You know how sometimes I rant about the conflation of correlation and causation? Sara is responsible for that. My reading-heavy, numbers-light education was a little lax on all that science-y stuff. Hers was not. She pointed out something really fascinating about a particular type of Jezebel content, and I asked to her explain it to all of you. Here she is:
* * * * *
Whatever it might make you think about me, I’ll say it: I read a lot of Jezebel. I think it’s smart and funny, that they take a much stronger stand on issues of sexual assault and abuse than do many other media outlets, and I can’t help but love the snark. While occasionally I cringe when they take an advertisement I have no problem with and dissect its misogyny, for the most part I find the writers intelligent, informed, engaged and hilarious.
The one area of their writing where this is consistently not true is what I will hesitantly call “science writing.” I say hesitantly, because while these articles are written about scientific studies, and purport to offer a scientific evaluation, they are anything but. Jezebel, usually so critical, regularly fails to extend their critical thinking skills to science, and I can’t help but wonder if it’s because no one who writes for them has a background in the field.
I’ll take one of my favorite recent examples. In June, Jezebel ran a story about two new studies showing that diet soda increases weight gain and risk of diabetes. At first, this seems fairly unsurprising – we all know diet soda isn’t on the food pyramid. Then, a portion of the study is quoted saying that this claim is based on a comparison of diet soda drinkers and non-soda drinkers, a claim Jezebel doesn’t question. How can the study make a claim specific to diet soda drinkers when what it really compared was soda drinkers and non-soda-drinkers? At the very least, you’d need to compare soda drinkers and diet soda drinkers to make any sort of claim about the effects of specifically diet soda. And what about other lifestyle choices? It seems pretty likely that people who don’t drink any soda at all might generally make healthier eating choices, or drink more water, or exercise more, or any other habit that might explain their relative health. By the same token, it seems possible that diet soda drinkers might generally eat more sugar than non-soda drinkers (and the study never points out that there isn’t actually sugar in diet soda) because they have a sweet tooth and are generally a little unhealthier.
My point is not that any of these things are necessarily true, it’s that the study is a poorly conducted one because it makes no effort to rule out any of these other correlations. And Jezebel, queen bee of critical reading, fails to consider any other claims besides those made in the study. Time and again, when Jezebel purports to discuss anything in the realm of science, they suddenly lose the ability to critically analyze the material they’re presenting, and they do their readers a great disservice in the process.
* * * * *
[It’s Emily again.] Here are a few other examples of Jezebel’s science-y coverage being a little heavy on conclusions and light on criticism. This one’s about girls and sports and professional success. This one’s about virginity loss and divorce rates. While Jezebel (rightly) undercuts the immediate assumptions about promiscuity, they don’t point out the dearth of discussion about other possible correlations and causes, like socio-economic class and religion.
Related Post: Sometimes Jezebel pisses me off, too. Like when they apply a hacksaw to a Bieber quote to deliberately create outrage.
Related Post: Depending on how you cut this data on body type and sex drive, the correlation/causation gets real tricky, and changes the outcome in drastic ways.