(Guest post by Matthew Ladner)
In the opening scene of Saving Private Ryan, Captain John H. Miller (played by Tom Hanks) leads a group of American soldiers in storming the beach at Normandy. Pinned down behind inadequate cover and facing extremely heavy enemy fire, Miller orders his men to move out, prompting one of his soldiers to ask where they should go.
Miller bellows to his men: “ANYWHERE BUT HERE!”
Facing the task of reauthorizing No Child Left Behind (NCLB), Congress also faces an unsustainable status-quo. Although fashioned with the noblest intentions, NCLB created a perverse incentive for states to lower their academic standards — an incentive that will become increasingly powerful in coming years.
The law reflects contradictory urges regarding education policy. On the one hand, some want Congress to act to improve education, and on the other, some wish to preserve the tradition of state and local control of schools.
In NCLB, Congress attempted to finesse this contradiction, but failed to do so successfully. The law requires states to test almost all students, and to have an ever-increasing percentage of them reach proficiency in all tests by 2014. However, NCLB leaves the content and the passing thresholds of these tests to the states.
You don’t need a Ph.D. in game theory to see the problem with this, just a little common sense. As “proficiency” requirements have risen, states have begun to dummy down their tests to avoid federal sanctions.
Congress must now make an actual choice about which level of government should predominate in education policy — or the price will be very high.
Scholars have already noted the beginnings of a “race to the bottom,” as states lower passing thresholds and otherwise make the tests easier to pass.
Congress has set up a looming train-wreck in public schools. If allowed to play out, every child in America may “pass” state proficiency exams by signing their names to completely meaningless tests. American schools would return to the dark age in which no one has meaningful data concerning school outputs. Parents, teachers, administrators, policymakers and taxpayers will be flying blind on school reform. The racial achievement gap will be closed- but only as a complete fraud.
Congress can either become far more intrusive in the setting of state education policy, or let the states lead. Federal education policy must move in one direction or the other, but cannot remain stationary.
Senator Barry Goldwater opposed the first bill to provide federal funds to public schools, which was the antecedent of NCLB. Goldwater’s warning rings prophetic: “Federal aid to education invariably means federal control of education.”
Evidence regarding the effectiveness of federal aid to education has remained scant through the years. In 2004, Secretary of Education Rod Paige bluntly noted “We have spent $125 billion on Title I programs for disadvantaged students in the past 25 years, yet we have virtually nothing to show for it.”
Today, the total federal share of the K-12 education budget remains under 10 percent of the total K-12 budget, but serves as a vehicle for a huge number of federal mandates on schools. Fully 41 percent of the administrative costs for state education bureaucracies are spent on complying with federal mandates, the General Accounting Office estimates.
Despite this record, some see the race to the bottom as an opportunity to expand federal control over local schools. Some have begun to make the case for “national standards.” The logic is simple: states can’t lower standards that they don’t control.
We need to move precisely in the opposite direction. Above all else, first the federal government should do no harm in education policy. The mess of NCLB inspires no confidence in the ability of Congress to fashion standards, even if it were constitutionally appropriate, which it decidedly is not. With a good deal more modesty, the federal government might take on the role of maintaining transparency as a condition of federal funds, rather than undermining it.
The states have begun to function as labs of reform on education. States like Florida and Massachusetts have made a good deal of progress in recent years, far beyond anything which can be attributed to NCLB. Ironically enough, NCLB threatens this progress by all but ordering states to dummy down their tests- or else.
Jay P. Greene's Blog 
As always, Matt makes a cogent argument, but I think he overstates the problem here. The incentive to dumb-down tests or standards is not unique to NCLB. It is a risk under any accountability regime. And while we can cite instances of dumbing-down, the overall trend under accountability reforms has been to gradually raise standards. The theoretical risks that Matt describes have largely not appeared in practice.
Yes, NCLB has various flaws, but so do most practical alternatives. Rather than going “anywhere but here,” it is most likely and probably wisest for us to muddle through, fixing someobvious flaws while accepting other defects.
Jay is correct that there is always pressure to dummy down accountability tests, whether the federal government contributes to that pressure or not. The sad fact however is that the federal government is adding to that pressure in a spectacular fashion. I don’t agree that these risks have not appeared in practice, and if you look at the AMO objectives, the improvements are all backloaded to the 2011-2014 period.
In other words, the worst is yet to come.
Jay seems willing to fix this flaw, but the administration and Congress do not. To date, none of the major reauthorization proposals have even admitted that this problem exists.
Dumbing down has clearly occured in some places. A few months ago Indiana schools were bragging about how their pass rates were up. Not long afterward, I had occasion to look a little more closely at Indiana’s scores for an unrelated analysis. I found that while the pass rates were up, the scale scores were flat.
But it’s not clear to what extent the dumbing down is attributable to NCLB. I talked to an old hand in Indiana testing politics, and he gave me the full twenty-year story of how standards are constantly being set and dumbed down, set and dumbed down.
And any dumbing down has to be weighed against the rise in standards elsewhere. I would say the most important rise in standards is the simple imposition of transparency. Before NCLB, many states engaged in no statewide standardized testing, or did not make the results available in a useful format. Every one of those states has to be counted as having increased, not decreased, the rigor of its school accountability, because the mere presence of a quantitative standard creates transparency about results. And I’m sure some states that already did testing are doing it better now. Plus, consider that we’re about to get our first national standard for reporting graduation rates, under the auspices of NCLB.
As for backloading, isn’t that one of the key ways useful things get done in Washington? You do what you really want to do now, and to get people on board and/or sell the policy to the press, you make impossible promises whose fulfilment is backloaded to Never-Never Land. In this case, the imposition of transparency is the real policy, and the promise that all students will become proficient is the bait. I expect that when the time comes, the AMO objectives will be quitely kicked down the road another decade.
Suppose we never get those magical AMO gains. How is that an argument against NCLB? The real story of NCLB is that the public schools got a big cash payoff in exchange for submitting to quantitative measurement of their results, allowing all sorts of research to be done on school performance that couldn’t be done before, leading directly to public accountability for results through simple transparency, even if the formal sanctions associated with NCLB never bite. It seems like a good deal to me.
Greg’s final question assumes that the race to the bottom scenario doesn’t develop. Currently, the clear escape hatch to the AMO ramp up that will start in the next year or so is to lower your cut scores. No one- neither George Miller nor the administration-has put forward a proposal to do anything about this problem.
I’m not against testing and transparency. Far from it. It is crystal clear however that the federal government can not do anything to stop states from dummying down, but have in fact provided a huge incentive to lower standards.
My final question does not assume that the race to the bottom scenario doesn’t develop. One of the points I was trying to make, and I guess I wasn’t entirely clear about this, is that the mere presence of a standardized test is valuable even if the cut scores are politicized; the scale scores will still be available and independent researchers can use them to produce reliable information.
Even the mere collection of demographic data mandated by NCLB is vaulable. In my recent study of Florida vouchers, I used NCLB data for demographic variables, because it was the only up-to-date source of data on the number of LEP students in each school.
I agree with you that transparency is good, and even that some states have become more transparent because of NCLB. My concern is that absent some serious revision in the law, that will not remain the case.
Where’s the empirical evidence for Matthew’s statements? Without some significant data to support this comment, the point hangs on a very slim thread.
[…] more cogent critics complain that it creates incentives to dumb down the proficiency standard until everyone is […]
Captain John H. Miller (played by Tom Hanks) DID NOT Say this in Saving Private Ryan.. You need to watch this film again!
Unless someone has been tampering with the film’s audio on youtube, he did say this: