I have a new post over at Policy Shop. It concludes thusly:
On average, SAT scores in all subjects track household income very closely: the poorer you are the lower you score, the richer you are the higher you score. Ultimately, it is only by changing that disparity that we can hope to make a serious dent in the class composition of higher education. That is going to mean directly tackling income inequality and poverty so that poor children are not behind their wealthier peers when college time rolls around. Finding ways to help the rare poor kids that manage to do well despite their impoverished childhoods is a worthwhile endeavor of course, but it will not address the big drivers of higher education inequality, and should not be seen as anything more than a very tiny part of a larger solution.