman seems to be excellent.
A week or so ago I started Topic 11. As soon as I saw the word "Random" in the title, I knew I wanted to use the Random Rectangles activity, which derives, I think, from the Activity-Based Statistics project. (I want students to understand that they cannot expect themselves to function like a random manchine.)
On the next page, in Activity 11-1, as soon as I read the paragraph about the call-in survey, I knew I wanted to use Moore & McCabe's summary in IPS (pp.251-252) and some exercises from the UCLA Statistics web page, about nonresponse bias; which led to the matter of response bias using a recent newspaper article which supplied evidence that what people SAY they do (regarding washing their hands after using a public restroom) and what they DO are not the same thing.
The next paragraph in Rossman, about the Literary Digest poll in 1936 led to selection bias, and the detailed discussion of this poll in FPPA (Freedman, Pisani, Purves, and Adhikari).
On the next page, Rossman's discussion of the inadequacies of "physical mixing" led to a re-examination of the 1970 draft lottery data, which we had looked at before, and was biased--apparently because of inadequate physical mixing.
The point is that one can admire a book for what it _doesn't_ include, too. Rossman probably could have written a larger book, but it would not have been better; he has left it rich in starting points that lead to other neat places.
============================================== Bruce King Department of Mathematics and Computer Science Western Connecticut State University 181 White Street Danbury, CT 06810 (email@example.com)