Correlation vs Causation
Photo by Dmitry Ganin on Unsplash
I was reading a book recently called The Purpose Revolution and on page 77 the author referenced a study that reported people who feel connected to their life purpose get more sleep. It was an interesting statistic; however, I couldn't help but question it. Do people who feel connected to their purpose get more sleep? Or is it that people that get more sleep are more likely to feel more connected to their purpose? Or is this just two unrelated data points where causation does not exist at all?
After looking into some of this research, I learned the conclusions were made from cross referencing sleep data (specifically having sleep apnea or restless leg syndrome) with self-reported statements about purpose. In other words, you cannot clearly say which one came first: the good sleep or the purpose. Unfortunately, this is the problem with throwing out statistics, especially in today's political climate. If you don't fully understand the difference between correlation and causation, and how data works in general, it is easy to jump to conclusions.
When two data points correlate, it means they appear to change in the same way. In most cases, this is in a linear sense. When one data point increases, we also see the other data point increase. In the case of the sleep study, people who felt more connected to their purpose were getting more sleep, and people who were less connected to their purpose were getting less sleep. This is correlation. Correlation is easy to spot - just look at the data: if they both go up and down at the same time, there's correlation.
Causation, on the other hand, means there is a cause-and-effect relationship between two items. This relationship requires more rigorous testing and controlled environments to be able to determine; just because one data point goes up (purpose) does not mean it is causing the other data point to go up (sleep). Fortunately, for these two sets of data, assuming causation is not likely to cause anyone harm: we all could use more purpose and more sleep. Unfortunately, politicians have recently been throwing out health data and assuming causation without proper testing or studies. In these cases, it leads to misinformation and can harm the people who listen to them. So, without getting too serious on a Sunday morning, I just encourage you this: the next time someone tells you X causes Y, take a step back and ask that question for yourself: is there causation here? Or maybe do we need more information before we can draw conclusions?
So, do we try to get more purpose in our life? Or should we try to get more sleep? In this case, why not try both? Or at the very least, next time you’re tossing and turning at 2 AM, instead of counting sheep, try counting your reasons for getting up in the morning. Who knows? You might just snooze your way into a deeper sense of meaning.
For a good laugh, check out this website that shows the strong correlation between two random things. It even uses AI to generate a fake explanation and a fake research paper to back up the data. My favorite is the correlation between searches for Never Gonna Give You Up and Netflix’s stock prices.
