Two recent studies in the news have been clear examples of the correlation vs. causation question that’s part and parcel of covering observational research studies.
And they’re worth taking a look at because the correlations are inherently interesting, even though they almost certainly aren’t causal.
First up is “The Effect of Sexual Activity on Wages”, which was published by Germany’s Institute for the Study of Labor. Predictably, and probably in part due to its terrible title, this study generated lots of headlines like “Have more sex, make more money”, from The Wall Street Journal‘s Marketplace, and Cosmopolitan‘s “The More Sex You Have The More Money You Make“.
Well, not exactly. The study found an association between sex and wages. As self-reported sex increased, so too, did income. Of course, that doesn’t mean that having more sex causes people to make more money, a point that wasn’t stressed clearly enough in some articles for Scientific American blogger Evelyn Lamb’s tastes.
But it’s still interesting to think about why the two might rise together, as Lydia DePillis did on Wonkblog:
The authors write that high levels of sexual activity are likely an indicator of good health, which also tends to correlate with higher earnings. It’s also possible that causality runs the other way: Earning more makes you more sexually attractive.
The second study had to do with person’s preferred brand of beer and whether it might signal an ER visit in their future. It turns out that Budweiser, the King of Beers, is also the king of injuries. About 15 percent of people who end up in the ER intoxicated and with an injury have been drinking Budweiser. Runner up among dangerous beers was Steel Reserve, a malt liquor that holds less than 1 percent of the U.S. beer market but accounted for 14 percent of ER-treated injuries among people who were drinking. Several malt liquors were associated with injuries.
Of course, a beer’s brand doesn’t make it more likely to cause a fall or a car accident, but its price and its alcohol content probably do. Some malt liquors have almost twice as much alcohol as regular beers.
As study author David Jernigan of the Johns Hopkins School of Public Health told Anahad O’Connor of The New York Times:
The findings could have policy implications, potentially influencing labeling requirements and marketing for higher-alcohol beers.