Performance standards that are now being crafted could reshape the way children learn.
Imagine American schools in which students didn't have to spend time filling in workbook pages and practicing for bubble tests, but instead worked on extended projects, discussed complex problems, and generally thought their way through a demanding curriculum aimed at the kinds of knowledge and skills they will need as citizens and workers of the future.
Imagine American schools in which the same high expectations were applied to poor, minority, and immigrant children as to the children in upscale suburban schools. Imagine American schools in which teachers were trusted professionals, committed to seeing that their students met the high expectations set for them and able to deliver on this commitment because of their access to high-quality continuing professional development. Those are the schools that the New Standards Project (NSP) aims to bring into being across the country.
Vision for the '90s
The New Standards Project is a consortium of 17 states and a half dozen leading school districts, which together serve nearly half of America's schoolchildren. State and district partners are working with a cluster of specialists—learning and teaching researchers, curriculum specialists, assessment and testing experts, staff development professionals, and leaders in systemic education reform—to design and implement a system of performance standards, authentic assessments, and professional development intended to change the way the American school system works.
We do not believe that American schools are doing worse than ever before. In fact, evidence supports the idea that our schools are, on the average, doing substantially better than they were 60 or 70 years ago.
But schools are not doing what we will need them to be able to do in the future. We have a curriculum—and indeed a conception of learning and knowing—that is more in touch with the 1920s than our modern day. We also have a system that is organized to teach a demanding program to only some of our children and that fails to develop fully the energies and capacities of most of the people who work in it—especially teachers, but other educators as well.
A system designed for the 1920s won't work for the 1990s and beyond. Our goal is to build a revitalized education system using assessment as a tool for transforming instruction and learning. Ten principles guide our work. Their essence is captured in a “Social Compact,” whose basic premise is this: “While students should make every effort to meet the high standards set for them, our 17 state and 6 school district partners pledge to provide the opportunities all students need to prepare themselves well.”
To transform this promise into reality, our partners need world-class content and performance standards, a performance-based examination system that embodies those standards, together with rubrics and procedures for scoring students' work reliably and fairly. Such tools are virtually useless, however, without teachers, content specialists, and other educators who have a firm understanding of how to construct and apply our examination system to improve curriculum, instruction, and, most important, student performance. To build this understanding, we must create a professional development system that will transform the way educators view teaching, learning, and assessment.
Content Standards
A number of professional organizations and groups are collaborating to develop content standards in their subject areas. The New Standards Project will build on these, adopting frameworks that satisfy criteria set by our governing board. The standards we adopt will be internationally benchmarked and derived with broad public involvement, and we are working closely with a number of groups to ensure that our emerging examination system reflects the best current thinking about essential skills and knowledge both within and across disciplines.
International benchmarks are derived from other countries' examinations, official curriculum guides, widely used texts, and new curriculums. Our process for benchmarking differs from the one used in current international testing in that we are aiming to establish what each country's own goals and outcomes are, not how each performs on a common denominator set of test items. Our analyses will go beyond exams to the curriculum and textbooks that are currently in use in other countries and to actual student work in classroom settings. Later, student work from other countries may be scored by New Standards teachers, and New Standards' student work, in turn, will be scored by teachers in other countries.
Assessing Performance
To move from the conceptual to the concrete, we need actual examples of student work that exemplify outcomes specified by content standards. Performance standards provide examples and explicit definitions of what students must do to show that they have learned to an adequate level the specified skills, strategies, and knowledge. Performance standards will help educators and the broader public make consistent judgments about the range and quality of behaviors that exemplify desired goals for learning. Without performance standards, the meaning of content standards is subject to interpretations, which, if allowed to vary, would undermine efforts to set high standards for the majority of American students.
But performance standards cannot be obtained without tasks or activities that elicit the desired behaviors. Unlike traditional multiple-choice tests, performance assessments require individuals to engage in tasks that mirror as closely as possible the conditions under which a particular competence is performed in “authentic” settings. Tasks and projects of this kind will be the centerpiece of the NSP assessment system.
New Standards Portfolios
work chosen by the district, school, teacher, and/or student;
prescribed projects and other extended learning activities; and
responses generated by NSP matrix examination tasks.
Over the next three years, the New Standards Project will develop performance-based matrix examinations in mathematics, English language arts, and science that will be administered in grades 4, 8, and 10. (Work in other subject areas will begin in response to partner demands.) The Amusement Park Task (see sidebar) is a good example of the demands posed by performance assessment and an example of the types of questions to be asked on the matrix exams.
The initial Amusement Park prompt invokes mathematical thinking and decision making, whole number computation, and measurement. The task also has a social dimension: Students are asked to consider the needs of a younger buddy. A second question on the task asks students to write a letter explaining their choices, showing us the integration of mathematics and writing as called for in the NCTM curriculum standards.
The design of the NSP portfolios is intended to protect the integrity of the matrix examinations by discouraging their use without the larger portfolio in which they are embedded. Scores for individual students cannot be derived from an NSP matrix exam because a given student takes only a subset of all the tasks, which together form a complete exam. In our system, the full picture of a student's attainment can be obtained only by analyzing a student's performance on the full range of tasks contained in the portfolio. Together, the matrix examinations and the cumulative accomplishments records, or portfolios, will yield results that can be aggregated to tell a story about a population in a state, district, or school.
