Correlation Does not Imply Causation

Today, while I was taking a writing class, the teacher encouraged us to read more because "good writers all read a lot". I brought the question, "does lots of reading make people better writer? Or is there something else?" Immediately, I had people jumping up and down saying that it does and that it is reading. They came up with all sorts of explanations about how reading helps one write better. There were explanations from "that is how you learn” to all sorts of theories on how the brain works and how it affects one's writing. Never once did someone realize that not all readers are good writers.   Never once did they take a second to think that maybe reading doesn't affect one's writing or that maybe there is something else that good writers have in common. I cannot tell you how many times I have seen this sort of thing. I therefore feel the need to explain this important concept and correct a few common ideas brought on by this kind of reasoning.

Think about this statement: “Children raised in homes with more appliances tend to perform better in school. Therefore, appliances improve intelligence.” The reasoning behind is obviously flawed. More appliances in a home usually mean a higher family income, a third variable that is not controlled in the experiment. This is called “Correlation proves causation”, a logical fallacy in which two events that occur together are claimed to have a cause-and-effect relationship. Of course, everyone realized this flaw because the implied meaning is so outlandish. But when people encounter things that are more plausible, like what was brought up in my writing class, people don’t notice. Similar examples are:

•       Teens involved in violent crimes tend to play violent video games. Therefore, playing violent video games causes teenagers to get involved in criminal behavior. – Whether or not this is true is still being debated, but this logic here is flawed.
•       Increased TV watching results in shorter attention spans.
•       Women who take combined hormone replacement therapy (HRT) have a lower-than-average incidence of coronary heart disease (CHD). Therefore, many doctors thought that HRT was protective against CHD. –This was later proven false. Randomized controlled trials showed that HRT caused significant increase in risk of CHD. Women undertaking HRT were more likely to be from higher socio-economic groups with better than average diet and exercise regimes.

This often leads to misinformed government policy that will end up hurting millions of people, if not all people. Instances of this include:

• Economic indicators are now often watched instead of the economy itself. During the beginning of this recession, the government was saying that the economy was in good shape and was citing all sorts of indicators to prove it, but a look out the window will show that all was not well and the economy was in recession. 
• Then, there is the manipulation of the indicators, making them further unreliable. GDP is a prime example. GDP growth is usually used to show that an economy is growing. But that has led to policies that boost the GDP, but don't help the economy and the delaying of policies that don’t directly influence the GDP but boost it. GDP = private consumption + gross investment + government spending + (exports − imports). So governments have been spending massive amounts of money, trying to use the government spending section to make up for the loss in other parts of the equation. This does not really benefit the economy. Then they also decided against freeing up trade because the increased imports and exports would cancel each other out, despite the fact that it would lead to a livelier economy and that will encourage more spending and investment causing real growth.
• GDP itself is also flawed. GDP suggests that breaking windows so that more windows would need to be purchased (consumed) would make everyone wealthier. The thinking is because more windows are being consumed, the GDP goes up, which in turn means everyone is richer. It couldn't be farther from the truth. GDP does not include accumulated wealth. So the higher GDP only replaces what was lost but the resources used to make the new window could have been used for something else, therefore, it makes everyone poorer. 

So the next time you encounter reasoning like this, don’t automatically assume causation in situations that are really just correlations. However, it doesn’t hurt to look at it a lot more closely. Consider the correlation, perhaps, a symptom.