We estimate an education production function in which attainment depends up
on parental inputs, peer group inputs, and schooling inputs. We find that p
arenting is much more important than schooling. The most powerful parental
input is parental interest in education for which OLS does not give upward
bias as Plowden (1967) suspected. We also find a strong peer group effect.
The school pupil-teacher ratio does not enter significantly. A simulation s
tudy of the properties of our estimators indicates that our conclusions are
robust.