It was anecdotal because your example started as a story about your kids. The statistics you quote here are pretty much meaningless, so it doesn't change a thin

It was anecdotal because your example started as a story about your kids. The statistics you quote here are pretty much meaningless, so it doesn't change a thing. Let me explain why. If the SA schools perform consistently higher, and assuming for a school to be "SA" involves a bit more then starting at 7.45, then you need to compare the performance of SA schools that start at 7.45 with the performance of SA schools that start at 8.30. There aren't any? Then there is no way to tell. Not via statistics, at least. Your comparison of statistics changes way too many variables to make any conclusion. There's simply no way to tell if an SA school might not perform even better if they started later. Especially since the statistics you quote are not objective but "top X", which if they're already near the top, you couldn't measure improvement even if it was there. Similarly, without quoting the proper statistics, there is also no way to tell if the schools starting at 8.30 might not perform worse if you changed nothing but starting them at 7.45. The numbers you quote simply do not say anything about this. (also you're never going to prove anything about an effect in highschoolers by quoting irrelevant statistics about elementary schools) Either way, your numbers don't prove what you are claiming, yet you still hold to this claim. If your claim is not anecdotal, then it is simply wrong.