We Know What Causes That

Dr. Franz H. Messerli, a recognized expert in hypertensive cardio-vascular disease, has published over 600 papers and written 20 books. In 2012, in the prestigious New England Journal of Medicine, Dr. Messerli wrote a short article about Nobel prizes and chocolate consumption. His research uncovered a strong linear correlation (r=0.791, P<0.0001) between the number of Nobel laureates a country produced per capita and the average chocolate consumption. He noted a possible link between dietary flavonols in chocolate and the reversal of age-related cognitive impairment. In order to think more clearly, he suggested, people everywhere should eat more chocolate.

While the article was intended as a funny reminder not to allow correlations to lead to false claims, not all of Messerli’s colleagues appreciated the humor. Many complained that the New England Journal of Medicine would publish such a thing, rushing to discredit the idea that eating chocolate would lead to more Nobel prizes. One even noted that there was a strong correlation between the number of IKEA stores and Nobel Prizes in a country, although offering no suggestions as to why that might be. If you’ve ever assembled IKEA furniture, you might have some ideas on this.

Food correlations are popular because we all eat. If eating chocolate in search of a Nobel Prize is not your thing, perhaps getting a doctorate in Civil Engineering piques your interest. Data shows a correlation (r=0.958) between the per capita consumption of mozzarella cheese and the number of CE PhD’s awarded. Since mozzarella is a key ingredient of pizza, this may explain its legendary popularity with graduate students.

Some of us are old enough to remember another popular correlation - the so-called “Hemline Index”. Economist George Taylor is often given credit, but while his 1929 thesis on post-war economic growth connected it with skirt length, he never proposed a general theory. As recently as 2010, economists Marjolein van Baardwijk and Philip Hans Franses of the Erasmus School of Economics examined this urban legend in light of hemline data from 1921-2009.

The accepted premise was that in prosperous times, people feel good and skirts get shorter, while empty bank accounts spoil the fun and cause hemlines to drop. The researchers were focused on the historical periods where the theory didn’t appear to hold. Their data analysis indicated that the economy leads the skirt length by about 3 years, although the conclusions were somewhat confounded by the weather. Even a full bank account doesn’t necessarily prompt a woman to wear a mini-skirt in the winter.

Not all researchers spend their working hours munching on pizza and chocolate while studying women’s skirts, although I’m not ruling out the possibility of finding data to support that conclusion. The Internet offers plenty of strange correlations, many of which should (but don’t always) fail the common-sense test.

As a kid growing up, I always looked forward to summers – you got 3 months off school and could do pretty much whatever you wanted while your parents provided free room and board. The other great thing about summers was that as the temperatures rose, the ice cream truck began making its daily rounds through our neighborhood. My Dad would grumble about how the ice cream truck’s wares were overpriced, so he would always lay in a summer stock of ice cream in our freezer. Confirmation bias notwithstanding, I just know that hot weather and ice cream sales have a strong positive correlation.

Mention hot weather to anyone in law enforcement, and I doubt if ice cream is the first thing that comes to mind. It is a proven fact that hotter weather begets higher crime rates – violent crimes in particular. Causation theories on this abound – increased levels of hostility and aggression, more social encounters that could turn violent, teens out of school with little to do and no supervision – people might even be fighting over ice cream.

It's easy to see how the comparison of two simple data sets, widely available on the Internet, can lead to some crazy theories. Comparing crime rates to ice cream sales would show a positive correlation, leading one to believe that eating ice cream makes one prone to criminal acts. Similarly, rising utility bills from increased air conditioner use would appear to make us more violent, at least until we identify the third variable – temperature. You get the idea.

At one time or another, most of us have been introduced to the Birthday Paradox. The basic idea is that if you have 23 or more people gathered together – like at a cocktail party or in a classroom – the odds are better than 50-50 that two will share the same birthday. The math is straightforward, although the result is quite counterintuitive. Someone inevitably points out that all birthdays are not equally likely, which is true, but has little effect on the outcome.

If you’re curious, the month with the most birthdays is August. Counting back 9 months puts us into the holiday parties, cold weather and cozy fires of December, a perfect setting for starting a baby.

And we know what causes that.

Author Profile - Paul W. Smith - leader, educator, technologist, writer - has a lifelong interest in the countless ways that technology changes the course of our journey through life. In addition to being a regular contributor to NetworkDataPedia, he maintains the website Technology for the Journey and occasionally writes for Blogcritics. Paul has 50 years of experience in research and advanced development for companies ranging from small startups to industry leaders. His other passion is teaching - he is a former Adjunct Professor of Mechanical Engineering at the Colorado School of Mines. Paul holds a doctorate in Applied Mechanics from the California Institute of Technology, as well as Bachelor’s and Master’s Degrees in Mechanical Engineering from the University of California, Santa Barbara.

---