4 ms·

> And they are concentrated in certain types of places (e.g., poor urban ghetto) such that the parts of the country that aren't these places have vanishingly sm

by est31 6y ago

> And they are concentrated in certain types of places (e.g., poor urban ghetto) such that the parts of the country that aren't these places have vanishingly small probabilities of being hurt by a shooting. I said "many places" not "all places". Indeed, shootings are concentrated to a few areas with high crime rates. Doesn't mean that people living in those areas don't have to watch out. I replied to an anecdote about people living close to such a tavern having to watch out. As for gun murders being a rare death cause, indeed if you average over all age ranges they are. However, for younger people they are a major concern. CDC data [0] lists homicides in the top 4 reasons for death for age groups from 1 until 35. Which weapons are used for these homicides? A different statistic tells us most times it's guns [1]. And yes, focusing on falls, car accidents and diabetes would be awesome. But preventing gun deaths is important too. [0]: https://www.cdc.gov/injury/wisqars/pdf/leading_causes_of_death_by_age_group_2016-508.pdf https://www.cdc.gov/injury/wisqars/pdf/leading_causes_of_dea... [1]: 6.4k murder victims in 2019 from handguns, 3.3k from unspecified firearms, 1.5k from knives or cutting instruments, all other causes below 1k. https://www.statista.com/statistics/195325/murder-victims-in-the-us-by-weapon-used/ https://www.statista.com/statistics/195325/murder-victims-in...

NoImmatureAdHom 6y ago
> I said "many places" not "all places"... Doesn't mean that people living in those areas don't have to watch out. I agree that there is some non-zero number of people in the U.S. who have to watch out for gun murder. The point I'm trying to make is that they make up a really small proportion of the population. Anecdotes are fun. > As for gun murders being a rare death cause, indeed if you average over all age ranges they are. However, for younger people they are a major concern. CDC data [0] lists homicides in the top 4 reasons for death for age groups from 1 until 35. Which weapons are used for these homicides? A different statistic tells us most times it's guns [1]. You're right that, among untimely deaths, gun murders move up the ranking if you're looking at young people. But this misses the main thing, which is that untimely deaths of any kind are super rare among young people, even rarer than overall. Why should we care if gun murders make up 50% of untimely deaths among young people if the likelihood of untimely death is sufficiently low? Your line of reasoning seems to indicate a 10% chance of untimely death with 1 percentage point of that being gun murder is preferable to a 1% chance of untimely death with 0.5 percentage points of that being gun murder. As an aside, thinking about this (and many other things) in terms of ordinal rankings will often lead you down the wrong path. "Top four reasons" tells us nothing at all. If there are 5,000 categories constructed and they are roughly evenly distributed, "top four" tells us nothing. If 99% of the probability mass is in #1, "top four" tells us nothing. It's like the classic "The U.S. is [embarassingly] #44 in math achievement". This means nothing if the top 80 countries are tightly clustered around 98 math achievement points. > But preventing gun deaths is important too. My point is that it isn't important. There are much, much bigger fish to fry from a public health perspective. The marginal benefit of throwing one dollar at "gun violence" is << the marginal benefit of throwing that same buck at something like trying to get construction workers to wear proper safety harnesses.
- WalterBright 6y ago
  Statistically, for kids, the dominant causes of death are car accidents and drowning. The rest you can pretty much not worry about. Another way to look at it is if you read about it in the paper, it is rare enough you don't need to worry about it.
- est31 6y ago
  > As an aside, thinking about this (and many other things) in terms of ordinal rankings will often lead you down the wrong path. "Top four reasons" tells us nothing at all. If there are 5,000 categories constructed and they are roughly evenly distributed, "top four" tells us nothing. If 99% of the probability mass is in #1, "top four" tells us nothing. That is correct, but please note that it's still a significant component. In the 15-24 age group, the top 1 cause has 13.4k deaths, the top 2 has 5.7k, and homicide has 5.1k. That's not very far away in relative terms.
  - NoImmatureAdHom 6y ago
    > but please note that it's still a significant component. I guess my overall point is it's not important. Even if the ordinal ranking is informative in this case (your argument), the overall likelihood of any kind of death is so low that we shouldn't care very much about the ranking.
- chefandy 6y ago
  Getting hit with a bullet isn't the only risk of living in an area with community violence. We're not robots who can coolly base our response to danger on a statistical risk assessment. Living with community violence has serious psychological consequences which result in significant, measurable problems for those who endure it... particularly children. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2700237/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2700237/ Unsurprisingly, the communities who most frequently endure violence are not a representative sample of the US population— they are drastically more likely to be poor and not white. Black communities are particularly subject to this violence, and considering that discrimination against them was not only legal, but legally mandated until around a decade before the first millennials were born, the dominant culture within the US has a moral obligation to address this violence because it's responsible for its creation. https://www.ajpmonline.org/article/S0749-3797(18)31907-X/fulltext https://www.ajpmonline.org/article/S0749-3797(18)31907-X/ful...

