Right now, some states seem to be [backing off from accountability]. Virginia originally set a 2004 deadline [for schools to pass its test]. They moved it to 2007. Some states have lowered the bar. Do you worry that states may be wavering?

What we know about this from experience in Texas is that it is tough stuff, and it's not tidy at all. A lot of bloodletting, a lot of pressure points. ...

I'm reasonably certain some states are [wavering], because it is politically tricky, and requires great political will. It requires some real powerful forces in there to keep things on track, and I'm not sure that's embedded in every situation. So we are going to have states that are going to waver. And what we're going to try to do is support those, and try to help as much as we can.

You said "politically tricky." What does that mean?

The idea of public education, and public education law as promulgated by our state legislatures, is very fragmented. There are so many different ideas about what should happen. It'll be difficult to find ten people to agree on almost anything. So because of the wide differences in the perceptions of how things should proceed, it creates a chaotic situation politically.

So it's easier to back off?

Yes, it's easier to take the road that is the least tricky, like a lot of compromise. But in this particular case, there are some truisms, and there are some ideas that appear, and we must stay the course whether they are popular or not. For example, accountability. So a lot of compromises here may not take us to the place where we need to go.

So you're saying, "Stay the course." ...

Stay the course. Absolutely, take the pain, stay the course.

You say, "Take the pain." Suppose a state puts a test in place for graduation and 50 percent of the kids cannot pass and therefore cannot graduate -- should they take the pain?

What we've got now is a situation where we can make several choices. One choice is we can say, "We'll deal with the test. Fifty percent of the people are not passing, so it's got to be the test." Another possibility may be that 50 percent of the people who are not passing have not been taught well, so let's look at the curriculum, let's look at the instructional programs. So there's lots of ways to handle that.

But how would you handle that? Fifty percent -- wait a minute, my son's not going to get a diploma? He's been in school 12 years.

OK, we assume that we're talking about a test that is valid and is in line with the state standards that are adopted. ... If we can take that assumption, and that the test is valid and reliable and objective, we'll look somewhere for the problem. And the problem probably is in the instructions.

Do you worry that we'll end up blaming the victim, blaming the kids? "Well, they didn't learn it"?

Let's say if they don't do well on the test, then we can conclude that they didn't learn it. Now, who should we blame for that, whether it's them or whether it's instructional circumstances? I believe that the school can make the difference, and so that's where I would look for the responsibility. This does not say that the community, the parents, the children -- all of them -- don't have some responsibility there. But I think the school is the organization that has to cause all of these other things to happen.

Why do you and the president feel that we need mandatory testing of all kids in grades 3 through 8 in both math and reading?

For several reasons. The first reason is, we have to know, with some reliable valid source, whether or not students are learning. And so far we believe the best way to do that is to take the standards that were adopted by each state and have tests constructed that measure the extent to which these standards have been or have not been achieved. That is why the test is necessary to answer the question, "Are children learning relative to the standard?"

But your proposal says that the state can choose the test and the state can decide what constitutes passing. Doesn't that make it a little bit questionable?

No. In our country, we made that decision when the Constitution was drawn up. This is a state responsibility. This isn't a federal responsibility, to set standards for states. So that argument's already been settled. So what we want to do is operate within the framework that we have, which is 50 independent states with the independence to develop the standards that they see would be good for that state.

Some of the critics have said, "Sure, but then the state's just going to say, "Hey, Secretary Paige, we passed. We did it. Take our word for it."

I think that's almost an insult to the states. That's to say that these guys can't be trusted, "They're crooks and they want to trick the federal government." I think every state wants the best for their children. Some are stronger than others as far as an academic situation is concerned, but I think all of them want the best for their people. And I think, with some assistance from the U.S. Department [of Education] and other agencies, that they can be trusted to do what's best for their children.

It raises the question of too much testing. American kids are already tested more often than kids in any other industrialized country. Are we testing our kids too much?

I don't know the answer to that question, because I don't know exactly how much testing is going on. But I would suggest the testing that we're talking about is to first set the standard: determine what each child should know and think and experience and accomplish and be able to do. Once the state has decided that, then the next logical question is, "Are they learning to do that?" And the only way you can answer that is to develop some assessment system to give you the answer to that question.

