What are the misuses of statistics?

Answer 1

The misuses of statistics are too numerous to list. One of best books published on this subject, "How to Lie with Statistics", by Darrell Huff, published in 1954 is still in print, much longer than any textbook on the subject of statistics. An excellent starting point, but certainly not the only source on the subject, is wikipedia: http://en.wikipedia.org/wiki/Misuse_of_statistics Statistics, if I narrowly define it, as values resulting from numerical calculations made from a set of data, are not the problem. Any set of collected data will have just one mean value, one standard deviation, and other measures. The measures can be easily checked and verified. But in a broader context, the discipline of statistical analyses, includes the planning of data collection, the verification of collected data, the assessment or evaluation of potential errors in collection and results, and misinterpretation, either intentionally or unintentionally. The discipline of statistics when correctly applied is an unbiased interpretation of data in a critical manner. But, if the interpretation fits pre-conceived ideas, then frequently the critical examination is neglected. Some statistical analyses fit the category as simply speculative or controversial conclusions. Our first look at data, is with graphs, and the art of biasing graphs for the sake of making them more supportive our conclusions - is part of the misuse of statistics. See: http://www.roofable.com/2007/10/03/how-to-lie-with-graphs-the-ny-times-as-real-estate-case-study/ You will find many other examples by doing Google searches in the internet.