CCE staff and partner reflections on our collaborative work to create schools where learning is engaging and rewarding, and every student is set up for success.
In Search of a Better Accountability System
Through more than 20 years of No Child Left Behind (NCLB), we have lived with a uniform definition of accountability, that of a standardized test used to make determinations of student learning and school and district progress. It is time to question some of the assumptions underlying this practice.
Assumption #1: The primary purpose of accountability is measurement of learning
Currently the primary focus of accountability systems, using standardized tests, is to provide data on student and school performance so as to sort, rate, and rank the performance of students, schools, and districts. The utility of this approach is compromised as data is usually sent to districts several months after students have taken the test and often when the subsequent school year has already started.
In a different paradigm, accountability should be a key tool to promote student learning and educator growth at the front end of the learning cycle, while providing a valid, reliable, and timely measurement of student learning at the back end. Imagine an accountability system that is built around extended, curriculum-embedded, and teacher-generated performance assessments, in which students demonstrate proficiency in ways that will be expected of them later in college and career. In this latter case, students are engaged in a learning experience that also provides assessment data on their proficiency over target competencies, while educators are gaining experience in designing rich learning experiences that are also measurement tools.
Assumption #2: The best lens for measurement should be a uniform standardized test administered at a single point of time
Using standardized tests, which represent one snapshot in time, as the sole point in making decisions about student and school progress has been found by researchers time and again to be an improper use of assessment with little scientific basis. The form that standardized testing takes – largely multiple choice questions and some short answer open response questions, does little to promote deep learning. Test development conducted by external agents inevitably leads to discontinuity within the local curriculum while depriving educators of valuable professional learning experiences in the design and scoring of meaningful assessments.
Imagine instead a system in which teachers design and administer engaging performance assessments embedded within the local curriculum. Extended assessments that result in research papers, multi-media presentations, and analyses of lab experiments provide a richer picture of what students know and are able to do than could a standardized test. Making judgments about student learning and school quality based on a body of work – a select number of pieces of student work from a number of assessments within a given discipline, provides a much richer and more accurate picture of student learning than a single, disconnected standardized test.
Assumption #3: We need to adhere to the most rigorous definition of technical quality in measurement
Bowing to the pressure of ensuring the highest technical quality of validity and reliability leads to a bit of chasing “fool’s gold.” By taking this path, we have boxed ourselves in to an assessment system that measures a quite narrow amount of learning, excluding learning that is of increasing importance in today’s world – complex higher order thinking including reasoning, analysis, synthesis, problem solving, communication, creativity, and collaboration. If we are not measuring critically important areas of learning at the expense of attaining rigorous technical quality, one has to question the rationale for having the test at all.
There are other ways of attaining technical quality that support meaningful learning. Providing teachers with tools for ensuring task validity along with expert task reviews and engaging educators in cross-district sessions on scoring student work are two ways of addressing technical quality within a more student-centered accountability system. You can still achieve a measure of technical quality with a mix of local and common performance assessments, even if it is of a different nature than historical practice.
Assumption #4: The standardized testing model of accountability is the best pathway to attaining equity in outcomes across student subgroups
In actuality, achievement gaps across race, language, and income have remained stubbornly wide throughout the entire tenure of NCLB. There is little evidence that the standardized testing movement has led to greater equity, providing legitimate cause to question the wisdom of continuing down a path of accountability that has not achieved any significant degree of success in achieving its intended ends. On the other hand, there is early evidence that rich curriculum-embedded performance assessments enable a greater diversity of students to demonstration proficiency.
A move toward reciprocal accountability, in which local districts use multiple measures to report progress to the state and the state provides support to districts to build quality local accountability systems, is possible. New Hampshire’s Performance Assessment for Competency Education (PACE), California’s CORE partnership, the NY Performance Standards Consortium, and the emerging MA Consortium for Innovative Education Assessment are all examples of schools, districts, and states moving in this direction.
Today, our nation’s students are increasingly diverse. Correspondingly, our means of assessing students should correspondingly become more diverse. This does not mean lowering standards; rather, it means we need to assess student and school progress in ways that provide deeper learning experiences for students and rich professional growth for educators.