5 ms·
> This relationship remained significant when influences of caloric intake, urbanization, obesity, education and carbohydrate crops were statistically controlle
by zhivota 3y ago
> This relationship remained significant when influences of caloric intake, urbanization, obesity, education and carbohydrate crops were statistically controlled.
In other words, they did not control for wealth or income, which quite obviously confound this entire exercise. Probably this paper is actively worse than useless.
- nomel 3y ago> education Wouldn't this show up as a relation to education, since there's a very strong relation between education and income?
- voxl 3y agothe real problem with these studies is the critics. If it's a study that says veganism is good, then everyone does that or has a moral imperative for that to be true will come out and say "what more do those evil meat eaters want?" while simultaneously the meat eaters will lambaste the study as worthless on any point they can find. Of course, the situation works in reverse. The problem is that it's incredibly difficult to find genuine, unbiased, nuanced criticism of diet research. Some people are dumb enough to see research that actively admits to funding from some food industry and immediately they say "this study is total bullshit" which is just not how science works.
- SantalBlush 3y agoIt's methodology sniping. Find some detail in the paper that looks wrong to laymen, then dismiss the whole study as trash.
- appplication 3y agoThis thread is giving me Deja vu https://news.ycombinator.com/item?id=38830148 https://news.ycombinator.com/item?id=38830148
- 4death4 3y agoThis is the problem with almost all nutritional studies. They’re almost always based on correlation. But human life has millions of facets, and each facet is correlated with thousands of other facets. It’s nearly impossible to discover a causal relationship between two lifestyle variables unless the effect is very pronounced.
- hombre_fatal 3y agoSo, we can't know trans fats are bad nor any other facts that you take for granted? What about smoking? I don't think people who say this realize the bullet they would have to bite. Though I only seem to hear it from people who are about to reject scientific consensus on something that's inconvenient for them because they like the taste of certain foods, like butter.
- 4death4 3y ago> It’s nearly impossible to discover a causal relationship between two lifestyle variables unless the effect is very pronounced. Also, for many things (like smoking), we have established causal relationships by explaining the mechanism of action. E.g. cigarette smoke damages DNA, lead poisons the brain leading to neuron death, etc.
- sjducb 3y agoAlso smoking has a large effect so you do notice it in every correlative study too.
- kmmlng 3y agoThe fact that you can pick up on negative effects through observational studies if they are large enough makes me especially suspicious about results that differ from study to study. Perhaps there is an effect there, but if we can't reliably pick up a signal, maybe it's not something I need to worry about.
- 3y ago
- huitzitziltzin 3y agoThe paper is total garbage but they do put gdp in the specification. No one should take this “result” remotely seriously though.
- coldtea 3y agoJust out of curiosity, are you a vegan?
- niemandhier 3y agoThey take GDPPP and other factors into account, and further group countries by similar overall lifestyle e.g. Mediterranean countries. I have not checked all the methods ,but I believe your criticism is to strong.
- coldtea 3y agoEducation is already a good proxy for income (thought they do take into account other criteria too for that).
- fulafel 3y agoIt would be interesting to see stats on how education vs income relate to meat consumption, since education seems to also correlate with adhering to a plant-based diet by ethical choice (concern for climate or animal welfare).