High Stakes vs. Democracy

K-12 Testing

by Ken Jones

In the current fever over using high-stakes testing for accountability, little discussion has been given to how this strategy affects the democratic purposes of schools. Most of the time, test scores are treated as ends in themselves, as if they can stand alone as clear indicators of school quality and health.

What if we took seriously the idea that schools exist to support and uphold our democracy? What would our vision for the classroom entail? And how would high stakes testing affect that vision?

One place that sheds light on this issue is Kentucky. In 1990, the state passed a sweeping school reform act called the Kentucky Education Reform Act (KERA). (See Examiner, ). At the center of this reform was a set of goals for students that addressed the needs of a democracy for preparing well-educated citizens, including focuses on thinking and problem solving, real-life applications, self-sufficiency, team membership, and interdisciplinary integration.

A statewide assessment system that would be “primarily performance-based” was called for in order to determine how well students were achieving these new educational targets. High standards meant going beyond the kind of static and fragmented knowledge typically measured by standardized tests. The stated intention was that this new form of assessment would “drive” curriculum and instruction in classrooms towards new and improved learning.

The vision expressed in state documents was of a classroom where committed teachers would engage students in meaningful and relevant learning experiences; critical thinking and depth of understanding would be valued and cultivated; and the roles of teachers and students would be characterized by the informed use of judgment, decision-making, and responsibility. Local school councils, consisting of teachers, parents, and school administrators, were empowered to make curricular decisions that would lead to greater responsiveness to the diverse needs of students. All necessary ingredients for the feeding of a democracy.

To give this reform “teeth,” high-stakes accountability for schools was connected to the state assessment, complete with rewards and sanctions for teachers. State authorities acknowledged that this might lead to “teaching to the test,” but asserted that this wouldn’t be so bad because the test would be more high-level and authentic and would prompt desirable changes in the classroom.

Kentucky thus became the first state to link high stakes to performance assessments as a means of school accountability. Now, ten years later, what do we know about this strategy? While the Kentucky Department of Education celebrates KERA successes in terms of scores on the state test, there is a more troubling story underneath the scores. This has to do with how the prerequisites of high stakes have lead inexorably to standardization and disempowerment. What we have seen is not the flowering of local curriculum leading to responsiveness to students, but a wholesale regimented responsiveness to the state. Moreover, the state test itself has changed shape so that it is no longer “primarily performance-based.”

At the heart of this loss is the requirement that high-stakes tests adhere to established psychometric principles of validity and reliability. When tests dictate whether teachers will lose their jobs or students will graduate, it is essential that the tests be able to stand up in a court of law, for surely these decisions will be contested by the inevitable “casualties.” It is this pressure that forces state departments to take a safe approach to testing, despite all best intentions about classroom visions.

Content validity requires that the assessment align with the curriculum. And yet, how can this happen when the state is in charge of the assessment and the schools are in charge of the curriculum? In Kentucky, it has meant that the schools, held highly accountable by the mandated test, have demanded that the state define the curriculum to be tested. Over the years, we have seen greater and greater articulation from the state. Now there is a very precise, and extensive, core content for assessment published by the state. Almost universally, schools spend a great deal of time “aligning” their curriculum to this state document to ensure that they are providing their students with the proper instruction. Thus, opportunity to learn has come to mean teaching to the test. So much for local empowerment.

Reliability, in the context of performance assessment, refers to the inter-rater consistency in scoring open-ended responses such as portfolios, performance events, and open response items. Scoring such items requires human judgment. This is a problem encountered when such assessments are used on a large scale for high stakes. It takes a great deal of time and professional development to form the widespread understanding of performance standards necessary to achieve consensus about “how good is good enough.” The immediate need for high reliability coefficients has been the demise of performance assessments in Kentucky. Gone are the performance events and the math portfolios; back are the multiple-choice questions, more oriented to objective answers than are the open-response questions. The writing portfolio, the only element of the state assessment which has demonstrated positive effects in classroom practice, has been retained, but has been scaled back. The vision has been lost.

The moral to this story is that the high-stakes use of tests dictates standardization. A democratic vision for the classroom will be undermined by the control taken by the state when it uses testing as a weapon. Let us not be deceived. It is the state that is empowered through high stakes, not local schools, parents, and teachers. This movement does little to foster the kinds of classroom activities, decisions, and judgments that we envision for democratic education. It is time for a new generation of school accountability that better suits a democracy.

Ken Jones currently works as the mathematics specialist for Jefferson County Public Schools in Louisville, KY. He is co-editor, with Betty Lou Whitford, of Accountability, Assessment, and Teacher Commitment: Lessons from Kentucky's Reform Efforts (SUNY Press, 2000). Available from CUP Services, (607) 277-2211; paperback, $21.95; hardcover, $65.50.