Class size

David Farrar writes at Kiwiblog that

[...] regardless it backs my view that the impact of smaller class sizes is minimal, unless it is a massive difference. In other words a size of 15 will make a big difference compared to 30, but a class of 25 compared to a class of 27 will not.

The following few pages come from the book Mostly Harmless Econometrics: An Empiricist's Companion by Joshua D. Angrist and Jorn-Steffen Pischke, Princeton: Princeton University Press, 2009 and show that research does in fact indicate that class size does matter for educational outcomes. It also makes clear just how problematic getting good results can be. (Enlarge to make more readable.)

Mostly Harmless