STATS ARTICLES 2012
Cacao or Caca? How the media bit into chocolate Nobel prize link
Cindy Merrick, October 16, 2012
Who wants to be a Nobel Prize recipient? According to coverage of a recent note by Franz Messerli, M.D., in the New England Journal of Medicine, which claims to have found a link between a nation’s chocolate consumption and its number of Nobel laureates, you may be just a few Hershey bars short of one of the most prestigious honors out there. Or maybe it’s based on the country you were born in. We’re not sure, but more chocolate couldn’t hurt. But we’re also not sure whether the whole thing isn’t a joke - and neither is the media.
According to medical journalist Larry Huston on Forbes, who seems to take the story quite seriously, “You don’t have to be a genius to like chocolate, but geniuses are more likely to eat lots of chocolate.” He goes on to cite Messerli’s reported significant correlation between the chocolate consumption in 23 nations and their number of Nobel laureates, noting, as did Messerli, that when Sweden is removed (due to presumed patriotic bias, as well as lower chocolate consumption), the correlation is even higher! However, Huston neglects to mention, as Messerli did, the inconvenient problem that there is no individual data on chocolate consumption by Nobel laureates. It may be that the flavenols in the chocolate I eat give me an enhanced sense of well being, but it is another thing to suggest that my well being has a mechanistic effect on the work of scientists and writers laboring across town.
Most coverage (Washington Post, Huffington Post, New York Daily News, USA Today) seems to have cut-and-pasted from an Associated Press story which, despite a bit of gleeful amusement (ain’t science grand?), swallows Messerli’s article as if auditioning for the gluttonous role of Augustus Gloop in Willy Wonka. Even when not running wire copy, journalists – or editors – couldn’t stay away from the secret sauce treatment. “Secret to winning a Nobel Prize?” inquired Time. “Eat More Chocolate.” The Secret To Genius?” asked NPR “It Might Be More Chocolate.”
Might this have been nothing more than a bit of fun, which perusal of scientific journals – if not actual studies – has shown scientists to be sometimes capable of? “No,” said MSN Now, “this isn't some wonderful hoax: The study was authored by Franz Messerli, a genuine doctor, and published in the New England Journal of Medicine, a genuine medical publication.” A “genuine” doctor! It must be true then.
But Reuters appears to have cottoned on, noting that “Messerli admitted the whole idea is absurd, although the data are legitimate and contain a few lessons about the fallibility of science.” Business Insider also got the joke, quoting American physicist and Nobel laureate Eric Cornell: “National chocolate consumption is correlated with a country's wealth and high-quality research is correlated with a country's wealth. So therefore chocolate is going to be correlated with high-quality research, but there is no causal connection there."
The purported logic goes like this. Start with the assumption that chocolate improves a person’s cognitive function. Then suppose that a whole population could see such an improvement, if enough of them were high consumers of chocolate. Then conclude that this could be indicated by a strong correlation between that population’s overall chocolate consumption and some comparative measure of that population’s cognitive function. Maybe a nation’s proportion of Nobel laureates is a good indicator of that nation’s cognitive function.
Not much scrutiny is required to show how quickly Messerli’s argument is, despite being wrapped in scientific prose, the stuff of Willy Wonka. For one thing, the data on chocolate consumption were drawn from a handful of international associations that track such things, dating back to 2002. The Nobel Prize has been awarded each year since 1901. How accurate can it be to correlate one decade’s worth of chocolate consumption data to over 100 years of Nobel Prize awards?
Second, the Nobel data cited in Messerli’s article are from a Wikipedia page, which itself cites a BBC News story that compiled the list of laureates by country from the Nobel organization’s database. In its description of the list of Nobel recipients by nation, the Wikipedia page states that “Where the [Nobel] website mentions multiple countries in relation to a prize winner (country of birth; country of citizenship; country of residence at time of award) each of those countries is credited as having won the prize.” So, for example, Erwin Schrodinger, who won the Nobel Prize in Physics in 1933, is likely listed under Austria (where he was born), and under Germany (where he did the work for which the prize was awarded). Simon van der Meer, who won the Nobel Prize for Physics in 1984, was born in the Netherlands, and received the prize for his affiliation with CERN, in Switzerland. It is not hard to see how such cosmopolitanism among Europeans could easily run up the tally for Nobel prizes among nations that, perhaps just coincidentally, account for approximately 40% of the world’s chocolate consumption. Add to this the fact that the Nobel organization’s literature committee is widely criticized for its obvious euro-centrism, and hopefully you find yourself thinking that the correlation looks a lot more like a coincidence.
Last, let’s restate the key empirical flaw in the claim: there is no data on chocolate consumption by individual recipients of the Nobel Prizes. In the end, the individual impact of chocolate on Messerli’s single indicator of cognitive function across populations is a nearly complete unknown – drawn only from chronologically mismatched data about one or more nations in which the recipient of the prize was known to have lived.
The upshot for a lot of journalists: you were hoodwinked. One look at Wikipedia could have saved you.
Cindy Merrick is a PhD candidate in Mathematical Sciences at George Mason University and an intern at STATS.org.