[bsder]

Allocate the same teacher/student ratio and randomize the students in a public school and at a private Montessori and get back to me with the data.

Most private schools do better only because they weed out the expensive problems and force them back to the public system. The moment you randomize the students, the private schools drop back to the mean (or, generally, worse).

I find this unfortunate, because education is in dire need of some real, evidence-based, advances. We have a lot of new data about achievement and learning.

However, putting it to practice requires money, time and a LOT of effort. And you will have to fight the parents, too.

[dragonwriter]

> Most private schools do better only because they weed out the expensive problems and force them back to the public system.

The single biggest factor is that private schools, as non-default choices, automatically filter for parental engagement in education, even before considering the filters they put in place in terms of admissions criteria.

Students with parents engaged in their education do better.

[frabbit]

> Students with parents engaged in their education do better

IMHO this is the crucial thing. And it is one of the aspects of Montessori that some parents who are simply aspirational-consumers are a bit thrown by initially.

[frabbit]

Private Montessori schools often have a much higher pupil/teacher ratio than public schools: the age-mixed structure of the class which encourages children more adept in particular tasks to work with those who are less adept makes this metric less important.

[humanrebar]

> Most private schools do better only because they weed out the expensive problems and force them back to the public system.

This happens with public schools except the mechanism is raising housing prices until problematic people aren't even in the neighborhood.