The Conundrum of Associations

BPAShould we reduce our spending on science, space and technology? A curious question, right? Why does it come up? Because the more we spend, the greater the number of suicides by hanging, strangulation and suffocation. Since 1999 the number of such deaths has increased in exact parallel to the increase in spending. Indeed, the correlation factor is an astounding 0.99. We owe this knowledge to Tyler Vigen, a criminology student at Harvard Law School who can best be described as a statistical provocateur. He has developed a website called Spurious Correlations which is dedicated to mining data in order to reveal amazing but meaningless correlations. The point of course is to show that just because two variables track each other closely over time doesn’t mean that one causes the other.

Since 1999 we have increased spending on science, as we should have, and we have seen an increase in suicides. But one has nothing to do with the other. The population has increased and suicides are proportional to population.

Another interesting correlation is between ice cream consumption and murders. It is most unlikely that ice cream turns people into killers but hot weather may increase aggression and of course it also increases ice cream sales. While coming up with such spurious correlations is amusing, it also serves an important purpose. That’s because associations are commonly used to imply that there may be a causative relationship. In some cases there may be, but one has to be very careful before jumping to such a conclusion.

Consider bisphenol A, the chemical used to formulate polycarbonate plastics and epoxy resins that line food cans. It has been accused of causing everything from obesity and cancer to heart disease and developmental problems. There is indeed an association between blood levels of BPA and heart disease but that may well be because people with heart disease have consumed more canned foods with high salt and fat content. They would have higher blood levels of BPA leached from cans, but the culprit is the sugar and fat content. In fact the Canadian Food Inspection Agency carried out a survey of BPA in canned foods and concluded that an adult would need to consume approximately 14 kg of canned vegetables each day (approximately 50 servings) to reach an exposure to BPA that may pose a safety concern.

On the other hand, there are some associations that may eventually prove to be causative. FDA in the U.S. recently learned of a cluster of acute liver failure cases in Hawaii. It turned out that the cases were associated with having used a dietary supplement called OxyElitePro advertised as an aid to losing weight and building muscles. The product contained aegeline, an alkaloid extract from leaves of the Asian bael tree. Researchers believe this may be the culprit, although genetics likely play a role because most of the victims were of Pacific island ancestry. Although the case is not iron clad, the supplement has been taken off the market.
Another association that has recently been publicized is that between the murderous rage of the California mass killer Elliot Rogers and his addiction to creatine which he was taking for body building to make himself more attractive to girls. There is virtually no evidence that creatine can lead to rage like steroids can, but nevertheless the association has been widely publicized. It likely has no greater connection to causation that than that between per capita margarine consumption and the divorce rate in Maine, another spurious association discovered by the clever Harvard statistician.


Joe Schwarcz

Leave a Reply

Spam Protection by WP-SpamFree

Blog authors are solely responsible for the content of the blogs listed in the directory. Neither the content of these blogs, nor the links to other web sites, are screened, approved, reviewed or endorsed by McGill University. The text and other material on these blogs are the opinion of the specific author and are not statements of advice, opinion, or information of McGill.