[0xff00ffee]

On one hand: YAY SCIENCE AND DATA!

On the other hand: Ahhhh, figs.

The sample size is incredibly small (n=35), but it appears to be a well-constructed & controlled study. This merits further investigation. And that's exactly what the abstract ends with.

My husband and I switched to soy milk about 15 years ago. Anecdote, he had his testosterone tested recently and it was fairly high [he's completely bald], I have not had mine tested ever. We like soy milk because we get the raw stuff without the gross thickeners like xantham gum or carageenan, everything else seems to add that (oat, rice, hemp, almond, etc.).

But data is data.

Hopefully this isn't one of those "un-reproducable studies" but it was done by the NIH and not a university.

[snapetom]

Larger sample sizes increase the possibility of Type II errors. Too small versus too large is a common line researchers walk. n=35 is completely reasonable and common in this type of study.

[0xff00ffee]

Really? Huh. I thought confidence interval mattered which is proportional to sample size, e.g. bigger is always better.

So more samples is worse for a study like this?

Seems counter-intuitive, but I'm not a stats guy.

[snapetom]

For things like polling, yes. Bigger is better. But scientific studies deal with the question "does this have an effect." In that aspect, you introduce Type I and Type II errors, and sample sizes play into the chances of making them.