To the Editor:
It may be a mistake in public policy to continue the heavy dependence on standardized testing for high-stakes purposes to drive a punitive system of accountability for teaching and learning. I can think of no substantial body of research that supports the idea that we can force teachers or students to succeed by negative reinforcement (punishment).
Similarly, I have not seen evidence that suggests that the standardized testing of the status of students’ abilities contributes much to our understanding of what should be done to teach better or to learn better. Penalizing teachers or students for failing to succeed appears to be dysfunctional, and I believe immoral, given the weak association between the assessment of student status and the central processes of education—teaching and learning.
This punitive use of the data from standardized tests may be eroding confidence in the education enterprise. As instances of cheating are uncovered with increasing frequency—cheating not only by students but also by administrators and teachers for their own advancement or self-protection—it will be difficult for any of us to maintain confidence in our system of education.
Assessment of education based upon the isolated use of standardized tests is dysfunctional as a model for teaching and learning, especially so when more positive models are available to inform these processes.
The recent report from the Gordon Commission on the Future of Assessment in Education (which I chair), “To Assess, To Teach, To Learn: A Vision for the Future of Assessment,” strongly suggests that assessment for education can be used to inform and improve teaching and learning.
We have made considerable progress in the generation of formative-assessment data from which to inform teaching and learning transactions. Assessment can be used to analyze and document teaching and learning processes in addition to the measurement of achieved status.
This report, along with a second from the commission, “A Statement Concerning Public Policy,” advocates the use of systems of examinations: different kinds of examinations, distributed over time, embedded into instruction and in different contexts. The reports question the relative value of the focus on the measurement of how much content and skill has been learned, as opposed to the appraisal and documentation of the development of competence.
Edmund W. Gordon
John M. Musser Professor of Psychology, Emeritus
Yale University
Richard March Hoe Professor of Psychology and Education, Emeritus
Teachers College Columbia University
Pomona, N.Y.
The writer chairs the Gordon Commission on the Future of Assessment in Education.
