News for hackers. Stuff that matters.
Re:Depression and suicide linked to air pollution in new glo... by AstralStorm on Thursday December 19, 2019 @05:39AM (#21833668)
Problem being of estimating NNT to see the effect. If the effect size is small you will need huge numbers to detect it

Plus unfiltered air is elsewhere, as a confounder.

Re:Depression and suicide linked to air pollution in new glo... by dhshahsndeisjwn on Thursday December 19, 2019 @05:45AM (#21833711)
This is my hobby horse! A confounder would be an unobserved (unmeasured) variable that influences both the allocation to treatment and the outcome. In this case unfiltered air elsewhere is contaminating the magnitude of the treatment effect (in a conservative direction) but wouldn't undermine the validity of a properly conducted experiment.