How To Lie With Statistics

Introduces the reader to the niceties of samples (random or stratified random), averages (mean, median or modal), errors (probable, standard or unintentional), graphs, indexes, and other tools of democratic persuasion.

'A great introduction to a crucial topic' Bill Gates

'Perhaps the most popular book on statistics ever published ... It's a marvel ... gave me a peek behind the curtain of statistical manipulation, showing me how the swindling was done so that I would not be fooled again' Tim Harford

In 1954, Darrell Huff decided enough was enough. Fed up with politicians, advertisers and journalists using statistics to sensationalise, inflate, confuse, oversimplify and - on occasion - downright lie, he decided to shed light on their ill-informed and sneaky ways. How to Lie with Statistics is the result - the definitive and hilarious primer in the ways statistics are used to deceive.

With over one and half million copies sold around the world, it has delighted generations of readers with its cheeky takes on the ins and outs of samples, averages, errors, graphs and indexes. And in the modern world of big data and misinformation, Huff remains the perfect guide through the maze of facts and figures that are designed to make us believe anything.

'A hilarious exploration of mathematical mendacity.... Every time you pick it up, what happens? Bang goes another illusion!' The New York Times

'A pleasantly subversive little book guaranteed to undermine your faith in the almighty statistic' Atlantic