How to Lie with Statistics

Chapter 3 covers the small things that often go missing from published statistics. Huff notes that the problem with missing information is that it usually needs to be pointed out to general audiences. Because members of the public don’t know they should look for these details, those who make deceitful statistics can present wild claims without being caught.

Huff begins by returning to the problem of samples by talking about those that are too small. In statistics, the use of large sample sizes produces the best results. Misleading statistics sometimes favor small sample sizes. The reason for this, Huff says, is that “with a large group any difference produced by chance is likely to be a small one and unworthy of big type” (40-41). An insignificant difference in a population appears more significant in a small sample than in a large one, which makes it look more meaningful and dramatic for advertising purposes, for example. He uses flipping a coin to make his point: In his small sample of flips, one side came up more often than another, but if enough flips are done, the probability of half heads and half tails emerges.

Huff continues by talking about the issue of how big is big enough for a sample size. This depends on factors such as the size of the population or the variation within said population. Huff also warns that the necessary size might be larger than it appears. He uses the field of medicine as an example and warns against statistically unproven treatments.

In the next section, Huff gives the reader ways to look for inconclusive statistics. The first is to examine the degree of significance. In statistics, 5% is the worst the significance can be while being reasonably accurate. Another way is to look at the range or deviation. Huff returns to the issue of averages and says that a misunderstanding of averages, often by leaving out this information, leads to severe consequences. For example, parents might react negatively when their children aren’t meeting the “average” for development as they age. This can happen when a well-made statistic loses information, such as its range, before it reaches its audience. Huff builds on his last point by noting the issues of using “normal” to measure how good something is. He discusses the work of Dr. Kinsey again, saying that pointing out that a sexual practice is statistically “normal” does not mean that the researcher approves of the practice.

The final section of the chapter discusses missing information in statistical charts. Huff says that important information often goes missing, such as the numbers on the sides of the chart or what those numbers represent. He concludes that charts or other statistics that lack key points that would allow the reader to understand what is being measured are deceitful and don’t provide an accurate picture.