Sunday Commentary: More on What Can Obama Really Do?

A few weeks back I asked what Obama could really do to fix education in this country. While people have high hopes, the federal government has quite limited power when it comes to education. But I wanted to put this idea to the test by thinking through a few reforms that might be implemented at the federal level and what outcomes might result.

National Standards
If you believe half the buzz you hear in the blogosphere and in the press, national standards will be in place by the end of Obama's time in Washington. Precisely what the point of national standards would be without a national test I am not sure -- so I would have to assume that a national test would follow soon after. Assuming both are implemented, it would open the way for more accurate accountability testing. I think it is fair to assume that accountability based on more meaningful tests would also be more meaningful. Standards make it more likely that teachers will teach a certain topic, but I think policymakers often think they are of greater import to teachers than they actually are. More coherent standards and better tests would influence schools, but would not completely remake them. Besides, without expanding testing, the majority of teachers would continue to teach untested subjects.

Encourage Merit Pay
The federal government could do this in a number of ways, the easiest of which would be to expand the size and scope of the Teacher Incentive Fund. As of yet, we have little to no evidence that merit pay positively impacts teaching practices or student achievement (or that we can accurately measure teacher performance for that matter). But the results of the first randomized field trial should start trickling out over the next year, so this could change. For the sake of argument, let us assume that merit pay results in teachers working harder and students scoring higher. How much can the federal government afford to spend on bonuses? I have to believe that offering substantial bonuses to all of the top teachers in the country would cost at least $10 billion per year. Given that salaries are largely determined at the local level, it is easy to see this might place downward pressure on wages from local education agencies -- the union asks for a raise of $5,000 and the superintendent offers $3,000 and points out that good teachers can make another $10,000 in merit pay. So it remains possible that merit pay could significantly affect education, but I cannot envision it totally reshaping the system with any sort of feasible implementation -- particularly given that the majority of teachers teach untested subjects and we continue to rely largely on test scores for many of the largest experiments with merit pay.

Increase Title I Funding
This is intentionally vague. Title I monies are directed to many different projects -- from buying computers to paying companies to tutor students after school -- that a change in Title I funding is unlikely to yield earth-shattering results. Beyond stimulus monies designed to shore up troubled programs, it is hard to envision an increase in funds without a specific and targeted focus.

I have heard a number of other ideas bantered around as well, including but not limited to:
-lengthening the school day
-lengthening the school year
-reducing class sizes
-increasing the number of charter schools
-offering vouchers
-increasing teacher pay

But accomplishing any of these goals is complicated because they involve superceding local rules and regulations. And none are simple to accomplish. Duncan cannot simply mandate more charter schools and watch them magically appear, for example -- he has to pull on the right policy levers.

I could list off a thousand things that Obama or Duncan might do, but any policy the federal government puts in place will affect student learning indirectly. Changing the way a teacher teaches is incredibly difficult to do from that far away, and so is changing the amount that a child learns. As a result, any changes made at the federal level are likely to result in only minimal changes. In a best case scenario a number of policies would yield a number of small changes that would accrue over time . . . resulting in significant improvement a decade or so later.