to be fair, other countries also have listed as school shootings any event that involved shooting and a school; eg: Australia with a population ~ 15x smaller than the US:
https://en.wikipedia.org/wiki/List_of_school_shootings_in_Au...
has a total of six (6) incidents this century (since 2000) .. somewhat less than a factor of 15 less than the US for that same 26 year period.
Note that three of the six incidents of the past 26 years involved guns being fired but no one being hit (no grazes, injuries, or deaths).
I think the connotation of a school shooting is somebody going to a school and shooting numerous students in a mostly random fashion with the intention of creating terror. That's why guns are so scary in this context - it's trivial to kill a person with a gun, a knife, a car, or a brick, but it's much easier to kill 10 people with a gun than it is with the other instruments - well except perhaps a car.
But when you aren't listing just these sort school shootings, but instead listing any homicide that occurred within the vicinity of a school, why is it reasonable to exclude the various incidents of non-gun homicides at Australian [1][2] (or American for that matter) schools? It just feels like a false narrative. Because what matters if we're just speaking of safety at schools is how often people are killed within the vicinity of schools, though obviously subdata including the share of each weapon in homicides would be useful/informative.
[1] - https://en.wikipedia.org/wiki/Murder_of_Lilie_James
[2] - https://www.dailymail.com/news/article-15912963/Father-dies-...