Hacker News new | ask | show | jobs
by mikeskim 3888 days ago
A highly biased sample of less than 300 is supposed to really say anything of any significance for the general population? I think not.
1 comments

And this is thoroughly discussed in the article. It's a sample of men from Harvard. However, it is a very long, in-depth study.

Just as you can argue that it can, at best, indicate correlation - and only apply to white Harvard men... it's also possible that the lessons learned are applicable to people outside that narrow scope. It's possible. But then it's also possible that the internal opinions, biases and struggles of Vaillant are bleeding into the conclusions as well.