Rant on understanding what you’re studying

This article on increasing mortality rates, i.e., decreased lifespan, among certain American women is a good one to for econ and math professors to give to their stats students as a quiz. “Identify errors….”

In an email to a friend, I described this article as  half-written” because it so obviously makes statements that are completely wrong and non-sensical if someone had bothered to read them out loud.
 
Here’s a taste:
“Life is different for women without a high-school degree than it was a few decades ago, and in most cases it’s a lot worse,” she said. “It’s really just a perfect storm.” 
That’s following a discussion that says that education is driving the seemingly crazy increase in death rates. Of course, that’s nonsense because there’s no plausible mechanism for education qua education to lead to early death. There probably are a lot of other things correlated with or even caused by being a female high school dropout, like being poor, fat, a smoker, frequently pregnant and drunk (you *know* those two are linked as tightly as my punctuation indicates), and a meth-head. Add that to all the other crappy side effects of being poor, and it’s no wonder. But dropping out doesn’t cause these problems — handing out degrees won’t fix them. Yes, education helps, but that facile answer sidesteps the most uncomfortable of assessments in America:  people are different at the same time as they are all equal. (This is my nod to the possibility that the causes of dropping out are probably also causes of those other drivers of poor health.)
 
Earlier, the article did the same thing with location. Absolutely lazy-ass crap to write that. Unless there’s poison in the ground or radiation in the air, location is almost certainly 100% not causally related to anything that’s going on. Sure, it’s correlated, but so what? 
 
Two closing thoughts, and then I’m going to post this rant:
 
1. This is another example of Prof. Kingsley’s admonition against doing the experiments you can do instead of the ones you should do. (I have always thought that there was a subtext there about folks not always knowing the difference, but that’s mine not his. As smart as he is, I’ve never heard him utter a disparaging word about the brainpower of other scientists.)
 
2. Clay Shirky, NYU Prof among other things, recently tweeted that he realized the problem with daily journalism is that the deadline doesn’t care whether you understand what the real story is. (Of course, CNN has that problem in spades.)
 
These comments about location and education “causing” the observed death rate are either sloppy reporting/research (someone doesn’t understand how things fit together) or there are folks who frankly aren’t done with their work — if these are the proffered “explanations.”

Leave a Comment