Download How to Lie with Statistics by Darrell Huff PDF

By Darrell Huff

Darrell Huff runs the gamut of each popularly used form of statistic, probes things like the pattern research, the tabulation technique, the interview procedure, or the best way the implications are derived from the figures, and issues up the numerous variety of dodges that are used to idiot instead of tell.

Show description

Read or Download How to Lie with Statistics PDF

Similar logic books

Geomorphological Hazards of Europe

The Geomorphological risks of Europe comprises an outstanding stability of authoritative statements at the diversity and reasons of average dangers in Europe. Written in a transparent and unpretentious kind, it gets rid of myths and concentrates at the simple evidence. The ebook seems on the identified distributions, strategies and the underlying rules and specializes in the necessity for a real realizing of the clinical info in order that a true contribution to endanger administration might be made.

The Logic of the Plausible and Some of its Applications

So basic and imperfect because it might sound this e-book has made use of information on invention and discovery accumu­ lated in the course of a life-time. these folks who will be tempted to stress basically its imperfections should still learn the correspondence exchanged among Cantor and Dedekind on the finish of the 19th century; they might then detect how tricky it used to be, even for an excellent guy, the author of the set conception, to suggest impeccable ends up in a totally new box.

Incompleteness in the Land of Sets

Russell's paradox arises after we think of these units that don't belong to themselves. the gathering of such units can't represent a suite. Step again a section. Logical formulation outline units (in a regular model). formulation, being mathematical items, might be considered units themselves-mathematics reduces to set concept.

Extra resources for How to Lie with Statistics

Example text

You may get a fifty-fifty result, but probably you won't; your result, Uke mine, stands a good chance of being quite a ways away from fifty-fifty. ITI-ES a J ~ 4 ~j tain to come out with a result very close to half heads-a result. that is. which represents the real probability. Only when there is a substantial number of trials involved is the law of averages a useful description or prediction. How IllallY ill euough? That"s a tricky one too. It depends among other things on how large and how varied a population you are studying by sampling.

L: ~ ~~ --r~I~ RecDrd .. Rang. • • . '" V_ +:. 1. ~ , ~ ,\ ~ Recard LDws- -.. J f .. A• J J ASO.. --' CHAPTER 4 Much Ado about Practically Nothing IF YOU don't mind, we will begin by endowing you with two children. Peter and Linda (we might as well give them modish names while VI-"e're about it) have been given intelligence tests, as a great many children are in thecourse of their schooling. y be right at that. Anyway, you learn somehow that Peter's IQ is 98 and Linda's is 101. You know, of course, that the IQ is based on 100 as average or "nonnal," Aha.

I came upon another of those statements that collapse under a second look. It appeared in an advertisement by a group of electric companies in 1948. «Today, electric power is available to more than threequarters of U. S. farms... :' That sounds pretty good. Those power companies are really on the job. Of course, jf you wanted to be ornery you could paraphrase it into "Almost one-quarter of U. S. " and by using it the companies have HOW TO LIE WITH STATIS'11CS been able to say just about anything they please.

Download PDF sample

Rated 4.78 of 5 – based on 27 votes