You can't create a successful politics of support for public education on the basis of asking voters not to care about skills, and tests that measure skills, at all.

One of the most consistent findings about low performing schools and students is that "home variables" (parental income and education, etc.) are more predictive than "school variables." But, having said that, we as a society can have much more effect on the school variables than on the home variables, so it's important and valuable to focus on the question of which interventions in schools are most effective and which are least effective.

I think serious research tends to be associated with higher academic quality, more prestige, more resources, and even, heaven help us, better teaching, to a greater extent than you might think. Folks who don't have an active intellectual life become, though the long years of just teaching, less intellectually alive and exciting.

Profound question is how do you measure the non-skills component of what goes on in schools: values, curiosity, critical thinking, and so on. That's very tough. Maybe everything worthwhile can't be measured.

You can't create a successful politics of support for public education on the basis of asking voters not to care about skills, and tests that measure skills, at all.

In the current environment, attributing low student performance to teacher tenure is one of the great unproven causal links out there. The relationship just hasn't been examined very carefully, but we should all recognize that in higher education the strongest institutions generally have the most robust tenure systems, and in elementary and secondary, the states with the strongest teacher unions (and tenure systems) tend to have the highest student performance.

In journalism, especially, we tend to deal with large, complex systems by finding especially interesting people and story lines to focus on.