The standard academic failure mode is to make a number of assumptions for tractability that severely lower the relevance of the results (and the more pernicious failure mode is to hide those assumptions).
Perhaps, but at least these assumptions are stated. Most work leans on similarly strong assumptions (for tractability, brevity, or lack of rigour meaning you don't even realise you are doing it) but doesn't state them.
Really good to see some concrete suggestions and experiences, thank you for sharing
Are you claiming that there are other deceptive statements in this post?
I agree that this is an unfair claim but the post wouldn't lose much force without it
Often recruiting is the bottleneck in biomedicine so you want to maximise the power for a given number of participants
Looks like everyone except the two confirmed cases (father and daughter) tested negative.
I am >98% certain this was due to a shared exposure to an infected bird rather than human-to-human transmission (base rate + no meaningful evidence otherwise). Similarly, at no point did I think there was a >5% chance this would turn out to be human-to-human transmission.
I would recommend that we do not post H5N1 cases without evidence of mammal-to-mammal transmission or very large clusters. We should expect to see a lot and they should not meaningfully update our view of the situation.
Owen should have told the org this situation is unacceptable and that they should arrange accommodation at their own expense.
Thanks for the response. I think this position is completely reasonable, and I'll repeat that I (currently) do not think there's any cover-up or the like here. But I think at this point, we've both acknowledged that it would be easy to misconstrue actions as a coverup, which is also a problem IMO.
Someone doing something similar to the community contact points at CEA but more clearly separated. Much like the people you list in the post as others who have volunteered to be named as contacts but are doing this voluntarily.
Have you considered A/B testing changes? As you note, looking at engagement numbers before/after isn't capable of assigning causality.