• berrytopylus [she/her,they/them]@hexbear.net
    link
    fedilink
    English
    arrow-up
    20
    ·
    edit-2
    1 year ago

    A lot of this reporting is a big misunderstanding of statistics.

    As the study says

    About 45% of the population had zero beef consumption on any given day, whereas the 12% of disproportionate beef consumers accounted for 50% of the total beef consumed

    Now just as a thought experiment, do you think that almost half of the US never eats any beef? No, of course not. But on any given day? Sure, quite possible. People’s diets vary.

    A randomly selected person might have a McDonald’s hamburger for lunch and a steak for dinner and be part of the 12% on the first day but then eat mushroom ravioli for lunch and pizza for dinner on the second day and be part of the 45%.

    And there might be certain demographics that are more likely to make up that 12% on a given day but that doesn’t mean there’s a particular nonchanging group of high consumers.

    I’m not going to dig into the study here but just as an example, let’s say Dog Breed X is 1.5 times more likely to bark than Dog Breed Y is. You can’t hear a dog bark and say “Ah it must be Breed X then!”, you can only say “Ah, it’s more likely from Breed X than Breed Y”.

    • BynarsAreOk [none/use name]@hexbear.net
      link
      fedilink
      English
      arrow-up
      15
      ·
      1 year ago

      I recommend you check the paper and not just go by the usual shit science reporting from MSM, the authors address this issue

      One limitation of this work is that it was based on 1-day diet recalls, so our results do not represent usual intake. Averaging both days of data available on the NHANES would not address this problem, would reduce our sample size by 15%, and would mix recall methods between an in-person interview (day 1) and one done on the phone (day 2). Still, as a check, we examined day 2 and found the same associations with gender and MyPlate guidance. Other associations were similar in magnitude, though not always significant. Another potential limitation is that the NHANES is a US study, and the data we analyzed are from 2015–2018. Thus, these results may not be generalizable or useful for targeting interventions in other populations, and do not capture any changes that have occurred in the correlates of beef consumption since the COVID-19 pandemic. At this point in time, however, post-pandemic NHANES dietary data are not available.

      • berrytopylus [she/her,they/them]@hexbear.net
        link
        fedilink
        English
        arrow-up
        5
        ·
        1 year ago

        You say “address” as if they were able to appropriately fix the issue, rather than addressing it as a limitation of the study. Limitations are fine, I’m just trying to explain the big one here in an easier to understand way because the reporting makes it seem like it’s a consistent 12% eating a shit ton of beef.