On January 25, 2019, the CDC released a report on school homicide trends.
Characteristics of School-Associated Youth Homicides — United States, 1994–2018
The overall 22-year trend for single-victim homicide rates did not change significantly. However, multiple-victim incidence rates increased significantly from July 2009 to June 2018.
Overall, media reports were solely relied upon for coding demographic and circumstantial details for 80 (18.6%) of 431 incidents.
Sounds bad, doesn’t it? But…
First off, their own Table 2 gives different incident numbers. In text, they claim 431. Table 2 says says 393 single-victim incidents and 33 multi-victim incidents. That’s only 426 total incidents. So right from the start, I take issue with their “data.”
In fact, eyeballing that chart, single-victim incidents/deaths… appear to trend downward, from roughly .052/100K to around .034/100K; a difference of .018/100K. That’s a drop of almost 35%, which strikes me as significant.
Multi-victim incidents — over the entire period — have increased (eyeballing again) from the neighborhood of .004/100K to .010/100K.
Deaths in multi-victim incidents definitely seem to have increased significantly, from around .002/100K to .048/100K.
Now allow me to explain why that is absolutely meaningless.
By their own admission, at least 47.42% of these school incidents did not occur at school or at a school-related event. I have to guess at the number of non-school incidents. They tell us that there were 33 total multi-victims incidents. They tell us 10 people died in non-school multi-victim incidents. But they never tell us. How many non-school incidents there were. To be “multi-victim,” I’ll assume a minimum of 2, so there may have been 5 multi-victim non-school incidents. Or it could be as low as 2 incidents. Either way, nearly half the “school homicides” weren’t at school.
By their own admission, 210 people –40.85% — of the victims were not killed at school or at a school-related event.
I was hoping to find a breakout of weapons type by incident location. They don’t do that. So I went looking at their references to see if I could find the data myself. But when I saw this…
2016–17 and 2017–18 multiple-victim cases were identified through manual Internet searches using phrases such as “school shooting” and “multiple victims and school,” as well as supplementary review of web-based firearm injury data sets (i.e., Everytown for Gun Safety and the Gun Violence Archive) to identify cases matching the School-Associated Violent Death Surveillance System case definition for multiple-victim youth homicides.
Everytown for Gun Safety. Regular readers may recall that I like to play find the school shooting with Everytown’s alleged school shootings list. When I started almost a year ago, only 71% of their “school shootings” were in fact school shootings by their own definition. The last time I checked, only 40% of the listing met their own definition. At least one “shooting” hadn’t occurred at all; anywhere.
Since the CDC compiled their “school-related” homicide data from a source which gets that wrong 60% of the time, I am forced to assume that the numbers I get from this report are worse than we thought. Nearly half of these researchers’ school incidents aren’t. And the one’s that they do claim happened at school… come from sources that invent fake listings.
Dear Bog: a fake report based on fake stats based on fake news. This error-ridden CDC report on school homicides really needs to be retracted. Their discussion cites numbers that differ from their own tables, they claim one trend is insignificant when it’s actually a 35% reduction (by their own numbers), nearly half their “school-related” incidents aren’t school-related by their own data, and they used a source known to have an error rate of as much as 60%. We paid Holland $106,380 a year for this garbage?
There was a reason for the Dickey Amendment, and we obviously need to keep the CDC out of this.
(More Tip Jar Options)
Ed. note: This commentary appeared first in TZP’s weekly email alert. If you would like to be among the first to see new commentary (as well as to get notice of new polls and recaps of recent posts), please sign up for our alert list. (See sidebar or, if you’re on a mobile device, scroll down). Be sure to respond when you receive your activation email!