A recent study suggesting a link between coffee drinking and longer lives has prompted a flurry of coverage—some snarky, some cautious, but mostly celebratory. (We see you there, reaching for another cup of coffee.)
The study published at the prestigious New England Journal of Medicine is about as good as observational epidemiology studies go, but it’s limited by virtue of being observational. Last month on our Crux blog, Gary Taubes wrote a hard-hitting piece about the problems with observational studies. A major limitation of surveying people about their lifestyle habits is that correlation does not imply causation. It can’t prove coffee drinking actually led to living longer. There are always confounding variables. Read More
If you remember anything from statistics class, it’s probably that correlation ain’t no causation. Just because two numbers happen to go up at the same time doesn’t mean that one is causing the other to rise (or fall, or hold steady, or whatever). If there isn’t a plausible explanation for how the two might be connected, and proof that that explanation is indeed the cause, all you have is a couple of lines on a chart.
So it’s a move in the right direction when people try to suss out connections between two variables they have a hunch are related. But seeking such a connection can lead to some pretty convoluted reasoning. A new paper—unpublished, but released for discussion—claims that the passage of states’ medical marijuana laws cause decreased traffic fatalities, following on the researchers’ intuition that people might smoke pot instead of drinking alcohol if marijuana were more readily obtainable, and that driving while high is less dangerous than driving drunk. Their logic goes like this: if medical marijuana laws make pot more easily available, and people smoke more pot after laws are passed, and they buy less alcohol because of that, then there would be fewer traffic deaths.
As the 2012 presidential race ramps up, campaigns are courting voters not only at the traditional county fairs and town hall meetings, but online—and generating, in the process, an enormous amount of data about who potential voters are and what they want. At CNN.com, Micah Sifry—an expert on the intersection of technology and politics—delves in the Obama team’s extensive efforts to mine and manage the data in a way that could help them better interact with voters and home in on important issues. He writes:
What’s the news: In a study published last week, researchers showed they could reconstruct video clips by watching viewers’ brain activity. The video of the study’s results, below, is pretty amazing, showing the original clips and their reconstructions side by side. How does it work, and does it mean mind-reading is on its way in?
New tools for conservation?
What’s the News: Maybe it’s you—or maybe it’s the dice. A technique that relies on concealing individual transgressions while revealing greater truths is letting biologists get to the bottom of South African farmers’ killing of leopards.
A house decimated by the 2010 earthquake in Chile.
What’s the News: Enormous earthquakes are rare; there have been only seven quakes with a magnitude 8.8 or above since the start of the 20th century. Of those seven quakes, three of them have happened in the past seven years: off the coasts of Indonesia in 2004, Chile in 2010, and Japan last month. Some researchers think this earthquake cluster marks the start of a period of megaquakes, while others believe that the earthquake cluster is simply a statistical fluke, with these unusually massive quakes just happening to occur within a short amount of time, according to recent analyses (PDF) of Earth’s earthquake history presented at the Seismological Society of America’s annual meeting last week.
What’s the News: Many evaluations of scientific excellence singling out specific universities or departments, but two European researchers have taken a different approach: They rated the top scientific cities by looking at what proportion of published science articles are highly cited. Cambridge, Massachusetts, came out as the winner in physics and chemistry (no surprise there—MIT and Harvard) for having lots of influential papers; London was tops in psychology; Moscow was the chemistry and physics loser; and Taipei, Taiwan was the low achiever in psychology.
How the Heck:
What’s the Context:
Not So Fast: As the researchers note, the study fuzzes over any distinctions that emerge on a smaller scale than a city—for instance, the maps don’t show any difference between a city with one superstar who publishes 10 influential papers and another city with a group of 10 researchers who each publish 1. And since the scoring is based on citations, it’s subject to biases based on renown, language, and resources; the same paper published by a famous researcher at Oxford will get more notice than if it were published in Nigeria.
Reference: arxiv.org/abs/1103.3216: Lutz Bornmann and Loet Leydesdorff, Which Cities Produce Worldwide More Excellent Papers Than Can Be Expected? A New Mapping Approach—Using Google Maps—Based On Statistical Significance Testing