dents—both standardized tests and textbooks tests—emphasize and mutually reinforce low level thinking and knowledge, and were found to have an extensive and pervasive influence on math and science instruction nationwide.5
If current assessment practices prevail, reform in school mathematics is not likely to succeed.
Proponents of mathematics education reform have expressed the view that the goal of more and better mathematics learning for all students cannot be realized if assessment remains wedded to what is easy to measure and what has traditionally been taught. The messages sent by new views of content, teaching, and learning will be contradicted by the values that such assessment practices communicate. Some teachers may attempt to respond to both messages by preparing students for the tests while, simultaneously, trying to offer students some taste of a richer, more challenging curriculum. Other teachers may continue to teach as they have always taught, reasoning that the tests used to make important decisions about students' lives, teachers' salaries, and educational quality indicate what is truly important for students to learn. If current assessment practices prevail, reform in school mathematics is not likely to succeed.
Suppose assessment practice were to change in American mathematics classes. What if, at the end of a unit, students wrote an essay explaining how two situations could both be modeled with the same exponential function, instead of being tested on skills such as solving equations and choosing among definitions? Imagine students being assessed not only with a final examination taken in class but also on how well they could conduct and report a group investigation of the symmetries in a wallpaper pattern. Suppose students were allowed to use calculators, computers, and other resources on most tests and examinations, including those administered by external agencies. If such changes were to occur, many mathematics teachers would shift their instruction to prepare their students adequately for such assessments.
Reformers have proposed a host of innovative approaches to assessment, many of which are described in subsequent sections of this report. Leaders in the educational policy community are joining the chorus, arguing that minimum competence tests and basic skill assessments, like those commonly seen today, work against efforts to improve schools.6 Low-level tests give coarse and deceptive readings of educational progress. Worse, they send the wrong message about what is important.7 Assessments need to record