3 ms·

Critiquing this article's approximation of a field-biology study would be a great exercise for young science students. I envision a roomful of kids shouting "wh

Critiquing this article's approximation of a field-biology study would be a great exercise for young science students. I envision a roomful of kids shouting "where is your control group"? I know nothing of mosquitoes and I haven't had my coffee and yet I can reel off three alternative hypotheses that need to be checked, from the prosaic ("experimenter cannot reliably identify viable mosquito larvae", "water temperature was incorrect and mosquitoes died") to the less prosaic ("plastic box leaches chemicals into water which poisons mosquitoes"; "a tiny frog, perhaps attracted by the alluring sound of this buzzing box, visited the pond and took the opportunity to snack on mosquito larvae". Oops, that was four. I'll get my coffee now. Also, citations. Just one, pretty please? Surely a thousand people have published on this problem before. What was wrong with the version of this approach that they probably tried back in 1953?

jamesaguilar 12y ago
God forbid someone should sketch out a prototype and share about it before running a multi-million dollar RCT, right?
- foxylad 12y ago
  You're right we shouldn't shoot this down in flames, but I think most of us would consider evidence with a sample size of one as un-evidence. Providing better evidence will not take a multi-million dollar RCT - all you'd need is ten ponds and a fortnight of checking them once a day. If that went well, you would have the confidence to push this harder, and maybe get a local university biology department involved to carry out a better designed study. And if that produced good results, you'd be ready to knock on the Gates Foundation door.