by est31 6y ago

NoImmatureAdHom 6y ago

> I said "many places" not "all places"... Doesn't mean that people living in those areas don't have to watch out. I agree that there is some non-zero number of people in the U.S. who have to watch out for gun murder. The point I'm trying to make is that they make up a really small proportion of the population. Anecdotes are fun. > As for gun murders being a rare death cause, indeed if you average over all age ranges they are. However, for younger people they are a major concern. CDC data [0] lists homicides in the top 4 reasons for death for age groups from 1 until 35. Which weapons are used for these homicides? A different statistic tells us most times it's guns [1]. You're right that, among untimely deaths, gun murders move up the ranking if you're looking at young people. But this misses the main thing, which is that untimely deaths of any kind are super rare among young people, even rarer than overall. Why should we care if gun murders make up 50% of untimely deaths among young people if the likelihood of untimely death is sufficiently low? Your line of reasoning seems to indicate a 10% chance of untimely death with 1 percentage point of that being gun murder is preferable to a 1% chance of untimely death with 0.5 percentage points of that being gun murder. As an aside, thinking about this (and many other things) in terms of ordinal rankings will often lead you down the wrong path. "Top four reasons" tells us nothing at all. If there are 5,000 categories constructed and they are roughly evenly distributed, "top four" tells us nothing. If 99% of the probability mass is in #1, "top four" tells us nothing. It's like the classic "The U.S. is [embarassingly] #44 in math achievement". This means nothing if the top 80 countries are tightly clustered around 98 math achievement points. > But preventing gun deaths is important too. My point is that it isn't important. There are much, much bigger fish to fry from a public health perspective. The marginal benefit of throwing one dollar at "gun violence" is << the marginal benefit of throwing that same buck at something like trying to get construction workers to wear proper safety harnesses.

WalterBright 6y ago
Statistically, for kids, the dominant causes of death are car accidents and drowning. The rest you can pretty much not worry about. Another way to look at it is if you read about it in the paper, it is rare enough you don't need to worry about it.
est31 6y ago
> As an aside, thinking about this (and many other things) in terms of ordinal rankings will often lead you down the wrong path. "Top four reasons" tells us nothing at all. If there are 5,000 categories constructed and they are roughly evenly distributed, "top four" tells us nothing. If 99% of the probability mass is in #1, "top four" tells us nothing. That is correct, but please note that it's still a significant component. In the 15-24 age group, the top 1 cause has 13.4k deaths, the top 2 has 5.7k, and homicide has 5.1k. That's not very far away in relative terms.
- NoImmatureAdHom 6y ago
  > but please note that it's still a significant component. I guess my overall point is it's not important. Even if the ordinal ranking is informative in this case (your argument), the overall likelihood of any kind of death is so low that we shouldn't care very much about the ranking.
chefandy 6y ago
Getting hit with a bullet isn't the only risk of living in an area with community violence. We're not robots who can coolly base our response to danger on a statistical risk assessment. Living with community violence has serious psychological consequences which result in significant, measurable problems for those who endure it... particularly children. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2700237/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2700237/ Unsurprisingly, the communities who most frequently endure violence are not a representative sample of the US population— they are drastically more likely to be poor and not white. Black communities are particularly subject to this violence, and considering that discrimination against them was not only legal, but legally mandated until around a decade before the first millennials were born, the dominant culture within the US has a moral obligation to address this violence because it's responsible for its creation. https://www.ajpmonline.org/article/S0749-3797(18)31907-X/fulltext https://www.ajpmonline.org/article/S0749-3797(18)31907-X/ful...