Skepticism

Skepchick Quickies, 8.26

Jen

Jen is a writer and web designer/developer in Columbus, Ohio. She spends too much time on Twitter at @antiheroine.

Related Articles

14 Comments

  1. My body (& I’m a guy) keeps telling me to lose weight & eat lots of fried food & pastry. I think my body might be schizophrenic so I stay out of the arguement.

    The proper way to convert acorns to a non-toxic food source is to let squirrels eat them & then eat the squirrels.

  2. @ “7 secrets your body’s trying to tell you”

    I hate how these articles always phrase stuff like this as mysterious “secrets” that your body knows and if you’d only stop being so stupid you’d see it’s “trying to tell you” not to get Parkinsons.

    That aside, most of what’s presented here seems to be reasonable advice whether or not the correlational data they’re citing is actually any good. Not exposing your liver to toxic chemicals is probably sound advice regardless of how stocky your legs are. So is getting exercise. Fish oil supplements though? I’ve read conflicting data on that.

    They’re making two claims in each of these “secrets.” One is that there’s a significant increase in the rate of acquiring a particular illness based on certain physical characteristics like finger lengths. And second is that their prescribed activity (doing exercises or eating a Mediterranean-style diet) lowers the risk for that illness. I’d like to actually see the studies they’re citing.

  3. All I could think of reading the 7 things article was “correlation ≠ causation” and now I want to read those studies and evaluate the data myself. Color me highly skeptical. I need more evidence than that.

  4. True enough that correlation is not equal to causation, but in these cases there could be a third element that is causal to the correlated variables…

    The problem is with their ‘1.5x more likely’ or ‘5x more likely’ statements. Without knowing how likely these things were in the 1st place, they seem a lot scarier. For example, if 1 in 100,000 people get disease x, that’s not very likely. Now, if 2 in 100,000 who have trait y get diease x, we can say that people with trait y are twice as likely to get disease x, but it’s still not very likely.

  5. Bets on the studies being taken out of context by the article?

    I’m curious how having more fat on your torso leads to heightened risk of *dementia* rather than *heart trouble*. Or maybe they assumed that wasn’t a secret, and the really weird connection between a beer belly and dementia was out of left field enough to count.

  6. Oh, and on the note of the article about American students falling behind in math and science; no shit!

    Part of the reason why the crop of grad students coming in to my university this year are 60% Indian, 25% Chinese, and 15% American.

  7. Good comments so far so I’ll try not to repeat. One thing I find perpetually annoying is they don’t make it easy to find the original paper. A researcher’s name or accurate title or even, gasp, a real citation would make it much easier to fact-check articles like this. Usually these things turn out to be misleading or crap or misleading crap, but it takes so much time to prove this it’s just not worth it.

    While the “secrets” were minimally cited, the solutions were not and thus are useless. This is just another journalist trying to fill some inches.

  8. @davew: This is definitely a trend. I have no idea whether it only recently started being acceptable to cite research without, you know, actually providing a citation, or whether it’s been standard procedure since the beginning of newspapers and I just never noticed before, but either way it’s fucking annoying.

    While it’s not always easy for an untrained person to interpret the p-values and whatnot included in paper abstracts, the text surrounding the scary numbers is typically pretty clearly-worded. Maybe that’s why journalists hate to provide citations; if people realise that paper abstracts are actually human-readable, they wouldn’t need science journalists to paraphrase press releases any more.

  9. @sporefrog:

    As far as my reading on fish oil goes, the Omega-3’s are pretty much widely accepted as helpful for reducing triglycerides, if you happen to have high triglycerides or maybe if you have a family history.

    Then there’s lots of other studies relating to Parkinson’s, depression, schizophrenia, etc., but those are pretty preliminary so far – small study sizes, or just using mice, etc.

Leave a Reply

You May Also Enjoy

Close
Close