Measuring Success: Using Assessments and Accountability to Raise Student Achievement

A Comment to the House Subcommittee on Education Reform

ACT, Inc.

Richard L. Ferguson, President

Introduction

Almost everything about U. S. education policy has changed since the 1989 Education Summit, when then-President George Bush and the governors met to respond to the challenges identified in A Nation at Risk (National Committee on Excellence in Education, 1983). A central catalyst for this change has been the evaluation requirements for Title I as stipulated in the 1994 Improving America's Schools Act. With encouragement from the federal government, virtually every state has developed standards in core academic areas that are intended to define what students should know and be able to do by the end of their K-12 education.

More change remains to be made, however. The states' standards are, in many cases, fundamentally disconnected from statewide assessments and from instruction. Many states need to overhaul their systems to ensure that their assessments directly address their standards while, at the same time, providing technically sound results. What is perhaps most important is that many teachers may feel ill-equipped to teach what the state standards define. Thus, their students may not have the opportunity to learn the very skills and proficiencies on which they will be tested.

President George W. Bush's education plan, "No Child Left Behind," contains significant elements essential to improving the quality of our nation's schools. The plan empowers states and schools to take important steps toward achieving educational excellence. We at ACT commend the president for coming forward with priorities that will allow the states to design their own programs to help each child reach his or her full potential.

The president's plan is a call for fairness in the educational enterprise, a call that—if embraced by all students, parents, teachers, and administrators—can result in meaningful gains in student achievement across the socioeconomic spectrum. Well-designed and thoughtfully used assessments are indispensable to effective monitoring of student and school progress. In the hands of skilled teachers, they can have a significant positive impact on classroom practices.

We at ACT appreciate this opportunity to delineate what we believe are the four essentials for making President Bush's plan a success. These necessary elements are: well-defined standards, high-quality assessments, multiple measures, and direct ties to instruction.

Well-Defined Standards

Much has been written in the past few years about the strengths and weaknesses of state standards. There is little question that the 49 sets of standards developed to date vary greatly in their specificity and in their emphases. We believe it is critically important that the standards defined by each state be clear and specific in their expectations of students.

A well-written standard should articulate both what is to be measured and the ways in which it can be measured. Otherwise, states will be vulnerable to assessments that do not measure what their state standards intend, and teachers will not necessarily be instructing students in the skills and proficiency domains defined by the standards. We should not hold teachers and schools accountable for ambiguous or vague standards.

In regard to students in grades 3-8 specifically, we also strongly believe that the standards the states identify ought to position the students well for acquiring subsequent skills and proficiencies in high school, in postsecondary education, and in the workforce. Because the skills students attain in these early years will form their foundation for lifelong learning, these skills must lead toward longer-term goals like readiness for college and the world of work. If we fail to establish a firm foundation of skills and knowledge in our children, we will limit their opportunities in later life.

High-Quality Assessments

If schools and states are to be held accountable for raising student achievement, they must also have high-quality assessments to measure their results. This means, first, that the assessments must be designed specifically for these uses. They must also provide scores that have been validated for use in making the kinds of decisions the states will make about students and schools. Test forms must be created specifically to measure growth one year to the next. For this type of annual assessment, new tests must be made available each year, and these tests must be equated to previous forms to ensure that the scores from year to year will be comparable.

In addition, we must provide states more options than they currently have. Existing tests will not serve the needs described in President Bush's plan. And the nation cannot afford to try to get by on the cheap. If we use the same test forms year after year, the overexposure will result in inflated, and thus misleading, scores. And we will run the risk of constricting the education of our children, who will be taught and will learn only what is on the tests rather than the entire domain of important proficiencies that they need to learn.

Multiple Measures

The president's plan does not prescribe the number or types of measures to be used to assess student progress. We at ACT encourage the use of multiple sources of information to support decisions, particularly those arising from evaluations of student progress. Improvement decisions need to be informed by all relevant available information. They should, for instance, take into account students' levels of proficiency before instruction; that is, while we maintain the expectation that all students will improve, we must allow for the fact that students will be at different points when they start to learn. Multiple measures will better inform us regarding the gains students are making and will support important judgments about the quality of the education they are receiving.

Annual administration of the National Assessment of Educational Progress is included in the president's plan as an additional source of information about student achievement. If NAEP scores are to be added to the mix, we should recognize some qualifications on their use: 1) that NAEP, because it does not provide individual scores, can be used only as a measure of group progress; and 2) that NAEP scores may or may not measure a particular state's standards. If a NAEP test significantly covers a state's standards, then the test scores could provide information relevant to the decisions being made. But if a NAEP test does not align with a state's standards, then NAEP and the state assessment will be measuring different things. The NAEP measure won't be consistent in indicating changes relative to the state standards. Thus, we recommend that NAEP results be used only in those cases where the tests align with state standards.

Direct Ties to Instruction

If schools are to be held accountable, they must have access to resource materials that incorporate the standards of their state into the curriculum. Teachers and curriculum leaders will need support so they can understand and be ready to teach to students the knowledge, skills, and proficiencies defined by the state standards and measured by annual assessments. Teachers should also have ready access to the intended goals of their instruction. They must understand that their responsibility is to teach to the broader domains of proficiency as defined by state standards, not to questions or problems contained in a single, isolated test form.

"Improving Teacher Quality" is another part of the president's plan that speaks to a vitally important element of teaching and learning. Effective teaching is a key to improving student achievement. In the final analysis, assessment and instruction must mesh, and only teachers can ensure that this happens.