Correlation Versus Causation
When two things are related, does one usually cause the other?
There is a tendency in fitness to group things together that appear related. For example, competitive long-distance runners tend to be thin. Bodybuilders tend to have more muscle mass. Yoga instructors tend to be more flexible.
Those related things are called correlations.
When people see related things, it’s natural to take the next step and assume that one CAUSES the other. Long-distance running CAUSES you to be thin, or taking yoga CAUSES you to be more limber. However, it’s not always as simple as one causing the other.
If someone is born with more muscle mass, they’ll see quicker and more dramatic results from a weight training program than a naturally thin person. Perhaps people with more muscle mass are more likely to engage in bodybuilding because that produces better results than other fitness programs.
There may be many underlying factors to consider. Correlation does not automatically lead to causation.
Here’s an example. The average height of NBA players is 6’6” tall. There’s a correlation between height, professional athletes and the NBA. But you probably wouldn’t claim that playing in the NBA CAUSES someone to grow tall.
The problem of correlation versus causation can create problems for everyone, even scientists trained to look for it.
In a study done by the University of Pennsylvania’s Medical Center in 1999, researchers made this bold claim. “Children who sleep with a light on in their bedrooms at night before the age of 2 may be at significantly higher risk of developing myopia - near-sightedness - when they become older than children who sleep as infants in the dark at night.”
That statement has enormous implications. Should night lights be banned to help children avoid near-sightedness? Will turning off the lights at night reverse the potential problem? Will a night light that turns off after a few minutes have the same effect?
Turns out, it wasn’t the night lights; it was the parents. When scientists at Ohio State University looked at the data and reviewed the study, they found that nearsighted parents were more likely to leave their children’s light on at night so that the parents could see better. Nearsighted parents are more likely to produce nearsighted children. The night lights weren’t the cause of the condition, the parents genes were.
That’s what’s known as a COMMON-CAUSE VARIABLE. The third factor C (nearsighted parents), was the reason that A (night lights were used more often) and B (children whose parents used night lights were more likely to develop near-sightedness.)
There are three additional common ways people incorrectly link correlation and causation.
REVERSE CAUSATION is that B causes A.
If someone is overweight, they’re more likely to suffer from depression. Reverse causation says that depression leads to putting on extra pounds. In fact, researchers found that putting on excess weight often leads to depression AFTER the weight is gained.
BIDIRECTIONAL CAUSATION is that A causes B and B causes A.
A classic example is people trying to preserve grasslands. Elephants eat the grass, so you would think fewer elephants would be an excellent way to promote more grass growth. In fact, elephants feed the grass with manure, so as elephant numbers increase, they create more lush grasslands. More elephants feed more grass. More grass helps feed more elephants.
Finally, there’s COINCIDENCE or SPURIOUS CORRELATIONS. Just because two things follow a similar pattern doesn’t mean there’s any correlation.
For example, there is a 94% correlation between the per capita consumption of cheese in the United States and the number of people who died by becoming tangled in their bedsheets. Of course, there’s no correlation; it’s just a coincidence. (You can see hundreds of spurious or coincidental correlations like that on tylervigen.com.)
In the world of health and fitness, you have to watch out for unfounded claims that “doing this causes that.” Don’t assume someone with a six-pack got that way because of a particular diet or workout routine. They may simply have great genetics that gives them an advantage no matter what type of program they use. Always look for clinical studies that test and prove the claims across a wide range of people. Then repeat this phrase. Correlation does not mean causation.
Near-Sightedness In Children Linked To Light Exposure During Sleep Before Age Two
Science News - The Children's Hospital of Pennsylvania - May 13, 1999
Call for a FREE Consultation (305) 296-3434
CAUTION: Check with your doctor before
beginning any diet or exercise program.