Ann Cook: A View on Performance-Based Assessment

Ann Cook, the co-director of Urban Academy in New York City, describes the school's approach to curriculum and assessment.

1. What is Urban Academy's approach to teaching and learning?
2. How do you assess student work at Urban Academy?
3. How does Urban differ from more traditional public high schools, especially from a student perspective?
4. How do you know that the Consortium approach works?
5. What is the relationship between standards, curriculum, and high-stakes tests?
6. How do you respond to the argument that we need reliable test data to compare how students are doing in different districts, cities, or even states?

1. What is Urban Academy's approach to teaching and learning?

We're very interested in students developing certain skills. We're interested in them developing an ability to work with multiple perspectives, to be able to analyze evidence, to be able to critique. In our history classes we don't use textbooks, we use primary source materials, so kids would be exposed to ideas such as, "What does this historian think about the reasons for the Civil War? How about that historian?" And then, what do they think -- what do the kids think -- about that evidence and how did they present that? Can they present it verbally? Can they argue it out in class? Can they present it in writing? Can they argue it in a written paper? Do they understand the nuances of different kinds of issues?

We want a way of teaching that invites and encourages and engages kids to be developing their own point of view and their own perspective based on evidence.

Back to top

2. How do you assess student work at Urban Academy?

We're part of a consortium. There are thirty-seven schools across New York state that are part of a consortium of schools -- public high schools -- that use performance assessment instead of high-stakes tests in order to organize our curriculum and determine whether students are ready for graduation.

It's a system of assessment -- not a single instrument -- that's based on a number of components. It goes on all year long, and culminates in tasks we ask students to do that demonstrate that they can actually do something: They can write research papers. They can devise and conduct and defend an original science experiment. They can apply mathematical concepts to real situations like measuring the height and volume of buildings or the distance between South Ferry and Staten Island using the Statue of Liberty as the point. These are the skills they take with them when they leave these schools, and they can go on and do college work.

Back to top

3. How does Urban differ from more traditional public high schools, especially from a student perspective?

Traditional education is pretty bound to the textbook and to moving kids through a body of information. We're less interested in moving kids through a body of information and more interested in having kids engage in some way with the material. I would say we're more likely to focus on depth rather than coverage. And we're more interested in, for example kids learning what it means to be an historian than reading a textbook that covers a whole lot of stuff, which they'll forget. We want kids to be able to deal with the ideas.

Coupled with this approach is that we're very interested in hearing what the kids have to say about all of it. So student voice is a very critical piece of this. I do exit interviews of the students before they graduate. In almost every case over the last fifteen years that I've been doing this, the kids say the same thing. They say, "Well, you were interested in my point of view, you were interested in my ideas; I got an opportunity to express myself and in expressing myself I realized I had an obligation to know what I was talking about. And I had a chance to play with ideas and then figure out what I thought about those ideas." I think that's a critical piece.

Back to top

4. How do you know that the Consortium approach works?

The consortium schools in New York have a population that in some cases is even more economically disadvantaged than the average kid. Throughout the consortium, 66 or more percent of the kids are on free or reduced lunch. So we have a disadvantaged population in the schools represented in the consortium.

If you look at that population in say New York City and you compare it to the population in the consortium schools, here's what you find: You find that the dropout rate for New York City schools is somewhere upwards of 30 percent. In some cases, it's closer to 60 percent. In the consortium schools it's under five percent. If you look at college as an indicator, New York City sends 58.1 percent of its high school graduates on to colleges. That includes four-year and two-year colleges. If you look at the consortium schools we send 90.1 percent on to colleges, and a very big percentage of those kids go on to four year colleges. Urban Academy sends 97 percent of its kids on to college, and almost all of those kids go on to four year colleges. Many of them go on full scholarship to some of the leading colleges in the country.

These figures are really the indicators, and I think we should pay much more attention to where kids started from when they first came in, and how far the school was able to move them, and then what it is that they do when they actually leave the school.

Back to top

5. What is the relationship between standards, curriculum, and high-stakes tests?

I think that first of all we have to be clear about this mythology about the connection between high standards and high-stakes tests. I'm all for high standards -- I don't know of anybody who's for low standards. The question is, do we get we want using the tests to drive this? That's the real crux of it. And I would argue that we don't -- that in fact, we get the opposite of what we want.

I think what testing does is change the curriculum. Testing focuses people on the test. The higher stakes you make the test, the more you impact on the curriculum. So you get rid of things like arts. And you get rid of the interesting stuff that's going to engage kids. And you then start attaching penalties if kids don't do well on the tests. Then you need to hold kids back if they don't do well on the test. And if we look at the data and the research, holding them back once will increase the possibilities that they'll never finish school by 50 percent. If we hold them back twice, based on these tests, we increase the possibilities they'll never finish school by 90 percent.

There are consequences to everything we do. If we were going to say that we want these tests, we have to be much clearer about whether these tests are going to take us in the direction that we really want.

Back to top

6. How do you respond to the argument that we need reliable test data to compare how students are doing in different districts, cities, or even states?

If what we want are tests to tell us something about this city compared to this city, or this district compared to this district, we don't need to test every child. We can do that with sampling. Gallups samples 1,500 people across the country to tell us what Americans think about any number of things. And we could do that with educational testing, too. So I don't think we need this very high-profile, high-stakes testing that is trivializing curriculum and pushing teachers and kids, I think, into the wrong direction in terms of what we really want out of education.

Back to top