Now, if you have other tests going on, maybe that's where you should look to relieve the testing overload, if it exists. But there's one type of test we think is a necessary condition. That testing is that test which is aligned with the state standard and answers the questions for the state's people how much progress has been made by each individual student towards accomplishing the state's predetermined standard.

Is Washington willing to pay for all this?

Washington is willing to help, as we've helped before, even before we made this suggestion. But this is a part of the teaching responsibility that each state has. ... Washington has offered some assistance now. In the legislation, we have ... some support to pay for the development of tests. But even if that should be looked at as a gift, it is the state responsibility to do this.

I asked a test maker about the "Bush-Paige plan," let's call it. ... And she said, "It's not going to happen. We do not have the capability of scoring, producing reliable good reports in a timely manner if that plan goes through. The resources aren't there." So if someone in testing [says that], who's committed to testing, to help a child instructionally, it's a terrifying idea. ...

You want to know what I have for her? "Watch." That's it. "Watch." It's going to happen.

She says the test makers can't do it, [that] they don't have the resources.

It's going to get done -- because it is the right thing. See, I make this observation. Why is there such a strong defense of the status quo when the status quo is failing us? What about the other ideas? What solution are they bringing to the table, these detractors? Are they satisfied with the progress we're making now?

There are wonderful people doing wonderful work, but the whole country's not moving forward at the rate that we would like. And so what we've proposed is a system to put a real boost in the educational delivery system in our country, which we're going to need in order to keep our leadership in the world. So all these naysayers, we'd like to have their support, and we'd like for them to re-think their position.

But your message is, "Watch what we do. Watch."

We're committed to doing this, because it is the right thing to do.

One of the phrases that keeps coming up, in terms of doing it right, is "disaggregating the data." Would you explain what that means?

Absolutely. Most of what has happened in the past has happened as a result of our dealing with averages. Let's take a school, and we're going to say the average score at this school is X. That doesn't tell us anything about the individual child. First, we want to disaggregate the data by the different groups in the school. Maybe we want to look at the disadvantaged population, might look at the Anglo population, African-American population, Hispanic population, Asian population. We want all these different groups.

And then we want to disaggregate it even deeper. We want to disaggregate it all the way down to the individual child.

Why?

Because we're teaching children; we're not teaching schools. For the school to be making progress doesn't tell us enough about what the individual children are doing.

So when you find out that one group's doing well and one group's not, what do you do?

I'll just give you an example. ... Let's take this school in a major city -- and this is a real example -- this school in a major city is a premier school. It is a school that has 40 or 50 National Merit scholars every year. It's a school of the first order, no matter how you measure it.

And it has very high average performance for the school. But when you disaggregate the data, you find out that the Hispanic students in the school are not doing well. So they are, in effect, left behind. Or the African-American students in the school are not doing well. You won't know this until you disaggregate the data, past the average, to this population. Now that you know this, you have a responsibility now to do something about this.

So that's the idea. Are there some sanctions here?

... Sometimes people move based on standards. Others might need sanctions to prompt them, to make them go. But I believe that once you know that there's a problem here, and it has been identified, I think most people want to go and do something about that. Now, for those who don't, then we might have to have sanctions to make them, to make a little movement there.

You've been an educator all your life. Business is playing a big part in this. Do you ever worry that business is too much in control of education reform?

In the first place, I don't think business is in control of education reform. And second, I think that business has an absolute responsibility to be involved. They are a part of the public, and the school system is there for the public. It is to serve the public benefit. Those who have children in the schools also derive private benefit from the school system. But the school system is there for the public good and to benefit the public, of which the business community is a very important constituent base.

Superintendents seem to talk the language of business. [We interviewed] Mark Edwards in Henrico County [Virginia]. He talks about how "We're running this, we're looking at investments and we're looking at the bottom line." Is the language of business appropriate for education -- test scores as the bottom line?

There was a period in time when businesses were not doing very well. ... And they went to the social psychology literature, they went to the literature of organizational dynamics and organizational behavior, and they found methods of doings things, which really converts to the question of, "How do you arrange for the human beings in the organization to be more productive?" So they benefited from this. ...

Now we say, "This is business." It isn't actually business. This is social psychology, and how you create more dynamic movement in the organization that is pushed forward by people. So I don't consider this to be business practices. I consider this to be good practices based on the discipline of organizational behavior.

