There would be no reason to assume that, given the statistic as written. I’m certainly not reading anything but at the plainest level, which includes suicides, accidental discharges, justified homicides, etc.
I knew someone in high school who was shot and killed, accidently, by his best friend when they were fooling around with a hunting rifle at home.
I think the Brady Campaign statistics, which are believeable to me, are to point out the level of gun violence in our country. The fact is that the US has by far, of all developed nations, more guns per person and more gun violence. So it is completely reasonable that 1 in 3 people have been affectected by gun violence and 22% of teenagers have witnessed or been affected by gun violence.
The sampling procdure for this study is detailed in this document. I’m not really a statistician, but it appears that they chose telephone exchanges to get a proportionate geographic coverage. As a final step, they weighted the responses by race, ethnicity, and income level to match census data. And, as other people have pointed out, the effect of sampling a small percentage of the population is pretty well understood. I don’t think critcism of their sampling methodology is really all that valid.
However, it looks like the plain English of the survey has quietly morphed. The original report (published in Pediatrics) specifies that 22.2% of 14- to 17-year-olds have had an “exposure to shooting,” in their lifetime. That’s defined, apparently, by answering afirmatively to: “At any time in your life, were you in any place in real life where you could see or hear people being shot?” The same authors talk about “witnessing or hearing a shooting” in the DOJ publication, but drop the “or hearing” qualifier when discussing the actual numbers (“more than one in five 14- to 17-year-olds (22.2%) had witnessed a shooting”). The Brady Campaign cites the DOJ report, but shifts “more than one in five” to “nearly one in four” and, oddly, “14- to 17-year-olds” to “teens.”
So we have “22.2% of 14- to 17-year-olds have, at some time in their life, been in a place where they could see or hear people being shot” morphing into “nearly one in four American teens have witnessed a shooting.” That seems problematic to me, mostly because of “seeing or hearing” turning into “witnessing,” but also because of “*could *see or hear” turning into “*did *see or hear.”
I don’t think the “could”/“did” distinction plays into this. I can’t see somebody being surveyed answering “could you see the shooting” vs “did you see the shooting” any differently. If I didn’t see it, I would answer both “no.” If I did see it, I would answer both “yes.” I doubt even a fraction of a percent is answering the question in the sense of “did you have the ability to see the shooting, but didn’t” manner, if that’s what you’re getting at.
The first part though, yes. By their definition, I am one of these 14-to-17 year olds. I don’t think you can live in a major US city (at least not Chicago) and not hear gunshots at some point. But it’s not “witnessing” in the way I would normally think of it, although the statement may be true by a strict interpretation. Why that nearly 1-in-4 claim bothers me, but the 1-in-3 doesn’t, I couldn’t put my finger on it.
No, but then nobody expects a single number to provide a meaningful depiction of anything. That’s why there are lots and lots of numbers on the Brady site.
Self reported “data” is useless. I’d be better persuaded by a calculation using the number of Americans shot, how many people the average American knows, and calculating the likelihood that someone you know has been shot. Saying “sure I know somebody” on a phone survey doesn’t convince me of much.
Your math isn’t just fuzzy. Your underlying statistics are fuzzy too.
First, there are over 100,000 fatal and non-fatal shootings a year. Second, any rational definition of the term “teenager” has to include more than the four years you are apparently using, and there are only 82 million Americans under the age of 20 (page 4).
It is a plausibility test. It won’t tell you the exact answer because that isn’t the point but it will tell you that the number specified isn’t even within the ballpark even when all of the assumptions are on the Brady Campaign’s side.
Let me phase it again using your numbers (the vast majority of which are accidental shootings and suicides).
Let’s assume the U.S. is composed of nothing but people under 20 (82,000,000 of them). There are 100,000 shootings of any type every year.
Therefore, there will be about 2,000,000 shootings in the U.S. between the time a person is born and the day they turn 20.
That means that 1 in 41 people are shot in some way in this worst case scenario. 40 out of 41 are not shot but some are witnesses to another person’s shooting. That requires that each shooting have an average of 10 personal witnesses with no overlap between the witnesses to have 1 in 4 people as witnesses to a shooting including accidental shootings and suicides.
Does that sound reasonable to you? The actual numbers will be many times lower of course because these are the worst case assumptions all in the Brady Campaign’s favor and it still doesn’t pass the plausibility test.
It didn’t sound reasonable before. I was merely pointing out that your numbers were way off.
My first example was the same as the second except I roughly doubled the number of shootings to also account for all accidents based on your cite and reduced the people under 20 count from 90 to 82 million (mine was just a rough estimate). The second one was just phrased differently to be more clear. The first one included everyone under 20 as well, not just current teenagers.
The numbers for the claim that 1 in 4 teenagers has witnessed a shooting just don’t work in any way, shape or form. There simply aren’t enough shooting to go around for that to be true. I think we can declare that claim firmly busted just based on known facts. The Brady Campaign shouldn’t use it because it makes them look either manipulative or stupid.
I can actually believe the claim that 1 in 3 people know someone that has been shot. I know more than a few myself. That is completely different claim and only requires a very small percentage of the population who have been shot to make it true because each one will typically know many people and there is no “witnessing” requirement.
Also notice that those claims don’t match with each other. Are we supposed to believe that 1 in 4 people witness someone getting shot by the time they are 20 but the numbers only go up to 1 in 3 if we take out the witnessing requirement and expand it to everyone they know for a person’s entire lifetime?
They’re making two different points; they don’t need to match.
The first is about the current generation of teenagers (not “people in general, by the time they’re 20”), and that 1 in 4 of them have “witnessed” a shooting – note that it may have been a stranger being shot, or, as someone noted earlier upon reading the underlying study, “witness” appears to include hearing a shooting, at which point, you likely have no idea who got shot.
The second is about the broader population, and refers to actual knowledge of a shooting victim.
One possible implication which one might take out of the two numbers (and which I imagine the Brady group would like you to take away) is that younger people are more likely to have been exposed to gun violence than older generations have.
But there are two huge flaws with that line of thinking. This is aside from the very basic fact that there aren’t enough shootings to go around to have that many witnesses no matter what (run some remotely plausible numbers any way you want and be generous to the Brady Campaign; it does not work out to the point of impossibility).
The idea that hearing a gunshot is the equivalent of a shooting is completely flawed. If you hear gunfire don’t know that anyone has been shot at all. You may just live down the street from a skeet shooting range and didn’t realize it or someone may just be firing up into the air for whatever reason. It could be thousand reasons and teenagers have active imaginations. That isn’t “witnessing’” a shooting at all to me even if you knew for a fact that someone did get shot. Hearing gunfire alone does not equal someone getting shot in any case.
The other problem with your comments is that violent crime has been going down since the late mid-1990’s in the U.S. and the downward curve has been steep. That should be common knowledge to an informed person but I can give you plenty of cites if you wish. All current teenagers were born when violent crime was dropping to historically low levels in the U.S. compared to previous generations.