When it comes to evaluating and validating assessments, including those used within statewide education accountability systems, the measurement community does many things well: We scrutinize the development of test blueprints and test items. We examine how effectively items function individually and how they behave as a group to generate estimates of students\u2019 ability in reading, math, science, and other content areas. We evaluate test alignment to standards, and how accessible tests are for all students. We compare results of the assessments across student groups to seek out and address potential biases.<\/p>\n
Despite these laudable efforts, gaps remain. Most critically, few resources are devoted to identifying data-driven ways to ensure the correct schools and districts are identified by the accountability system, and even fewer resources are spent determining whether the accountability system is improving students\u2019 educational outcomes. However, this type of outcomes-focused research is exactly what is needed to support well-designed interventions to assist students in the identified schools and districts.<\/p>\n
Ultimately, the purpose of state accountability systems is to identify schools and districts where assistance is most needed so that the state can allocate very targeted\u2014and often very limited\u2014resources. Recent state evaluations conducted by HumRRO can help identify best practices that should be used when gauging the effectiveness of an accountability system.<\/p>\n<\/div>