Here's sort of what you wanted to read about: Rand Corporation RCT of "Restorative Justice" which includes measures of achievement.
"The most troubling thing: There were significant and substantial negative effects on math achievement for middle school students, black students, and students in schools that are predominantly black."
I have no idea how to solve the politics of this. But on the mechanics, what you want is possible, if tough. Back around 2005, our small charter school would admit 9th graders who had scored ~30th percentile on state math exams. We created a high dosage math tutoring program, staffed by recent college grads who lived in top floor of the school, converted to a dorm. Huge gains (later validated by UChi RCT). Several kids then did AP Calc as seniors, admittedly taught by best math teacher.
Here's sort of what you wanted to read about: Rand Corporation RCT of "Restorative Justice" which includes measures of achievement.
"The most troubling thing: There were significant and substantial negative effects on math achievement for middle school students, black students, and students in schools that are predominantly black."
https://www.manhattan-institute.org/html/school-discipline-reform-media
Note the finding was buried in the RCT document itself.
More here.
https://www.the74million.org/article/eden-studies-and-teachers-nationwide-say-school-discipline-reform-is-harming-students-academic-achievement-and-safety/
I have no idea how to solve the politics of this. But on the mechanics, what you want is possible, if tough. Back around 2005, our small charter school would admit 9th graders who had scored ~30th percentile on state math exams. We created a high dosage math tutoring program, staffed by recent college grads who lived in top floor of the school, converted to a dorm. Huge gains (later validated by UChi RCT). Several kids then did AP Calc as seniors, admittedly taught by best math teacher.