A Critical Fork in the Road

Written by Tony Wagner (co-authored by Tom Vander Ark)

President Bush's strong support for the goal of raising learning standards for all students and focusing additional resources on improving schools that serve poor and minority children represents a significant step forward for this country. His proposals come at a point in time when standards and standardized tests have helped to create awareness of the achievement gap between students of different backgrounds and to frame the national challenge we face.

But many states face a critical fork in the road: to increase the focus on "passing the test," or to increase the focus on meaningful student learning. States that have become too dependent on high-stakes, high-standards tests have another option: build an accountability system that helps students focus on producing quality work, helps teachers focus on improving the quality of student work, and helps school systems create portfolios of quality schools.

Researchers at a recent national conference at Harvard University's graduate school of education reported that the dropout rate in many states has increased in the last few years, most likely as a direct result of high-stakes testing. According to University of Chicago researcher Anthony Bryk, when Chicago instituted a high-stakes 9th grade test several years ago, the 8th grade dropout rate increased significantly.

Meanwhile, growing numbers of students, teachers, and parents are concerned that the curriculum in their schools may demand more memorization than before, but is not intellectually more rigorous. In too many schools, the curriculum has been reduced to "test prep." An increased focus on teaching to the test further undermines many students' motivation to learn and will likely accelerate the flight of good teachers and principals from the schools that serve the neediest children.

The reason many of the best educators are tempted to leave the profession is that they know that bad schools can hide behind good test scores, and good schools don't necessarily have the best test results. As Deborah Meier, the founder of the highly acclaimed Central Park East Secondary School in New York City, has said, "I can change kids' lives more easily than I can improve their standardized-test scores."

The heart of the problem is that cheap, paper-and- pencil, computer-scored state tests do not begin to assess the competencies that large majorities of adults agree are essential today: the ability to comprehend difficult reading material and apply information to the solution of complex problems; the ability to write and speak clearly and thoughtfully; the ability to understand mathematical data and use math and technology as problem- solving tools; the ability to work effectively in teams; and, finally, respect for others and an understanding of our roles as citizens.

Effective teaching and high-quality assessments for these competencies do exist in some schools. A group of educators from around the country recently shared its work on performance-based assessment during another Harvard conference, this one sponsored by the Bill & Melinda Gates Foundation. They reported that even those students most at risk of school failure can perform at very high levels, given the right conditions: much smaller schools, teacher teamwork, a personalized learning environment, and many more opportunities for applied and hands-on learning.

Even more significant, hundreds of these new small high schools that have consciously created these conditions for all students may have mediocre test scores. But they also have virtually no dropouts and a college- attendance rate of more than 90 percent.

The best corporations have also developed means of assessing these more sophisticated skills; they do not rely on paper-and-pencil tests for important personnel decisions. They use simulations, interviews, and performance-based assessments to identify and hire the most qualified individuals. Such testing is expensive to develop and administer, though, and the results are often not readily quantifiable or easily compared across sample groups, which makes their application for broad-scale educational accountability difficult.

To create better incentives for learning and teaching at high levels, we believe American public education needs a fundamentally different system of accountability. The essential tasks of schools, districts, states, and the federal government must be re-examined and redefined. The following is an outline of what we think a performance-based accountability system might look like, and some of the short- and long-term steps that would be needed to create such a system:

What Schools Must Do. The first and most important task of every school is to ask two questions: What should students know and be able to do, and what specific forms of evidence of that will we look for? Rather than take a quiz (or a standardized test) on the parts of speech, for example, students in this "merit badge" type of system would be required to write a research paper and give a speech in order to graduate or move on to the next grade level. The New York Performance Standards Consortium has developed very sophisticated "merit badges" in a number of academic areas that are requirements for graduation in all 39 of the consortium's high schools.We believe American public education needs a fundamentally different system of accountability.
Redefining the Role of Districts. Charter schools are pioneering an accountability system that appears to have promise for all schools. Rather than managing schools, districts should establish clear school accountability through performance contracts with schools. These contracts would clarify academic goals, operating parameters, financing, and consequences for lack of performance (as described by Paul T. Hill and the recent Education Commission of the States study, "Governing America's Schools: Changing the Rules").

Districts would ensure that all students had equal access to a variety of high-quality schools. Pennsylvania is deploying this contracting strategy in the Chester-Upland district and recently chose three different companies to each manage one or more schools in that community. We believe this will be the new role for districts: managing a portfolio of schools with clear accountability, flexibility, and choice for teachers, parents, and students.

The Tasks for States. States can lend support to the development of performance-based assessment systems in a number of ways. First, they can support the development of digital portfolios of student work, so that multiple measures of student competencies are readily available. Second, they can encourage colleges and universities to adopt competency requirements for admission that are linked to students' digital-portfolio content. Third, they can encourage robust public-information systems that allow comparison of school performance data (as Just4Kids has done at www.just4kids.org and other important considerations (as the Seattle Times has done at www.seattletimes.com).

Public- information systems could be enhanced with school quality reviews, in which teams visit schools regularly to assess academic quality by multiple means. Rigorous peer reviews, including class visits; discussions with parents, teachers, and students; audits of student work; and the examination of all relevant quantitative and qualitative data, are far more reliable measures of school quality than test scores alone. Such a system has been used in Britain for years and has been successfully piloted here in the states of New York, Rhode Island, and Illinois.

Robust public-information systems also would allow states more freedom to license alternative systems of standards and assessments to networks of schools, rather than imposing a single arbitrary definition of "high standards." Some districts and many organizations, including New American Schools, a nonprofit developer of school improvement models, have already crafted excellent K-12 standards and curricula. They are all rigorous, but they are not all the same.

Finally, states must continue to license teachers—but not through standardized tests alone. Teachers should be required to periodically submit portfolios that document teaching mastery, similar to what is required for certification by the National Board for Professional Teaching Standards. States should also consider developing a tiered licensing system, as Allan Odden at the University of Wisconsin has proposed. Such a system would provide a career ladder for educators based on knowledge, skill, and responsibility.

Online Assessment: An Important Next Step. The federal government has the opportunity to develop or create incentives for online diagnostic systems for literacy and numeracy—tests that give teachers, students, and parents immediate results that can be used continuously to improve learning. Rather than the high-drama, weeklong spring events that have become the focus of the school year in many states, online testing would allow tests to move into the background and become another regular source of student-achievement data. Combined with classroom evidence and systemwide (or statewide) performance tasks, online assessment would help inform instructional decisions and improve performance-monitoring capacity.

The standards movement has brought us far. But misusing standardized tests will not get us better-educated citizens. To create significantly better schools for all children, we need to reinvent educational accountability from the local level to the federal level.

Co-Author Tom Vander Ark, a former business executive and superintendent of schools, is the executive director of education for the Bill & Melinda Gates Foundation in Seattle. For more information about the work of the foundation, click here

ResourcesTony WagnerJanuary 27, 2011Articles