[napoleond]

1 in 20 kids see someone get shot every year, so you would expect that by age seven that 7 in 20 would have

No I wouldn't, because:

a) I doubt the age distribution for "kids who see people get shot in a year" is constant. The linked article doesn't include age distributions, but they surveyed children up to age 17, and I would expect many more 15 year-olds to see people get shot than 7 year-olds.

b) I doubt the system is stochastic. It's probably more likely for the 1 in 20 from year 1 to see another few of the 7 shots in those 7 years (due to geographic and socio-economic factors) than it is for the other 19.

In other words, even for a country where gun violence is rampant, a study population where 35% of children below 7 have seen someone get shot is far from average.

[Alex3917]

Good points. In terms of lifetime prevalence, that study says that, "Similarly, 3.5 percent of 2- to 5-year-olds had witnessed a shooting during their lifetimes, whereas more than one in five 14- to 17-year-olds (22.2 percent) had witnessed a shooting."

So the population in question is definitely no where near average, my was just that strictly in terms of the amount of violence witnessed, the differences probably aren't as great as one would otherwise assume.

[rlanday]

What the hell!? One in five teenagers has witnessed someone being shot? I must live a sheltered life…