Hacker News new | past | comments | ask | show | jobs | submit login

I think this is an important question to answer, but the shehab paper (the first link) provides vertically no useful information. They didn't blind the assessors carrying out the cognitive studies, which on its own invalidates their conclusion. They also didn't randomize the ordering, but appear to have "investigated for order effects", found "no statistically robust difference", then proceeded to have every participant do the "post exposure" test second, making it highly likely that fatigue alone could explain their effects. With only 30 subjects, their tests for ordering effects are likely underpowered.

Also note that they did 11 different tests, but didn't appear to correct for multiple comparisons [1]

This would be a really interesting problem to study, but you'd probably want to look at academic performance in college students before and after transfer between universities e.g urban vs rural (presuming you can measure differences in pollution) between the two, you'd expect to see a relative drop in academic performance going from rural to urban, now that would worry me. But this study is unhelpful.

[1] https://en.m.wikipedia.org/wiki/Multiple_comparisons_problem




Just want to say it's impossible to tell which paper you're referring to.

The first link on the page isn't a paper. The word shehab doesn't appear anywhere on the page. The word shehab doesn't appear anywhere in the first paper linked. The three quoted sentences / fragments you provide don't appear anywhere in the first paper linked. I opened some of the papers and didn't see any with n=30.


Patrick quietly dropped the entire first paragraph in the examples list a few hours ago: https://web.archive.org/web/diff/20191118190006/201911190530...


Oh wow. I think that is a bad style. Once your article gets that discuss you should not simply delete an entire paragraph but rather comment on it...


Thank you.


If the effects are permanent then your results would be inconclusive if studying the same people transferring between polluted and non polluted universities.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: