15
	Assessment Conversations as a Tool for Reform
	Paul G. LeMahieu
	University of Delaware and Delaware Department of Education
	Marshá T. Horton
	Delaware Department of Education

What role might extended, open-ended tasks play in assessment? Through the good efforts of many individuals and the reform of large-scale assessment initiatives, we have begun to ask such questions. While there is yet much to be done to bring technical quality and intellectual rigor to this particular aspect of a more general reform movement, it is clear that we will never again think of assessment in precisely the same terms as those that have dominated our thinking for the past century or more. What is less certain is what role assessment should play in order to maximize its contribution to broader reform efforts.

     Some see assessment as an agent of reform, as a lever that, when properly applied, lifts the system (and the individuals within it) to improved performance by increasing accountability. This view is based on a well-established logic about the force and influence of assessment and accountability: clear articulation of goals will make public the expectations for individuals and the education system; appropriate and adequate assessments will reveal the performance of individuals, schools, and systems; and appropriate sanctions and rewards will provide the motivation to improve effectiveness and productivity.

     This "lever view" is plausible and quite possible. We are not too optimistic, however, about assessments (reformed or otherwise) used solely in this way and detect an underlying cynical assumption: that those who work in our schools have the skills and the capacities to help their students learn better, but for some reason choose not to do so. Our experience within schools does not support this cynical view. We have encountered some who (sad to say) lack the requisite familiarity with national, state, or local standards for student learning; more who have not had the opportunity to develop or refine the skills necessary to ensure their students' success; and still more who are constrained by a lack of intellectual or material support from their school districts or systems for the kinds of practice necessary to prefigure accomplishment of the standards. We have not encountered many who willfully choose not to perform well.

     However, tasks like the ones in this volume, if we assume that these are assessment tasks, do offer a potentially powerful but very different role for assessment in reform. They might play an "instrumental" role, in which teachers work with such tasks in a variety of ways that challenge their practice. First and most obvious is the signaling function, helping to make the standards concrete. This is particularly important as the concepts (curricular and instructional) underlying the standards become more complex and more subtle.

     There is a second instrumental use for tasks like those documented here, also related to teachers' growth and development. Such tasks can be strategically used to stimulate and discipline what we have come to term the "assessment conversation." Making these tasks (as well as the student work that they elicit) central to professional discourse can challenge teachers' notions of students' capacities. Appropriate professional development focused on the use of these tasks can powerfully shape teachers' notions about what constitutes high-quality student work and what serves as adequate evidence of such quality. Our experience with the instrumental use of tasks like these is instructive: assessment and the development and use of tasks such as these can be approached so as to maximize the beneficial impact on instructional practice.

     Just as standards-based reform requires so much more by way of professional judgment, there is a need to warrant that judgment. Given the very complex systems within which educational professionals work, faith and trust must be located somewhere. That confidence can be placed in the people who work within the system or can be placed in mechanisms designed to control people and their behaviors. The current reform movement places faith in people rather than mechanisms. Whether one considers local democratization and site-based management; teacher empowerment and shared-decision making; the increased role of teachers in curriculum definition and selection; efforts at organizational development founded upon continuous improvement and learning community ideals; or the advent of assessment approaches that more explicitly privilege human judgment--in all of these cases the trust and the hopes for a high-performing system are vested in the teachers and other professionals who serve within it.

     However, the call for placing our trust in the empowerment of professionals requires that serious action be taken to warrant that trust. This is the basis for the persistent urging for investment in capacity building at the system level and in the professional development of individuals. As part of that professional development, we are optimistic about the potential of well-designed "assessment conversations" as instruments to shape teachers' expectations, beliefs, and practices.

     Assessment systems offer important opportunities to show that our trust in teachers is warranted and thereby facilitate beneficial change. Well constructed assessment activities and the efforts to employ them strategically in professional development activity permit teachers to engage in review of student work; develop shared notions of high-quality student performance; determine what constitutes adequate evidence of high quality; and ultimately to reflect upon the kinds of learning experiences that challenge traditional practice and produce the desired performances.

     We see immense potential for the use of tasks, such as those in this volume, as assessments within a process of ongoing and sustained development designed to support professional growth. We have done so involving large numbers of teachers who meet regularly (at least monthly for up to two days at a time) over an extended period (two years) to develop assessment tasks and to pilot and refine them in their classrooms. In these assessment-development conversations, teachers share and closely examine students' work. A professional discourse is provoked by the simple guiding question, "What can we tell about this student as a learner in mathematics?"

     The ensuing discussions are closely documented and analyzed for "points of evaluative judgment." These points of judgment represent the beginnings of a framework or rubric for evaluating student work. Over time, insights grow in number, depth, and sophistication, and are synthesized into a framework that teachers use and refine.

     Can an evaluative framework derived from one collection of student work be applied meaningfully to others? Will it hold up across a diverse array of students (with respect to abilities, levels of performance, cultural and linguistic backgrounds, gender, etc.)? Can it be applied with discipline and consistency (yet insight and sophistication) by many teachers? These and similar questions are addressed by taking the evaluative framework and applying it to new and broader collections of student work. At the same time that an evaluative perspective is being validated in this way, the sophistication, consistency, and reliability of shared interpretations are also being refined. Over time, these conversations, begun with a group of teachers involved in assessment development, are replicated in professional development activities on a broader and more extensive scale.

     In our experience, two things have invariably been realized through these "assessment conversations." First, all who participate in them leave with higher and more clearly articulated aspirations for student performance. This should not be the least surprising, as the derivation of criteria and expectations for quality in mathematical performances is essentially additive. One teacher sees certain things in a piece of student work, while the next recognizes some (but perhaps not all) of them and adds others. These assessment conversations proceed until the final set of aspirations (criteria of quality) are far greater than the initial one or that of any one teacher at the outset. Simply put, these assessment conversations increase the expectations of all those who participate in them.

     The second effect of these assessment conversations is that a shared framework for interpreting and evaluating student work emerges. The aspirations and expectations associated with this framework become commonly understood by the teachers and more consistently applied to all their students. Again, the nature of these conversations (long-term shared encounters and reflections) supports this outcome.

     These two outcomes of assessment conversations--elevated aspirations and more consistently held and applied aspirations--are key ingredients in a recipe for beneficial change. Educational research is nowhere more compelling than in its documentation of the relationship between expectations and student performance. Where expectations are high and represent demanding yet attainable goals, students strive to respond and, ultimately, they do achieve. Assessment conversations, focused upon tasks such as those in this volume and student work produced in response to them, provide a powerful device through which to warrant investment in the human side of the educational system. It is when assessment is used to provoke conversations of this kind that we find cause for optimism about the role of assessment in reform.

Paul G. LeMahieu currently serves as Director of the Delaware Education Research and Development Center and as Associate Professor of Educational Studies at the University of Delaware. He is currently the principal investigator of Delaware's Statewide Systemic Initiative for Mathematics and Science Reform. He also holds a senior staff appointment in the Delaware Department of Education as Special Undersecretary of Education for Policy Research and Development. LeMahieu has received a number of major awards for his contributions to educational theory and practice from the American Educational Research Association, the Evaluation Research Society, the Buros Institute of Measurement, the National Association of Test Directors, and the Association for Supervision and Curriculum Development. He is a former member of the Mathematical Sciences Education Board, having served on its Executive Committee and as Chair of its National Forum on School Mathematics.

Marshá T. Horton is the Associate State Superintendent of Assessments and Accountability for the Delaware Department of Education. In this position, Horton coordinates the design and implementation of the Delaware State Testing Program and coordinates the department's teacher certification and licensure responsibilities. She has served on the National Reading Research Center National Advisory Board, the Sweet Briar College Board of Directors, and the New Standards Project Literacy Advisory Panel. She recently co-authored a paper with Dr. Paul LeMahieu on standards-based accountability and has served as a consultant to many school districts and national organizations.