School Killings and The Problem With Relative Numbers

The mysterious Revere looks at school killings today (or, more accurately, looks at a study looking at school killings). He/she/they opens with an arresting factoid:

The second leading cause of death in the 5 to 18 year old age group in the US is homicide. These are school aged children and the first thing that comes to mind are the big names like Columbine and Virginia Tech. But we know there are other school-related homicides that kill only one or two. Moreover there seem to be more of them than we remember from years past. But are there?

You may or may not be surprised to learn that school killings are not a huge part of this effect. But read the post for the details.

I just wanted to comment on that firt sentence, which plays into a pet peeve of mine. Hearing that homicide is the second-leading cause of death for children 5-18 certainly sounds scary, as it's supposed to. It creates an impression of some vast epidemic of child murder sweeping across the country like a biblical plague.

That's really a false impression, though, created by the use of relative numbers. This doesn't reflect a murder epidemic, it reflects a triumph of modern medicine.

I'm too lazy to do all that much looking for the exact number Revere used, but you can get the essential idea from the following graph from a CDC report showing death rates over time by age group:

i-dfd48433286a362192a72d3642d63e29-death_rates.jpg

That's the death rate per 100,000 people, plotted on a semi-log scale. The line at the very bottom is the 5-14 age group (not precisely the same group Revere cited a number for, but close enough). Boys that age die at a rate of about 20/100,000, and girls a bit less. That's a full order of magnitude down from my own 35-44 age group.

So, the answer to the question "Why is homicide the second-leading killer of children?" is "Because there aren't that many things that kill children." Something has to be the second-leading cause of death among the young, and murder is it. It's not that there are a lot of child murders-- the absolute number is quite low-- it's that there aren't many other things killing significant numbers of children.

And that's really a remarkable testament to the quality of modern medicine. If you look at the numbers on that graph, the child mortality rate has dropped by a factor of three since 1955, and if you look back even farther, I'm sure the effect becomes more dramatic. Diseases that used to regularly kill young children have been all but eradicated thanks to antibiotics, vaccination, and general improvements in medical technology.

We live in a remarkable world, relative to the rest of human history. The use of relative numbers for death rates in this case tends to obscure that fact, and create an alarming impression that just isn't true.

More like this

The second leading cause of death in the 5 to 18 year old age group in the US is homicide. These are school aged children and the first thing that comes to mind are the big names like Columbine and Virginia Tech. But we know there are other school-related homicides that kill only one or two.…
Lott has an article in the National Review Online where he claims that the Washingtonian DC handgun ban caused crime increases: Crime rose significantly after the gun ban went into effect. In the five years before Washington's ban in 1976, the murder rate fell from 37 to 27 per 100,…
Canada. Gun law in 78. Homicide rate (per 100,000 population) 74-78 2.7 3.1 2.9 3.0 2.8 average 2.9 79-83 2.7 2.5 2.6 2.7 2.7 average 2.6 (a t test on the statistical significance of the difference of the means gives p=.01) Thomas Grant Edwards said: From "Gun Control and Rates of Firearm…
I was curious to see what kind of defense Matt would put on against my suggestion of additional regulations to address the problem of gun violence and homicide in the US, and I was a bit disappointed to see that the response is largely a "no problem" argument. I had actually come into this debate…

I was in India over the holidays for a wedding, and one of the topics of discussion was a school shooting, apparently the first in India. (Two kids are bullied, one day snap and shoot bully.) All of the commentary I read inevitably included some line like "This is not America, where such shootings are commonplace." Nice to see the impression the rest of the world has of us.

Population control by killing schoolchildren is more efficient than wasting enormous social investments and then killing them as soldiers or storing them as prisoners. Contraception is a sin unsuited to delectation by the lower classes.

Chad: You are quite correct. The fact is that overall mortality in that age group is very small. The figure on age specific relative mortality was in the CDC MMWR but the actual contribution to overall mortality is small as a result of mortality in that age group being small. However these are preventable deaths and from the standpoint of potential years of life lost larger than they appear. We also attach greater importance to mortality in children. One can question whether a child's life is worth more than an older adults, but I assume the tendency to do so is hardwired into us. If we didn't have a reflex concern for the next generation we might not have survived as a species.

However the point that peeves you is quite correct, which I acknowledge.

Looking at the data, it seems that the dominate cause of death among this age group is accidents, specifically motor vehicle accidents (see Table 10, 10987 deaths in 2004 out of a total of 33421, as opposed to 5085 homicides).

I wonder what would happen if the driving age were made, say, 18? Or if the US attitude towards speeding were changed? (yea, I know, good luck with that....)

By Brad Holden (not verified) on 21 Jan 2008 #permalink

If you look at the numbers on that graph, the child mortality rate has dropped by a factor of three since 1955,

And, don't forget, your country's rate is still the highest in the developed world, by a significant factor.

Child mortality in the USA *could* be much lower than it is now. It's just that allowing access to basic necessities of life is not an American priority.

Uncle Al,
I'm not sure I can take that link with much more than a grain of salt. The author, Mike Adams, seems somewhat cooky in many ways. He may well be right but he doesn't cite any of his claims and looking through his list of cartoons he seems very anti-pharma pro-herbal. He comes off very biased.

"see also the US infant mortality rate, which is 2-3x the rest of the developed world."

That's because the US calculates infant mortality in a different way than other countries do. Children considered alive at birth by the United States are considered dead in other countries.

see also the US infant mortality rate, which is 2-3x the rest of the developed world.

That's also because we a diverse country with many different ethnic groups, some of which have lower average birth weights.

By BillBodell (not verified) on 22 Jan 2008 #permalink

wonder what would happen if the driving age were made, say, 18? Or if the US attitude towards speeding were changed? (yea, I know, good luck with that....)

Imagine the lives that would be saved if we outlawed driving altogether!

By BillBodell (not verified) on 22 Jan 2008 #permalink

That's also because we a diverse country with many different ethnic groups, some of which have lower average birth weights.

Dude, whether a baby is born with 4 kg or 3 kg or even 2 doesn't make a difference. That's the kind of span in my own family (all gleaming white, all living in the same country in Europe).

By David MarjanoviÄ, OM (not verified) on 23 Jan 2008 #permalink

That's because the US calculates infant mortality in a different way than other countries do. Children considered alive at birth by the United States are considered dead in other countries.

Are there so many of those?

By David MarjanoviÄ (not verified) on 23 Jan 2008 #permalink

This reminds me of my common comment to Cancer Conspiracy Theorists:

The reason cancer is so much more common now isn't radiation, it isn't pollution, though both of those may contribute.

The reason cancer is so much more common is: PEOPLE LIVE LONG ENOUGH TO DIE OF IT.