4 ms·

I think you're violently agreeing: my point was that "dozens" or "hundreds" of times that it goes wrong, across the entire US, even if they're all from the same

I think you're violently agreeing: my point was that "dozens" or "hundreds" of times that it goes wrong, across the entire US, even if they're all from the same year is a statistical fluke we likely can do nothing about, and does not reflect any underlying problem with the emergency services. This is in direct reply to the person above me essentially making the argument "Well, yes, but I can show you exceptions!" Naturally, the first thing we should do is look at the rate of exceptions and see if that is reasonable for the process that we're talking about. I was pointing out that even low balling the number of calls (a good way to get an estimate on the scale of a situation to reason about it), the number of cases the person was offering to present was such a small fraction, we could never get below that threshold of times things went weird. The reason for low balling the numbers in my estimate is that it made the resulting figures biased towards favoring the argument being made by the other person. If when the number start biased in their favor and still don't support their argument, it's reasonable to use an estimate to reject their claim. The main thrust of my argument was essentially: the number of cases the person above my post offered to present was so small relative to the number of cases happening that they're mere anecdote, not data representing any underlying problem. I would encourage you to understand what estimates are and how they're being used in arguments before making comments about numbers being nonsense or simply referring to them as made up. It simply makes you come across as mathematically ignorant.

rgrieselhuber 12y ago
The point is that you can do something about these "flukes" if you, as an individual, don't only rely on timely response by emergency services.
- ObviousScience 12y ago
  Again, of course there are cases for which knowing CPR or other training makes a difference - an obvious one is having basic first aid training around pools, particularly in the case of diving accidents leading to spinal injuries. Knowing the proper procedure to stabilize the person until paramedics can arrive on the scene does a lot to lessen the chance of secondary damage (eg, from the being moved wrong while trying to keep them from drowning). I don't think that anyone is arguing that having more people know CPR or first aid is bad, but rather, the number of cases cited were too small to even infer that kind of information from: it's quite possible that there's nothing that could have been done for many of the cases the paramedics were late to respond without actual medical equipment onsite, ie, having more than just bystanders who know CPR or basic first aid. The gap in size between the cited number of cases and the number of cases that happen is too great, especially considering that the cited number of cases were from a biased selection, eg, only cases that matched a particular profile. It's important that we not let emotion or our desire to do something about tragedies cloud our vision of what's actually helpful and what things actually matter. I would expect for every case cited to me where someone saved a life being a bystander, I could find one where bystanders with basic training and good intentions ended up making the situation worse. That is, of course, beside the point, because both collections of anecdotes would be just that, and not actual data about what's a good idea.

rgrieselhuber 12y ago

The point is that you can do something about these "flukes" if you, as an individual, don't only rely on timely response by emergency services.

ObviousScience 12y ago
Again, of course there are cases for which knowing CPR or other training makes a difference - an obvious one is having basic first aid training around pools, particularly in the case of diving accidents leading to spinal injuries. Knowing the proper procedure to stabilize the person until paramedics can arrive on the scene does a lot to lessen the chance of secondary damage (eg, from the being moved wrong while trying to keep them from drowning). I don't think that anyone is arguing that having more people know CPR or first aid is bad, but rather, the number of cases cited were too small to even infer that kind of information from: it's quite possible that there's nothing that could have been done for many of the cases the paramedics were late to respond without actual medical equipment onsite, ie, having more than just bystanders who know CPR or basic first aid. The gap in size between the cited number of cases and the number of cases that happen is too great, especially considering that the cited number of cases were from a biased selection, eg, only cases that matched a particular profile. It's important that we not let emotion or our desire to do something about tragedies cloud our vision of what's actually helpful and what things actually matter. I would expect for every case cited to me where someone saved a life being a bystander, I could find one where bystanders with basic training and good intentions ended up making the situation worse. That is, of course, beside the point, because both collections of anecdotes would be just that, and not actual data about what's a good idea.