4 ms·

Base Rate Fallacy on science.org ? https://en.wikipedia.org/wiki/Base_rate_fallacy https://en.wikipedia.org/wiki/Base_rate_fallacy

• neonate 5y ago
  That page explains the fallacy poorly. Like a lot of Wikipedia articles on technical topics, it gives an explanation that only makes sense to people who already know the concept (or at least I assume it does). For the rest of us, maybe someone could provide a clearer explanation?
  • SideburnsOfDoom 5y ago
    I think this is the same as my earlier comment here https://news.ycombinator.com/item?id=28273947 https://news.ycombinator.com/item?id=28273947 Quote: --- This number is meaningless without also knowing what percentage of the people are vaccinated. Compare: "80% of the people in hospital have $foo, and 80% of the general population has $foo" - $foo has no bearing at all on hospitalisation. "80% of the people in hospital have $spon, and 5% of the general population has $spon" - $spon is clearly linked to the majority of hospitalisations. "45% of the people in hospital have $bar, and 5% of the general population has $bar" - $bar is linked to higher rates of hospitalisation, _despite_ not being the majority of hospitalisations. Spoiler: for some age groups in the UK now, "being unvaccinated" is much like $bar. --- I would add now: "not $bar -" i.e. "being vaccinated" could make up "55% of the people in hospital" (or more) despite reducing the odds of it. and "not wearing seatbelt" -> "in hospital after a car crash" is probably like the $spon case. Those examples above of course come with the caveat "all other things being equal" which isn't the case if, e.g. that others have raised, the age breakdown of the whole population vs hospitalisation is different. That's why there are graphs of "over 50s, vaccinated vs. hospitalised" as a catogory.