# Chapter 7: Statistical "Bait and Switch"

Home > Chapter 7

Chapter 7

We are constantly bombarded with statistics, and often we don't fully understand what they mean. Advertisers, labs, and other wrtiers can manipulate this confusion. If they are unable to prove what they want to prove, they can establish something else and profess that the outcomes are the same. They just choose something that is related to what they want to prove, and our minds do the rest. This is called using a semiattached figure. For instance (Example 2), a study in USA Today headlined that teenagers without an enforced bedtime are more likely to become depressed or suicidal. However, what was actually proved was that teens need sleep. An enforced bedtime and sleep are not the same thing. However, the authors of the article knew that bedtime and sleep were related, so they just pretended they were the same thing.

Another misleading tool that is used is comparative statisitcs. You can claim that something is "34% better", but often the article will leave out what it actually better than. For example, a toothpaste brand could say that their toothpaste is "34%" more effective. However, the toothpaste effectiveness could have been compared to brushing with no toothpaste, or even not brushing at all. It really means nothing for the consumer and doesn't actually tell them anything about the brand in comparison to other brands. Sometimes uncomparable groups are compared, and the results are often very misleading. A large group may be compared with a very small group, and the statisics report how many times something happened in each group. It isn't surprising to find out that the occurance was much more frequent in the large group, but this is most likely just because there was more of a chance for it to happen than in the small group. For example, a large state university could be compared with a small private college to see which has more foreign students. Though the private college might have a higher percentage than the large university, it is reported that the university had many more foreign students than the private college. The large university only had more because they had a significantly larger student body, so the results are very misleading.

 Advertisements and articles may also make claims that generalize a diverse collection of data. In these cases, further information is needed, though not provided. Example three provides an instance where the further information was actually provided. The article claimed that 1 in 5 adults will suffer from mental illness. However, as you read further, the statistics are broken down. Out of the large age group labeled adults, ranging anywhere from 18 to over 50, the small portion of 18-25 was much more prone to developing mental disorders than the older portion of the adult category. By originally lumping all age groups of adults together, older readers would think that they were just as likely as a 20-year-old to develop a mental illness, when in fact they are much less likely. When no further information is provided, the conclusions drawn from the statistics are often very wrong.

Check Out These Example Articles:

Kids of working moms are more likely to get hurt

Study links teen depression to bedtimes

Mental order strikes 1 in 5 adults