3 ms·

> "Proponent aversion would be if you denied that the rate of suicides has been growing because Obama ... had made that claim" Another example might be gun/cri

> "Proponent aversion would be if you denied that the rate of suicides has been growing because Obama ... had made that claim" Another example might be gun/crime statistics from the NRA vs gun/crime statistics from gun-control advocates. There's the inherent suspicion that the statistics had been twisted to fit an agenda, because you've caught that proponent or someone from the same ideological group doing it before, and that therefore the statistics themselves are untrustworthy. I apply a mental correction factor when dealing with a group I think lies for agenda purposes -- I figure they're spinning things maximally in their favor, so what they claim forms one end of my error bars. Sometimes my expectations regarding the size of the error bars in that field lead me to believe that it's likely the data is actually centered all the way on the other side. In that sense, their statements can imply their own contradiction in some circumstances.

hga 11y ago
Here, though, most of the raw data comes from the FBI/DOJ, so pretty much everyone accepts it (even if some number of homicides are scored as "accidents", as happened in my home town) and can easily go back to it; the same is not true for many many other domains like the climate. But the statistics are indeed critical, or must be viewed critically. E.g. the Kellerman study that said "N times more likely to be killed with a gun in the house!!! than use it to kill an intruder" is utterly bogus from the start, because gun owners don't score kills as the only type of successful self-defense uses. That sort of critical thought, though, isn't particularly fashionable....
- lotharbot 11y ago
  even the raw data is subject to questions, two of which you point out: (1) is everything being categorized correctly, and (2) are we counting the right thing?

hga 11y ago

Here, though, most of the raw data comes from the FBI/DOJ, so pretty much everyone accepts it (even if some number of homicides are scored as "accidents", as happened in my home town) and can easily go back to it; the same is not true for many many other domains like the climate. But the statistics are indeed critical, or must be viewed critically. E.g. the Kellerman study that said "N times more likely to be killed with a gun in the house!!! than use it to kill an intruder" is utterly bogus from the start, because gun owners don't score kills as the only type of successful self-defense uses. That sort of critical thought, though, isn't particularly fashionable....

lotharbot 11y ago
even the raw data is subject to questions, two of which you point out: (1) is everything being categorized correctly, and (2) are we counting the right thing?