That's interesting, because Mark Edwards talks, and others talk about, you set the goal, you provide the resources, and then you measure to see if you're getting there, and you take action -- which is sort of a business model of organization.

But business went and got this model from that discipline. So I don't think business owns it. Business learned it. So the question is, how do you make organizations more productive? That's the question we need to get the answer to. And if that answer is the same answer that business got, it doesn't mean that that is a business answer. It is a social psychology, organizational behavior, organizational theory answer.

What's bottom line in the business of education?

The bottom line is the amount of learning that we get on the part of the students. Student achievement is the only purpose for which these systems exist. That's the bottom line; that's what matters. The transportation system gets students to school, the food system makes sure they're nourished, and the regulatory system makes sure we do things right. But the bottom line is, are the students learning?

Are the tests we use an adequate measure of whether kids are really learning?

Yes, it is. ... You use the word "adequate." So the question would be, are the tests adequate to measure the extent to which students are learning, against the standards that are being set? That speaks to the quality of the tests. Is this a valid test or is it a reliable test? Is it an objective test?

We know there are poor tests out there. We know that there's improper use of tests out there. But those are things that we're going to work against and try to get fixed. The bottom line is, we have to have some system that gives us the information that we need, which is the degree to which the goals, the standards, are being achieved by the students.

You've heard the stories. And I've seen it where, a month before the test, regular instruction stops and teachers start drilling kids in test-taking and so on and so forth. Is this good practice?

No, I don't think that's good practice. First of all, I think the test and the content that's being taught should be the same thing. It shouldn't be that this is a set of content that's going to be on the test, and this is something that we're going to be teaching during the regular school year. These should be the same thing. So then when you're teaching to the test, you're teaching what should be tested in the first place when you made your decision about the standards. The test should reflect that.

Now, should students learn things other than what's on the test? Of course. So this is a minimum evaluation system. "This is what we must expect -- we require this." But we want more.

If you reflect a little bit on the high-stakes movement, there are at least 19 states now which have endorsed a graduation test. "You must pass this or you do not graduate." And there are states which say, "You pass this, or you don't move from fourth grade to fifth grade." The idea of basing promotion on results of a test -- is this a good idea?

I think the highest stake of all is that we continue to put students into the public who have not achieved the requisite requirements, as far as student learning and literacy is concerned. People are talking about the high stake of moving from one grade to the other. I would caution them that the highest stake of all is for our country not to be able to supply the level of education that our nation needs, in order for us to keep our national well being, and our national standing in the world. That's the highest stake.

So is that "Yes," making kids pass the test to graduate is a good idea?

I think the alternative to that would be, the alternative question would be, should we have students graduate from school that have not met the standards? Should we? If the answer to that is yes, then we don't need the test. But I think the answer should be no -- "No, students should not be permitted to leave high school absent achieving the mastery required to leave high school."

Would you find some other way of proving that they've achieved that mastery?

I think the other way would also be a test. So you're talking about a different kind of test. But there has to be some type of assessment to determine whether or not this achievement has occurred.

"The soft bigotry of low expectations." ... It's a phrase that the president used when he was governor of Texas.

And I think that it's a powerful statement. ... That ties into a statement that goes like this: "Leave no child behind." Now, our system has been doing pretty good, and we're moving many children forward. And we applaud that.

But right now, we're focusing on those who are not moving forward, [who] are being left behind. Many of those are being left behind because, simply, some person makes a judgment that these children are either, quote, "hard to teach," quote, not capable of certain academic achievement. And once that is in the mind of the system and the people who are doing the teaching, then we think that that is a bigoted position.

The alternative would be, give every child the maximum of opportunity to go as far as they can go with the ability that they have. But for an individual to conclude that a child can go no farther than some predetermined notion that they have in their head is a bigoted position, in my estimation.

I'm scratching my head. Why has it taken us so long to say, "Hey, we need standards?"

That is a mystery to me. Nowhere else in the nation, in other institutions and enterprises, would that be an issue. Let's think about that for just a minute. We're saying that, in schools, we should have standards. And that turns out to be a controversial discussion. Would that be the case in the hospital district, or in the public highways or in the building industry? Everybody else takes standards as the ordinary. And we're taking standards as something that is extraordinary? I think that gives you some clue on how much we need to change.

How are we doing on our standards, would you say?

I think we're doing rather well. ...

