Statistics Workbook For Dummies with Online Practice, 2nd Edition

Chapter 21

Ten Ways to Spot Common Statistical Mistakes

IN THIS CHAPTER

Knowing how to sniff out incorrect statistics

Being on the lookout for misleading statistics

You shouldn’t accept every statistic without question. When you know what questions to ask and what to look for, spotting problems in statistics and feeling empowered to decide whether to believe the numbers you see becomes much easier. It also helps to be as specific as possible in terms of critiquing a statistic. Saying exactly why a statistic is incorrect or misleading is much more effective than just saying that a statistic is “bogus” or “biased.”

In this chapter, I summarize the big ideas for spotting errors or misleading statistics, show you how to report them with the right lingo, and give tips for avoiding them altogether when you do your own statistical analyses.

Scrutinizing Graphs

Graphs can be misleading in a number of ways: by what they show and by what they leave out. Here are some tips for looking at graphs and spotting misleading info:

• The values on certain types of graphs should sum to 1. For example, a pie chart or a relative frequency histogram/bar chart. In general, for any chart/graph where the group is broken down into categories that do not overlap, and percents are shown, those percents ...