By All Measures: 'A Very American Way:' Everybody's Getting Into the Act
Gregory R. Anrig is the president of the Educational Testing Service, the largest testing firm in the country.
Primarily known for college-level and professional tests, the åŸôŸóŸ, under Mr. Anrig's leadership, has taken strides in precollegiate assessment and in alternative forms of assessment. Since 1983, it has administered the National Assessment of Educational Progress, a federally funded project that tests a national sample of students on a range of subjects.
The firm is also one of four companies working with the California Department of Education to develop that state's new performance-based assessment program, and it has joined with the College Board to develop Pacesetter, a new project to provide courses, examinations, and teacher-development materials for high schools.
MR. ANRIG: It would be a mistake to think that national standards and assessments are synonymous with the New Standards Project. I think of the New Standards Project as one of the most promising projects going on in the field of national assessment and national standards. But there are others. We need to keep this in context so that it is clear there is not one proposal everyone is either accepting or supporting.
It is important that we are not going about this whole process as Japan would do, as Germany would do, or as any other European or non-American country would do. We're doing it in a very American way. Everybody's getting into the act. There are some limitations and problems with that, but it is also healthy and worth preserving.
There are a lot of flowers blooming at this point. I don't view them as competitive, and I hope they don't become that. Nor should they become a monopoly. It's very important to have a lot of people working on the issue of common standards; common standards are at the heart of what we're talking about.
This is a very dynamic, healthy, and promising period in educational reform. It started out with the White House advisory group focusing on the idea of a national test. That was a terrible idea. I'm very happy that it's no longer a prevalent idea.
There has now developed a consensus around this idea of a national or nationwide assessment system. The reason that consensus exists is nobody can define it, and everybody feels that it's what they're doing. And that's great; that's American, and we ought to keep it that way. We ought to have a good time doing this, and it ought to be something that invigorates people and makes them feel there's a chance for them to be part of it, rather than something that is exclusive to anybody, anywhere.
What is important is that the emphasis at last has shifted toward what is being taught and what is being learned. Education reform didn't begin this year. It didn't begin with "A Nation at Risk'' in 1983. It actually started back in the early 1970's, when parents began to feel their kids weren't learning well enough for the world that awaited them. So they put pressure on. It was the back-to-basics movement.
Well, the basics are back. The trouble is, the basics are no longer good enough. And the parents are again saying, "My kids are not learning enough for the world that awaits them.'' And they're right. The whole purpose of reform has been to help kids learn more for the world that awaits them.
Up to now, we've tried to accomplish that in a number of ways. The decade of the 1970's was a period of testing. Thirty-eight states had mandatory tests in place by the end of the 1970's. The decade of the 1980's was sort of the "more'' period--more courses, more requirements, more tests, more teacher requirements, more dollars for teachers' salaries.
At last in the 1990's, we're getting down to what the parents were originally concerned about--what happens in the classroom.
The reason standards-setting is so important is that it provides an opportunity we have not had up to now for some coherence between what is expected of the schools, what is taught in the schools, what is assessed in the schools, and, hopefully, how teachers are prepared and developed for the schools. That's why we need national standards. You can't have that coherence if standards are decided in 15,000 different school districts.
Today, a number of disciplinary groups are working on common standards. The history group is now funded. Work is under way in civics, English, science, geography, and, most recently, the arts. The governors didn't mention the arts in the national goals, but the arts people have come together--music, the visual arts, the performing arts, dance--and they have agreed together to develop standards. Secretary Alexander, to his credit, has funded them to do that.
The National Council of Teachers of Mathematics was perhaps the first to define common standards. They did the job very well, but they already think they should change it. And that's good; that's part of the dynamism going on. It's not carved in concrete. It's not Moses coming down from the mountain with his tablets of standards for K-4 and for 5-8 and for 9-12. Standards are going to change, and they ought to. This is going to be a dynamic period, typical of America.
In terms of coherence, there's another part of the equation that hasn't received much attention. What is the family's responsibility? We should be concerned about not only how teachers are prepared for this, but also how families are. The policy-information center at the Educational Testing Service is about to come out with a publication called "America's Smallest School: The Family.'' It includes a lot of data that are just beginning to get organized in an interesting way, that show how important the family is in this whole mix of things.
I agree with Ted Sizer's comment on high-stakes testing. When I was in the Army, the order was: "If it moves, salute it. If it stands still, paint it.'' Now if it stands still, we say, "Test it.''
We are putting a lot of dollars and a lot of hope on tests. But I don't know of a responsible person or responsible organization that works with tests that says tests should be the sole determining criterion for high-stakes decisions. They should be used in combination with other information, recognizing their limitations, recognizing the information they can provide. But they must be used together with other types of information.
In the Encyclopedia of Educational Evaluation, the word "assessment'' is defined not as a test, but in terms of getting information from a variety of sources about a variety of things that can help one make a better decision. That's what we ought to be talking about. And if we start tagging a single test on a single day to such high stakes as promotion, graduation, and in some states financial incentives and penalties on teachers and schools, then we are putting a very important human decision on a very thin reed. That's bad policy.
Here's an example. Until recently, one state had a 90-minute test for kindergarten promotion. I don't know what you'd do with a 5-year-old for 90 minutes that would be useful in making a decision about promotion. After a whole year for a teacher and a parent and a principal to make a judgment about that child, the decision to promote rested on the result of a 90-minute test. That wasn't only an unwise policy, it was unfair to kids.
It is also distressing that we are so obsessed with accountability, just obsessed. That's why there is so much testing. Most of it is for accountability purposes.
Accountability should not be defined as synonymous with tests. If we want information about how schools are doing, then let's get information, not simply scores from a single test.
We all agree there has to be accountability. We need accountability to help us judge whether our investment in schools is being wisely spent. The question is: How do you achieve it? A single test on a single day is not going to do that.
Rick Mills, the commissioner of education in Vermont, has a terrific idea. To provide accountability in Vermont, they're going to do two things: First, keep portfolios during the year of students' works, including test scores and examples of their actual work in mathematics; and, second, they're going to have town meetings in the school where parents and the public can see the work, go through those portfolios, interview the teachers, speak to the students, and reach a judgment.
Surely, that is a better form of accountability than looking in the local weekly paper for the test score for your school and finding it's a 4.8, whatever that means. We've got to go about looking at assessment and accountability in a much broader way than we do now. It's time for us to get off of our distorting obsession with accountability.
If we were going to design a test for instructional assessment, for instructional improvement, we would not design an accountability test. And yet, that is what we're using.
It won't be easy to change that. The kind of "thinking curriculum'' and assessment being proposed by the New Standards Project could lead to student improvement, but it's very hard to translate that into terms that legislators and the public can identify with. Imagine going before the joint education committee of a legislature and saying, "I've got this stack of papers that the kids did awfully good on.'' They'd say, "That's fine; what's the number?'' People want a quick answer, and, unfortunately, they often get a quick and dirty answer.
There is a widely held concept that testing can drive instructional improvement and better learning. I believe that using accountability tests to drive educational reform is reckless driving. It won't work, and it hasn't worked. If we were able to improve instruction and learning with such tests, we wouldn't have the mediocre achievement that we're faced with today. Remember now, in additional to all of the tests that are now administered at the school level, 38 states by the beginning of 1980 had mandatory testing programs. By 1990, 47 states had them. And, as Ted Sizer correctly points out, there has been some improvement at the bottom, particularly for black and Hispanic children. But there has been no improvement at the higher levels that we're concerned about, and the parents are concerned about. If tests could drive reform, we would have had it already. And to think that tests are a quick answer to this is misleading the public and misleading ourselves and misusing tests.
Another problem is that too much of the debate is ideological: All existing testing is bad; all new testing is good; all performance assessment is good; all multiple-choice testing is bad. That is not very useful.
Instead, we need to ask what it is we're trying to do with tests and what's the most efficient way to do it--in terms of not only dollar costs but instructional-time costs. A Texas superintendent testified at a hearing that a month of the year in Texas is spent preparing for or administering tests. That is one month of a nine-month instructional year. That's too much. That just doesn't make sense. So, when we talk about costs, we can't just think in terms of dollar costs; we must also think in terms of instructional time.
If we want tests for accountability, we need to ask: What's the best way to design such a test, given the costs, the realities, and the instructional time allowed? That's not the kind of assessment we would design to help kids learn more and help teachers teach better.
Just a word about the federal role. I got a call the other day from a national magazine reporter asking me who would be the best education president--George Bush or Bill Clinton. And I said, "Frankly, I don't think that's a very important question, because, while I don't want them to do anything bad, what they do really isn't very important. What's really important is what's been happening outside of Washington.''
When I was in Massachusetts, I figured out the federal government provided 4 percent of the cost of elementary and secondary education; I gave 4 percent of my attention as commissioner of education to the federal government. It worked out fine. We need to keep this in perspective. Where reform is going to take place, where our standards are going to be set, where better assessments are going to be developed, is not in Washington, D.C., but in Pittsburgh, Princeton, New York City, and in many other places that are working on these tough questions.
One last word on what has been most overlooked during this whole period: the social compact Marc referred to. Governor Romer uses that wonderful analogy of the sandwich--the top of the sandwich is the goals and standards, the bottom is assessment, and in between is all that you do to help the teachers and students succeed.
I'm reminded of the old advertising slogan: Where's the beef? There's nothing much in between. If all we have are goals, standards, and assessments to measure them, and nothing in between, we are doing a real disservice to the children and teachers in this country. Not enough attention is being given to that. And until it is, this effort can't be thought of as very serious or productive.
You can't turn around an organization and have it do better by telling it what it's not doing, or what it can't do. Instead, you have to give it hope, tell it what it can do, what it will do, what you will help it do. After a lifetime of working with kids, I don't believe you can get a child to do better by telling him what he can't do, or telling her what she can't do. Instead, you say, "You can do this, we'll help you.'' That's what this period of reform must be: a can-do, will-do effort, rather than simply putting new bars up there for kids and teachers not to be able to pass.
It may sound strange coming from me, but I think we're putting too
much attention on the testing part of this and not enough attention on
the opportunity side. Until we do, we're not going to have very good