I suppose if you're comparing 2018 to 2008, then sure. But that seems cherry-picked, since it looks like shootings were more common before then. Then you also have to consider that the population has been growing, so per-capita shootings might be down? And then there's the question of whether body-count is going up. I think combining all of these factors makes the data-set very noisy. I'm sure that OP has the best intentions and likely believes that schools are more dangerous than 10 years ago, but I don't think the data bares that out.
I think what OP said is pretty clearly correct. If you find a data point contradicting it, please share, but eyeballing what I think it'd be even per capita, I think OP is still right. If you chose a different data point than 10 years, maybe it'd show something else, but OP said 10 years so it seems fair to judge on the basis of what was said.
But certainly school safety is a bigger issue than just shootings.
My own emphasis, personally, would be on mental health issues being on the rise for young people.
Wikipedia suffers very much from recency bias in reporting.
Keep in mind that the site was founded in 2001, was growing rapidly through the 2000s, and (along with much the rest of the Internet) saw a huge boost in use and access with the spread of mobile devices.
Inconsistency in reporting is a general problem with crime (and other human activities), with reporting rates rising and falling for numerous reasons. But a standardised crime reporting database would be preferable.
Do you have a source for this?