But some have said [that] the standards are this hodgepodge. It's a laundry list; they're a mile wide and an inch deep. And it's a good idea, but we aren't thinking them through.

I don't agree we're not thinking them through. It is broad. But we have a broad nation. We've got 50 different states. We have 16,000 different school districts in these 50 different states. And so I think it's going to take some degree of broadness, if that's a good word, to encompass all of them. America, the United States of America, is a very diverse nation.

And what we're trying to do is make one set of rules that covers all of this. So I think, by necessity, it has to be broad.

And we'll leave it up to the individual states to go deeper?

And we would encourage them to go deeper.

Now, this administration wants this testing. They're not calling it national testing -- it's testing of every kid. ...

Because it is not national. It is state-by-state testing.

Is this a step towards national testing?

Absolutely not. Matter of fact, we are resisting that. We're operating inside the framework of the Constitution that we have, which makes this an individual state responsibility. And any state has the authority, indeed the responsibility, to define its own education system. It would be much simpler if it were a centralized system, but it is not. And so we're operating inside the framework that we have. We think it's a good framework. Our Constitution has done well for this country.

How are we doing, would you say? Are we halfway there?

I think we're a little less than halfway there. But I think we are moving rapidly in the right direction.

We do have the National Assessment of Educational Progress [NAEP]. What is its role going to be in telling us how we're doing?

First of all, the National Assessment of Education Progress literally is not a test. It is a survey, and it is based on sampling techniques from each state -- a small [sample] of students, fourth and eighth grade, in math and reading. And it's designed to measure the progress of the state, of the big school district, not the individual child. ... [NAEP] is designed in such a way to measure a state's progress, not the individual progress of the individual students in the state. ... So in that sense, it is not a national test.

But, for example, when the Texas results came out, and then we were able to see how Texas kids did on the NAEP, you'd say, "Oh, they seem to be on" --

And that's quite different from seeing how individual kids did. What the NAEP does is tell us that this state is making this kind of progress. And so when we see a state that is making great progress on that state's adopted standards and state-adopted tests ... and that state is also not doing very well on the NAEP, that gives us one clue that we need to look at this difference, and to determine whether or not the standards and the system there is in need of improvement.

Now, what I've been hearing is that, yes, these standards are moving along, and the states are making them better, deeper, tougher, more thoughtful. But the testing is driving the standards.

No. It is the opposite. ... The testing is driven by the standards. First, adopt the standards. Second, develop an assessment system to determine the extent to which these standards are or are not being met.

But the reality is that, in schools, those tests are given in May, in early May, before the teacher is finished with the curriculum. Isn't that testing driving the system?

No. Actually, let's take a live example. In Texas, the test is given in late April, and that's about the end of the school year for all practical purposes. They've had the instructions from the beginning of the year, so that the results are back at the end of May.

So what happens between April and May?

There's a lot of other instruction that's continuing during that time.

So you don't worry that the test is in the driver's seat?

The test that we are talking about is not in the driver's seat. Are there examples of tests being in the driver's seat? Of course. But those are barriers that we need to work against. The system that we are proposing is: first adopt good, solid, clear standards; second, adopt the assessment instrument to determine the extent to which these standards are being made. By the way, once we adopt these standards, it drives not only the test, but a lot of other things. These standards can drive the adoption of textbooks, adoption of teaching materials, training of teachers, or staff development, assessment.

The key is, what do we want to have happen? What do we want students to learn, or know, or be able to do? Now let's get the materials to make that happen. And it has to be in line with these standards that we've adopted and the assessment system in place, to determine how much success we're having in achieving what we have set out as standards.

What's the biggest obstacle to a successful standards and accountability movement?

I think the biggest obstacle is the various points of view that people have, and there are so many different points of view, so many different understandings. We've got to build some consensus around what we are talking about. And I think once it is clear what we are proposing, I think that we'll have, we'll gain more support for it.

What's your biggest fear?

What's my biggest fear? That we are not successful in derailing the idea that success should be measured by the amount of spending that you put in place.

It's not just money?

Absolutely. Money is required for what we're talking about. But what we've learned from the past is [with] just money, absent reform and organizational structure with standards, you don't get the job done. But we've not yet been able to convince a lot of people of that, and that's my biggest fear.

So you think we are moving in the right direction?

Absolutely. We have great leadership in the White House.