[lolinder]

And for every study that shows that there's another on the other side. Homeschool outcomes are incredibly diverse by nature and the numbers involved are so small that it's hard to have a study with any reasonable amount of power.

[seanmcdirmid]

Public education outcomes are pretty diverse as well, although there are good and bad schools (and home life status plays a lot in the results, which I assume is also true with homeschooling).

[lolinder]

True, but each individual school is big enough to have actual data associated with it. Conclusions about the system as a whole might not work, but you can make a reasonable guess about whether you want to send your kids to a specific school.

[seanmcdirmid]

Individual schools but we already know different schools vary greatly. As a form of selection bias, I bought my home in an area with schools where most kids go to college. The house was more expensive, so also needed to be richer, and it’s a two parent home, my neighbors are the same. It is t necessarily the school that’s good, it’s just selection bias all the way down.