I don't know how many times I have come across the advice that a person can improve his or her disposition through exercise. People who exercise are happier, statistically speaking, so the best step toward beating the blues is to actually take several steps. It turns out that this may not be completely true, and that the reason people thought this was true due to a common mistake—the opinion that statistical correlation equals cause and effect.
Not long ago I came across an article describing a study that challenges the idea that exercising makes people happier. Just because people who exercise are more happy does not mean that one causes the other. Each may be caused by a common third factor, such as genetics. People who are genetically wired to be more happy may also be genetically wired to enjoy jogging.
This mistake is one that I come across all of the time when I hear or read about scientific studies. Certain research is done, a correlation is found between two sets of data (like hours of exercise and levels of happiness), and an unchallenged statement is made in the media that research shows that one of these measured factors must cause the other.
At first glance, this should not be a big deal. Most people don't care too much what studies say, and they really only bring up the topics for small talk anyway. It is bothersome, though, because important decisions are often made based on the the implications of such research. Governments pass and enforce policies and laws because studies indicate something specific. People are prescribed medications based on the implications of research. A murder case can hinge on what research studies imply about the workings of a mentally unstable person's mind.
With all of this said, I know this is why most published research must go through peer review. Mistakes in data collection, computation, and interpretation can get filtered out there. But what happens if everyone in the peer review process makes the same assumptions in interpreting the data? From what I can see, that story ends with a bunch of depressed people in jogging shorts.
Subscribe to:
Post Comments (Atom)
3 comments:
Interesting question. As someone who works in research, I know all of ther researchers I work with consider all of the possible conflicting factors that can interfer with a pure interpretation of their data.
With human research, it's much more difficult because you can't put them in a laboratory in cages and control their genetics, social interactions, food, exercise, etc...
Interpreting the research you are talking about would have to be slightly bias because most of the other factors that can make one happy are not controllable in a scientific study when one is dealing with a human being.
I wouldn't be so hard on exercise. Generally speaking, I know I feel a little happier after I exercise.
Yeah, I figure that most research is actually done properly. Occasionally a story like this slips through, though.
I think the real issue is that there is a great deal of desire to have research interpreted in the media, and/or people reading research results are willing to make educated guesses off of the research.
Post a Comment