A constructivist approach to learning calls into question the adequacy o f norm-referenced, multiple-choice tests for assessing student learning. The authors describe one strategy for dealing with this problem.
On any given day, an outsider could walk into this public elementary school and observe students actively engaged in learning. Students might be working in cooperative groups, studying independently, participating in discussions, presenting a portfolio of their work, or listening to a short teacher- or peer-led lesson. A reading or writing workshop might be in progress. Students of different ages would be working together in the same classroom. Parents would be assisting students, and teachers would be guiding learning in myriad ways.
A scenario typical of many classrooms? Perhaps. As more teachers model instructional practice that reflects recent research on learning and as more states move to performance-based assessments, the nature of teaching, learning, and assessment is gradually changing. Still, this scenario would rarely be typical of an entire school. The coherent curricular, teaching, and learning philosophy common to this school originated 10 years ago as a separate program within the school. The developmentally based program, designed to give students more responsibility for their learning, has expanded over the decade to include the entire school.
Despite the rich learning environment and despite teachers' assertions that students routinely demonstrate their knowledge and skills, this school's median achievement test scores declined steadily for three years. Concerns about the declining scores led the teachers and principal to enlist our services for a classroom-based research study. Together we investigated a way of preparing students for standardized tests that would maintain the integrity of the school's curriculum and its methods of learning. This approach, a series of interactive workshops for children, was successful in three ways. Teachers learned, children gained confidence and skills, and test scores improved significantly. Yet responses to the improved test scores were mixed. Elation was tempered by suspicion that the results were too good; whispers of cheating escalated to allegations of test-rigging, a charge lodged by a group opposed to the teaching practices of the school. In the midst of feeling good about the children's improved self-confidence and performance, we found ourselves wondering, "Why bother?"
As we searched for answers to this question, we remembered a response to a question a student asked Annette Kolodny. Kolodny, now a dean and faculty member at the University of Arizona, talked about the responsibility an education imposes and asked rhetorically, "If you are not responsible for your knowledge, what on earth have we been educating you for?" Our renewed commitment to "bother" emerged from our belief that teachers need to use their knowledge of assessment in the service of children's learning. Although assessment entails much more than administering norm-referenced tests, knowledge of this form of assessment is critical and begins with an understanding of the uses and misuses of norm-referenced test scores. It broadens to an understanding of the ethical considerations inherent in designing test-preparation interventions that protect the integrity of the tests and honor the integrity of curriculum, pedagogy, and children. Armed with this knowledge and the awareness that these types of tests are prevalent in children's lives, it is difficult to imagine anything less than a commitment to act. This article explores a successful intervention that bridged a "thinking curriculum" and a traditional form of assessment and led one group of teachers to act.
Becoming Responsible For Our Knowledge
Norm-referenced, standardized tests, trusted by many as reliable and valid measures of student achievement, have been a routine part of American students' school experiences since the mid-1900s. …