> harmful in total (our results are presently worse than if there was no homework at those ages)
Most studies compare schools / classes assigning no homework to various "normal" levels of homework, so they imply that our results are presently worse than if there was no homework.
I do assign homework to middle school students. But I don't do it often, and it's typically of the form "think about an idea on the topic of _____ we can discuss" or "do the first couple items of this lab worksheet so that we can use our time in the classroom more efficiently tomorrow." I enforce its completion with social pressure, not the gradebook. "Dude, you were supposed to do the first couple questions on the lab worksheet!"
I have a feeling that there is better homework we could assign that would be useful. But we don't have a lot of evidence of what that would be. Proponents of flipped classrooms think that is the path forward, but the research quality is very dubious in primary education.