Pencils Out

Courtesy of Stanford University

Linda Darling-Hammond

One of the most contentious debates racking state houses this year isn't about Obamacare, inequality, or even jobs. It's the implementation of the Common Core State Standards, which set benchmarks for what students should know in math and language arts at each grade level. At the current count, 44 states and the District of Columbia have signed on.

Some conservative legislators have objected to what they see as a step toward federalizing education; Indiana has withdrawn from the initiative and 11 states are considering bills to slow or derail Common Core implementation. Another source of friction is the adoption of standardized testing to measure students' knowledge of Common Core standards and, in some cases, evaluate teachers. The assessments have been designed by two state groups—the Partnership for Assessment of Readiness for College and Careers (PARCC) and the Smarter Balanced Assessment Consortium—but a growing number of states are opting out and instead coming up with their own. Many education scholars also contend the tests don't assess students’ multiple intelligences and connect to real-world skills and tasks.

Stanford education professor Linda Darling-Hammond, who led President Obama’s education-policy transition team after his election in 2008, is one such figure. She is also a member of the Gordon Commission for the Future of Assessment and Education, a group of scholars overseeing the implementation of the Common Core. The Prospect sat down with Darling-Hammond to discuss the role of standardized testing in the adoption of Common Core and what the tests miss in evaluating student performance.

What is the role of standardized testing in the implementation of the “Common Core” curriculum?

When people talk about Common Core, they often mean the high-stakes tests attached to the standards and not the Common Core itself. Testing is not required by the Common Core. You can see that in the way the curriculum is being implemented across the country. In places like New York that include standardized testing in the Common Core, people are really talking about the tests and the high stakes attached to them. In a place like California, where the Common Core is being implemented without any high-stakes tests, there's much less anxiety and debate about the value of the actual standards.

Some states have opted out of using the tests developed by PARCC and Smarter Balanced. How does the goal of a national curriculum succeed if you have different testing regimes?

In my own view, the tests are not the most important thing. In fact, you have countries like Finland that have a national curriculum and have no external tests whatsoever, which surprises many Americans because they perform on the Organisation for Economic Co-operation and Development (OECD) tests near the top of the world. But they don't get there by driving students and teachers and schools with external tests. They just don't have them. They do all of their assessment locally within schools, around the national curriculum. So, the national curriculum does have guidance for assessments, but the tests should not be driving the curriculum.

Critics of standardized testing say the new tests are an improvement, but do not go far enough in terms of measuring the diverse skill sets and intelligences of students. In what way do the new tests assess a wider array of students’ knowledge and skills?

The tests are a step in the right direction for most states in that they include more open-ended items. In most cases, they include at least one or two performance tasks, which require the kids to take up a problem, do an analysis, write a response, and sometimes revise that response. There’s real engagement in the work.

But the Common Core state standards ask for much more. Students will be asked to collaborate, engage in the use of technologies for multiple purposes, communicate orally and in writing, do extensive research, apply mathematics and English language arts in complex problem-solving situations. The tests are not designed to reach all of those Common Core standards. They tackle the ones that are closest to what traditional sit-down tests can accomplish. Many of the answers will still be close-ended—that is, pick one answer out of five, or drag and drop your answer, or identify it from something that is already provided.

How do the tests compare to assessments in high-achieving nations?

In a lot of high-achieving nations, the assessments, first of all, are fewer. They are, in some cases, almost entirely open-ended—long essays and problem-solving questions that include performance components like designing and doing a scientific investigation, inquiring into a social-science problem … something very hands-on and illustrative of the capacity to do something with your knowledge. We've moved a little bit in that direction with these assessments, and that's a good thing, but we won't move as far as what is normal and common among high-achieving nations.

In the United States, we use tests in ways they were never intended to be used and in fact are prohibited by the standards that are supposed to guide test use in this country. To mechanically make a decision about whether a student can move forward to the next grade level, whether they'll graduate from high school, whether a teacher will continue to be employed or get merit pay, whether a school will be put into some kind of “failing schools” category and potentially closed down—these uses of tests are really irresponsible. They are unwarranted. By virtue of the fact that we know that tests are always partial and have a lot of errors associated with them, the testing standards always say that you should be using them with other measures. There's nowhere in the world where [this type of high-stakes standardized testing] is done except in the U.S., and it had never been done in the U.S. until the last few years.

Critics would say, “How else are we supposed to hold teachers accountable if not by testing what their students know?”

We have a lot of ways to evaluate and develop teaching. I think you can include information from assessments in teacher evaluation, but you also have to look at what teachers are doing. You have to understand how they're organizing the curriculum, what kinds of opportunities to learn they're providing kids, what kinds of feedback they give kids to improve. You can also include information about the assessments kids take. But you don't have to make a single, external, high-stakes test which poorly represents the skills you want to measure and is a mechanical trigger for sanctions and punishments. That's what we've done. That has distorted the system.

So, what do you see as the way forward?

I think in order to move forward we have to change the accountability paradigm in the country from “test and punish” to “assess and improve,” and we have to pursue the Common Core standards within that kind of framework. If we try to pour the Common Core standards into the old No Child Left Behind accountability framework, it will be like pouring new wine into old bottles.

You need a system that is designed to help teachers know a lot about what the assessments are going to be and to have them engaged in developing and scoring them. Kids should be engaged in doing activities that prepare them for complicated problems that they're going to confront in the world. The testing paradigm will have to evolve to engage students and teachers with the right work, and we should use the results to inform the continued improvement of the curriculum and the design of professional development. The Common Core State Standards themselves do express those more complex skills and abilities to a certain extent, so if we want to be faithful to that vision, we've got to change the way we think about assessment and accountability.

You may also like