Cross-posted at Education Week
In the more than a quarter of a century that our organization has been researching the world’s top-performing education systems, we have yet to see one that lacks a very strong instructional core. Conversely, we rarely see such strong instructional cores in countries that are not among the top performers. We take that to suggest that strong instructional cores are an essential component of top-performing education systems. But there is no state in the United States that has an instructional core as strong as those we see in the top-performing countries.
By instructional core, I mean the combination of state-mandated standards, curriculum frameworks, course syllabi, instructional materials, and tests and examinations that together define and measure what students are taught. When these systems are set to high standards, carefully aligned and well implemented, they are very powerful. The question I want to address in this blog is why the instructional core is so much weaker in the United States than in the top-performing countries. This is a non-trivial question. If we are right about the evidence, then no state in the United States will be able to match the performance of the top-performing countries unless it has an instructional core as strong as the instructional core we find in those countries.
Let’s start by looking at each piece of the instructional system. Until recently neither the individual states nor the United States had formal standards for student academic achievement. In the George H.W. Bush administration, that began to change when Diane Ravitch, then Assistant Secretary of Education for research, asked the subject matter societies to develop student achievement standards. That did not go as well as we all hoped, so we—NCEE—reached out to the University of Pittsburgh and put together a coalition of 21 states to put together a set of internationally benchmarked performance standards the states could use. That went very well until our state partners tried to implement those standards. They ran into a buzz saw from those on the left who saw the standards as a conspiracy to raise yet another barrier to the advancement of poor and minority students and from those on the right who saw them as usurpation of local control of the curriculum. Twenty years later, the Common Core State Standards effort, building on the base we had built with New Standards, developed another first-rate set of standards, also experienced considerable success and then ran into a buzz saw, this time mainly from those on the right who used the federal government’s support of the new standards to mobilize strong opposition to the Common Core. The Common Core may yet survive this onslaught, under whatever name, but the outcome is not yet clear.
With the exception of New York State, there is virtually no history of state-mandated curriculum in the United States, and considerable opposition to the idea. The notion of local control over the curriculum still has a lot of currency in this country. Indeed, as a practical matter, individual teachers have had a degree of control over what they teach that is unmatched in any of the other countries we have studied the world over.
In many countries, instructional materials are produced by the state and are, as a matter of policy, closely aligned with the state standards. In many others, they are produced by commercial publishers, but must be approved by the state on the basis of very clear and detailed guidelines. In some countries, the advisory boards for the commercially produced texts are chaired by top government officials to ensure they will be approved. In some countries, government inspectors regularly visit the schools to make sure the curriculum is completely aligned to the state standards and can shut down schools if they find that the curriculum and the materials are not aligned. None of this is true in the United States, of course.
Though some states subsidize the purchase of materials approved at the state level, as a practical matter, publishers publish what they wish and schools and districts in the United States buy what they wish. No state official is likely to come round later to shut down a school or a district for failure to align instructional materials with the state standards.
State-mandated assessment systems in the top-performing countries are not primarily designed as an integral part of their accountability systems. Their main purpose is to support high-quality instructional programs in the schools. They are typically much more expensive then their counterparts in the United States. That’s because they are dominated by essay prompts rather than multiple-choice questions, set to measure the extent to which the student has mastered the curriculum specified in the syllabi the state issues. Because they are essay-based, they have to be scored by human beings, which is what makes them so expensive. They are essay-based, for the same reason that our Advanced Placement exams count essays so heavily, because officials want to be able to test things that bubble tests cannot.
But there is more to it than that. The exam system design is a key element not just in finding out how well the students perform, but also in improving their performance. All the prompts are published after the exam results are published, and examples of the student performance earning high marks are also released. When the exams and the marked examples of responses earning high marks are published for all to see, they become part of the standards, making it possible for students, parents and teachers to get a visceral understanding of what student work needs to look like to get high grades. Tell a student that she needs to get a grade of 78 on the exam to get into university and you have told her nothing that will help her understand what she must do to reach her goal. Tell her that she has to produce work that looks like that and she has a clear target.
Let’s leave the standards aside, for the moment, on the assumption that most states will either end up embracing the Common Core by that name or some other name, substantially as written. It turns out that all the other elements of a modern instructional system, with the exception of the curriculum framework, are produced in our system by commercial publishers, and bought by states, districts and schools.
Some would have us believe that’s the problem. They say that we don’t have the instructional systems we need because of the perfidy, greed or incompetence of the publishers. I don’t believe that. The school market for instructional materials and tests is like any other market. It will produce what the market demands at the price the market is willing to pay.
So our question turns out to be: Is there a market for high quality, highly aligned instructional systems in the United States? If there is not, then there is no mystery why we don’t have such systems. The most widely used such system in the world, available in English, is the Cambridge International Examinations system, published by the University of Cambridge. It sells for about four times the cost of the systems about to be released by the two state testing consortia and many states are complaining that those will cost more than they can afford. There is a lot of evidence that the United States will not have high quality instructional systems because it is simply not prepared to pay what the rest of the world pays for them.
Bill Schmidt and others have shown that the materials that most of the publishing companies have released with stamps saying they are aligned to the Common Core are not aligned to the Common Core. No one doubts that the publishing companies could have produced materials better aligned to the Common Core. But it would have cost them money to do that. They chose not invest that money. My conclusion: They decided, based on long experience, that they could not recover that investment. Why? Because their customers would not distinguish between materials aligned with the Common Core from those not so aligned or because they did not really care enough about that to pay the higher price the companies would have had to charge. The evidence I just presented on the price resistance to better tests suggests that the publishing companies were right on.
Neither the test publishers nor the new state testing consortia are offering the kinds of heavily essay-based exams I just described. None of them will release all their question prompts after every test. None are prepared to release examples of high scoring student work with or without commentary. Why? I conclude that it is because there is no demand for high quality testing systems that are designed mainly for the purpose of improving student performance. The testing market has been completely dominated by a totally misguided belief that the best way to improve student performance is through test-based accountability.
My organization is looking for states interested in building a market for global-quality instructional systems. We will help you get there if you have an interest in doing that. You can reach me through our web site.