2 ms·

One of the worst scientific papers I've ever seen

3 points by Tomte 2mo ago

A_D_E_P_T 2mo ago
Just to play Devil's advocate here, high-level volleyball players tend to be extremely tall. The average male pro/olympic player is anywhere from 6'6" to 6'10". Some are, of course, even taller. There are well-known correlations between height/weight and lifespan, see e.g.: https://pubmed.ncbi.nlm.nih.gov/1600586/ https://pubmed.ncbi.nlm.nih.gov/1600586/ So it's not prima facie a foolish notion that high-level (pro/olympic-tier) male volleyball players suffer from a ~5-year reduced lifespan relative to a much smaller population. I've seen papers that are a hell of a lot worse, that's for sure.
newsomix9xl 2mo ago
As volleyball is a sport for the disadvantaged, this critique is the definition of punching down.
dlcarrier 2mo ago
I feel like this is par for any random paper. Never trust any conclusion's take on correlation vs causation.
jfengel 2mo ago
The article is oddly devoid of specifics. It just calls it bad, without quoting any of the article. The "key takeaway" is not from the article. It's the post's author imputing correlation, into an article that resolutely uses the word "association" over and over. It says explicitly "However, the cause of this association remains unclear." The discussion section speculates on possible causal links. That's what discussion sections do: take the new data and formulate hypotheses. Those hypotheses become the basis of future work. If you just want to know the data, skip the discussion. The data is in the "Results" section. The post's sarcastic summary appears to completely miss the point of the article: something is different about volleyball compared to other sports. That fact is interesting, and merits future investigation. So I think I'm missing what this post is about. The blog is titled "Statistical Modeling, Causal Inference, and Social Science". This paper strikes me as a bog-standard bit of social science. If it's not in keeping with other papers, the article is going to need to expand on that for me.

3 points by Tomte 2mo ago

Just to play Devil's advocate here, high-level volleyball players tend to be extremely tall. The average male pro/olympic player is anywhere from 6'6" to 6'10". Some are, of course, even taller. There are well-known correlations between height/weight and lifespan, see e.g.: https://pubmed.ncbi.nlm.nih.gov/1600586/ https://pubmed.ncbi.nlm.nih.gov/1600586/ So it's not prima facie a foolish notion that high-level (pro/olympic-tier) male volleyball players suffer from a ~5-year reduced lifespan relative to a much smaller population. I've seen papers that are a hell of a lot worse, that's for sure.

newsomix9xl 2mo ago

As volleyball is a sport for the disadvantaged, this critique is the definition of punching down.

dlcarrier 2mo ago

I feel like this is par for any random paper. Never trust any conclusion's take on correlation vs causation.

jfengel 2mo ago

The article is oddly devoid of specifics. It just calls it bad, without quoting any of the article. The "key takeaway" is not from the article. It's the post's author imputing correlation, into an article that resolutely uses the word "association" over and over. It says explicitly "However, the cause of this association remains unclear." The discussion section speculates on possible causal links. That's what discussion sections do: take the new data and formulate hypotheses. Those hypotheses become the basis of future work. If you just want to know the data, skip the discussion. The data is in the "Results" section. The post's sarcastic summary appears to completely miss the point of the article: something is different about volleyball compared to other sports. That fact is interesting, and merits future investigation. So I think I'm missing what this post is about. The blog is titled "Statistical Modeling, Causal Inference, and Social Science". This paper strikes me as a bog-standard bit of social science. If it's not in keeping with other papers, the article is going to need to expand on that for me.