At present, our matrix exam development is ahead of the larger portfolio initiative largely because we know a great deal more about how to do the former than the latter. Moreover, experience tells us that the use of portfolios in classrooms requires a considerable amount of professional development to create the kind of culture and resources needed to support thoughtful use by students and teachers.
Creating Leadership
Building an examination system that is technically sound gets us only part of the way toward our goals. We must also help educators acquire the expertise needed to use the exams to improve learning for all students.
The contents of the NSP portfolio (matrix exam tasks, projects, exhibitions, and other products) will require extensive engagement by teachers and other educators, especially curriculum and assessment specialists. Teachers, curriculum developers, and those involved with assessment will no longer work in their own separate niches. Educators from each group must share a common set of knowledge and skills related to task development, the analysis of student work, and how to conduct workshops to raise the skills of others.
Our examination system will be useless to schools unless they have a sufficient number of skilled scorers to analyze student work. The actual number of scorers this system must produce will vary with the number of schools and students that participate at each site. But to develop the needed skills in scoring, we are using a “trainer-of-trainers” approach. For the past two years, a team of 12 teachers from each partner state or district has participated in a professional development program emphasizing scoring and task development. These “Senior Leaders” are being trained in task and rubric development; portfolio design and implementation; scoring student work; linking standards and alternative assessment to curriculum and instruction; and relating NSP activities to other components of school reform. In turn, the Senior Leaders will provide scoring training for teachers and other educators, along with workshops on standards, authentic assessment, and their relations to school reform.
Once certified, the Senior Leaders will be responsible for conducting at least two scoring workshops per year. When fully certified, our initial group of 276 NSP Senior Leader candidates should provide training for roughly 33,000 “School Leaders” per year. That group of 33,000 would be a well-spring for future NSP Senior Leaders. NSP School Leaders will coordinate the administration and scoring of exams along with the implementation and analyses of portfolios at the individual school level. These educators will conduct scoring workshops in their schools and help their colleagues interpret the results and their implications for curriculum and school reform. NSP School Leaders will also act as a bridge between the principles and goals of the New Standards Project and other restructuring initiatives in their school.
Equity Initiatives
The substantial promise our project holds for advancing school reform cannot be fulfilled without special attention to equity. We are taking several steps to ensure that the standards we adopt and the tasks and activities we develop reflect the experiences and aspirations of diverse groups in our society.
The project will conduct focus groups with local and national equity advocacy organizations to explain our work and to get feedback about how we might address equity concerns more fully. In addition, we have made every effort to ensure that our various advisory groups include scholars and practitioners with expertise in culture, gender, and language issues. We have also established a specific advisory group to help us think about ways to design and administer tasks that will enable students who are learning English to display their competence in English and their native language.
In our view, the social compact captures the essence of what we are about with regard to equity; our partners and the entire nation have an obligation not only to set standards, but also to see to it that every learner receives the support he or she needs to have a fair shot at meeting them. A number of educators are beginning to grapple with the difficult problem of defining what the National Council on Education Standards and Testing (NCEST 1992) called school delivery standards—that is, standards for the kinds of instructional opportunities schools need to provide to enable students to reach the learning standards set for them. The challenge here is to define these opportunities in a way that avoids the creation of input lists (for example, number of library books and textbooks, students per class, and so on). Such input lists haven't worked to improve education significantly in the past. Instructional opportunities should be characterized in a way that remains focused on outcomes and recognizes that the number and kind of resources students require to meet the standards will vary with their needs.
We believe the New Standards Project can make a valuable contribution to attempts to define school delivery standards by creating a community of schools using innovative approaches to assessment, curriculum, and instruction to meet world-class standards. In the years to come, we will describe the kinds of innovations that work best for students, schools, and educators with different backgrounds. Our partners and the educators who represent them have already made an enormous commitment in time, energy, and resources to reach our shared goals. The strength of their commitment is highlighted by the fact that it has occurred at a time of painful budget cuts in education. But the states, districts, and educators in the New Standards Project see no better way to produce the kinds of schools our children deserve.
End Notes
•
1 The project is co-directed by Marc Tucker, President of the National Center on Education and the Economy, and Lauren Resnick, Director of the Learning Research and Development Center at the University of Pittsburgh. The New Standards Project began two years ago with support from the Pew Charitable Trusts and the John D. and Catherine T. MacArthur Foundation.
•
2 A matrix exam is given to a stratified sample of students so each student receives a specific portion of the tasks that make up the entire exam. The entire set of tasks for a particular matrix exam is designed to cover a subject area. By giving different subsets of tasks to students in a stratified sample (for example, a group selected to represent important population demographics), one can aggregate the results to make inferences about the attainment of the population from which the sample was chosen. One cannot use matrix examination results, however, to make judgments or inferences about individuals because the subset of tasks each person in a sample receives covers only a portion of the subject area being assessed. Results for individual students in our system cannot be obtained without combining matrix exam responses with analyses of student work that cover what is not reflected in the matrix exam tasks.