Publications & Resources
Creating Accurate Science Benchmark Assessments to Inform Instruction
Terry P. Vendlinski, Sam Nagashima, and Joan L. Herman
Current educational policy highlights the important role that assessment can play in improving education. State standards and the assessments that are aligned with them establish targets for learning and promote school accountability for helping all students succeed; at the same time, feedback from assessment results is expected to provide districts, schools, and teachers with important information for guiding instructional planning and decision making. Yet even as No Child Left Behind (NCLB) and its requirements for adequate yearly progress put unprecedented emphasis on state tests, educators have discovered that annual state tests are too little and too late to guide teaching and learning. Recognizing the need for more frequent assessments to support student learning, many districts and schools have turned to benchmark testing—periodic assessments through which districts can monitor students’ progress, and schools and teachers can refine curriculum and teaching—to help students succeed. We report in this document a collaborative effort of teachers, district administrators, professional developers, and assessment researchers to develop benchmark assessments for elementary school science. In the sections which follow we provide the rationale for our work and its research question, describe our collaborative assessment development process and its results, and present conclusions.
Vendlinski, T. P., Nagashima, S., & Herman, J. L. (2007). Creating accurate science benchmark assessments to inform instruction (CRESST Report 730). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).