How to Lie with Statistics: Difference between revisions

Revision as of 02:55, 22 July 2007

Revision as of 04:02, 4 August 2007

Line 2:

'''''How to Lie with Statistics''''' is [[Darrell Huff]]'s perennially [[best-selling]]<ref> "Over the last fifty years, How to Lie with Statistics has sold more copies than any other statistical text." J.M. Steele. "[http://www-stat.wharton.upenn.edu/~steele/Publications/PDF/TN148.pdf Darrell Huff and Fifty Years of ''How to Lie with Statistics'']. ''Statistical Science'', 20 (3), 2005, 205–209.</ref>

introduction to [[statistics]] for the general reader. Written in [[1954]], it is a brief, breezy, illustrated volume which explainsoutlining the common errors, thatboth canintentional beand doneunintentional, whenassociated describingwith the resultsinterpretation of statistical research, both intentionally and unintentionally,statistics and how these errors lead to a biased or inaccurate conclusionconclusions.

Over time is has become the most widely read statistics book in history and over one-half million copies have been sold in the English language edition alone. In [[2003]] the Department of Economics of [[Shanghai University]] published an edition in [[Chinese language|Chinese]].

The most recent edition — [http://htlws.it '''Mentire con le statistiche'''] – has beenwas published in Italian onin June 2007.

Some themes of the book are "[[Correlation does not imply causation]]" and "Using [[Random Sampling]]". It also shows how statistical graphs can be used to distort reality: