The Role of Performance Assessment in Developing Teaching as a Profession

By Linda Darling-Hammond, Maria E. Hyler

Illustrator: David McLimans

Illustrator: David McLimans

We are teachers who care deeply about the quality of our own teaching and the development of teaching as a profession. We write this from our conviction that performance assessments in teaching are a critical strategy for developing teachers’ expertise and for improving the quality of teacher preparation.

Linda entered teaching in large East Coast high schools, where she experienced the funding inequities, factory model design, and ill-informed top-down mandates that constrain teaching quality. Desiring to strengthen the profession, she was an early member of the National Board for Professional Teaching Standards. Later, she launched a support group for National Board certification at Stanford, helped design the Performance Assessment for California Teachers (PACT), and contributed to initial thinking about the now nationally available edTPA.

Maria was a high school teacher in California who became National Board certified after attending Stanford’s support group. When she later entered Stanford’s doctoral program, she worked in the National Board support group, helping other teachers going through the process. She co-taught with Linda in the teacher education program, was involved in the early implementation of the PACT, and is now a teacher educator working with edTPA.

We have seen how improvements in preparation and teaching can result from assessments that examine how teachers actually plan, teach, and respond to student learning. We believe these assessments strengthen teaching and empower teachers, and we think they will ultimately improve learning for students, especially those traditionally least well served in our schools.

The Puzzle of How to Improve Teaching

As a feminized occupation that is poorly compensated relative to other professions, teaching in the United States has long struggled to become a profession. Shortages of teachers are often addressed by lowering the standards for entry rather than increasing the inducements to teach. Ever since scientific managers took charge of schools in the early 19th century, there have been policymakers who have preferred a high-turnover teaching force that is minimally trained and compensated—and easily controlled by standardized tests, curriculum packages, and bureaucratic regulations.

Currently, with today’s new breed of “reformers,” there is a ferocious return to this managerial mindset. A raft of initiatives seeks to manage teachers from afar—to the point of evaluation systems aiming to make personnel decisions based on “value-added” test scores, despite evidence that these measures are frequently inaccurate.

The bureaucratic management of teaching might make sense if students learned in exactly the same way and at the same rate. But, as every teacher knows, students are not standardized and teaching is not routine. Teaching requires deep knowledge of how children learn differently and a sophisticated repertoire of skills deployed through professional judgment. Standardized teaching can never produce high levels of learning for all students. For students to be well served, teaching must become a real profession.

So why do we think performance assessments are important to this goal? Generally, professions have three features: They are morally committed to the welfare of those they serve; they share a common body of knowledge and skills that they use to advance the best interests of their clients; and they define, transmit, and enforce standards of professional practice. Professionals are willing to practice as part of a broader community that is pledged to meet professionwide expectations.

Professions make a compact with the public that allows them to manage their own work in exchange for holding themselves accountable for mastering the knowledge and skills that allow them to practice safely and effectively. The extent to which an occupation is micromanaged by rules from without is directly related to the extent to which it fails to maintain high, common standards of competence and professional practice.

For this reason, rigorous licensing and certification tests have been critical to the professionalization of occupations—from medicine in the early 1900s to nursing, law, engineering, accounting, architecture, and others thereafter. Licensing exams allow candidates to show that they know and can do the things that are expected of professionals in their field: analyze cases and write briefs, diagnose patients, design safe bridges. Although they are administered by testing companies, they are designed and scored by members of the profession.

These assessments are intended both to determine who can enter the profession and to guide the curriculum in professional schools, so the public can rest assured that the people to whom they entrust their lives and livelihoods have learned what they need to know.

Unfortunately, the tests mandated for teachers have not, for the most part, been designed by members of the profession; nor have they captured the essence of the knowledge and skills needed for teaching. They have largely been multiple-choice tests of basic skills or subject matter that, though useful for some kinds of screening, do not predict who will be able to teach well. Even “pedagogy” tests often assume that problems of practice can be described in a few sentences and answered by choosing one response out of five. These tests trivialize teaching rather than seriously engage teachers in the development of effective practice.

Performance assessments developed by teachers offer an important alternative. By evaluating teaching authentically, they represent the complexity of teaching and offer standards that can define an expert profession.

The Struggle for a Profession

The National Board—composed largely of accomplished teachers—was the first truly professional effort to set standards and create assessments for teaching. More than 20 years ago, the board rejected multiple-choice tests in favor of assessments of real performance. National Board portfolios combine student work samples, videotapes of instruction, and teachers’ analyses to assemble a picture of teachers’ practice as it is shaped by the particular needs of their students and the particular context of their schools. As in other professions, this evidence is scored by trained raters who are expert in the same field, using criteria that define critical dimensions of teaching.

Linda was on the board when these decisions were made and remembers how teachers had to fight for this approach. Because they won, the National Board process has been embraced by teachers, inspiring them to improve their practice in long-lasting ways—92 percent of candidates report the certification process has made them better teachers.1 Teachers who are involved in scoring the assessments also feel they learn more about good teaching because they interact with authentic evidence of teachers’ practice, examined through the lens of professional standards.

The Evolution of edTPA

The edTPA, also developed by teachers and teacher educators, is built on the National Board model and subsequent work by California educators. The PACT was developed by 12 public and private universities in 2002 when the state legislature required that all candidates be licensed through a performance assessment. It asks candidates to plan a unit of instruction, adapt the plans for English learners and students with disabilities, and track three to five days of instruction. Candidates discuss how and why the plans are revised as teaching unfolds, submit a continuous video clip of a teaching segment, and collect and analyze evidence of student learning. Candidates also describe and show how they develop students’ language proficiency and academic language in the discipline.

Together, these tasks, conducted over a week of instruction, provide vivid evidence of what beginning teachers can do in enacting the fundamental elements of teaching. The PACT is scored in a consistent fashion by faculty members, instructors, supervisors, cooperating teachers, and principals in partnership schools, all of whom say that they learn about candidates’ practice, the results of their preparation efforts, and the nature of good teaching.

More than 30 universities and alternative programs now use the assessment, which has proved so productive that programs have sustained it for a decade even when promised state funding never materialized. As teacher educators in California, we found that the PACT improves our own and our candidates’ practice. We’ve also found that, unlike other licensing tests, PACT scores do not show discrepancies by race or ethnicity, and that they do, in fact, predict teachers’ later effectiveness.2

As interest spread across the country, a national version began to evolve. First, a California teacher educator moved to the University of Washington and introduced the PACT. Faculty there and in other institutions liked the assessment and encouraged the state standards board to adopt the approach. Other California educators got jobs out of the state and took the idea with them. Word spread through sessions at the American Association of Colleges for Teacher Education. Ultimately, a group of teacher educators and teachers from across the country, in collaboration with staff at Stanford University, decided to create a national version, which has since become edTPA. As interest grew beyond the capacity of this small team, Stanford put out a request for proposals for an administrative partner and Pearson was selected, in part because of its expertise in supporting the National Board’s portfolio. The 2012 field tests for edTPA included 22 states, 160 institutions of higher education, and more than 9,000 teaching candidates.

Pearson’s involvement has prompted some in our community to characterize edTPA as the corporatization of teacher education, somehow replacing the relationship between instructor and students. Nothing could be further from the truth. Like the National Board portfolio, edTPA was not developed, nor is it owned, by Pearson. Like assessments in other professions, it was developed—and is guided—by a consortium of professional educators. These individuals make decisions about design, the scoring process, and the qualifications of those who are asked to score (accomplished teachers and teacher educators).

Instructors and supervisors continue to teach, observe, support, and evaluate candidates, as they always have. The assessment—which allows teachers to be evaluated in their own student-teaching or internship classrooms using curriculum they have designed—focuses on areas all beginning teachers need to learn: how to plan around learning goals and student needs, how to engage in purposeful instruction and reflect on the results, how to evaluate student learning, and how to plan for next steps for individual students and the class as a whole.

Experiences with edTPA

The vast majority of candidates and teacher educators feel this process strengthens their practice.3 In Ohio, when 32 institutions piloted edTPA and studied the outcomes, 96 percent of teacher candidates felt they learned from the experience, pointing especially to how it made them self-aware and more focused on student learning.4 As one observed:

I was analyzing student learning and developing lessons that met the needs of each individual student. It helped me develop lessons that were within the students’ ability level, but pushed them to think more in depth.

Teacher educators felt they learned as well:

It forced the teacher candidates to examine what they were doing as beginning teachers. It also forced me to look at the materials that I was including in my seminar and the relevance of these materials to my students.

Similar conversations occurred across eight Tennessee universities piloting the assessment. Marcy Singer-Gabella, a teacher educator at Vanderbilt University, described how instructors revised coursework and field assignments when the assessment revealed that candidates had difficulty analyzing student work and giving students usable feedback.

Vanderbilt graduate Nicole Renner noted:

Even though the TPA is used for summative assessment, it is also formative, and the main lesson of the TPA is exactly what new and pre-service teachers need to learn: “It’s about the students, dummy!” The TPA process shape[s] the candidate’s field experience [so that the] focus [is] entirely on students. Yes, we videotape lessons, and we refer to that as “videotaping ourselves,” but what we are really trying to capture on that tape is our ability to foster a student-centered learning experience.5

Stephanie Wittenbrink, a prospective special educator in Washington state, described the “radical perspective shift” she experienced due to edTPA, arguing: “All student teachers should complete such a classroom performance-based assessment. The field needs it and students deserve it.”6

How Teacher Performance Assessment Supports a Profession of Teaching

We share the concern of edTPA’s critics that teacher preparation programs are under attack. At this point, however, we believe that the most effective and ethical response is not to stick our collective heads in the sand and complain that nobody trusts us. Rather, we believe that our collective response should be to embed high-quality performance assessments in high-quality teacher preparation programs, and to ensure that our candidates demonstrate they can meet professionwide standards before being permitted to practice.

We agree with our colleague Jon Snyder, a longtime teacher and teacher educator at Bank Street College, that states must assure families that teachers meet a standard of competence because parents are compelled by law to entrust their children to educators in schools. Jon argues: “I firmly believe that a state saying ‘just let anybody in’ or even ‘we trust the preparation programs’ would be an abdication of its moral responsibility. Thus, I have no problem with states using better assessments to provide for a better future for our children.”

In the context of the current debates about teacher education quality, it has been inspiring to us to see educators step up and accept the challenge to create better assessments, rather than complaining about narrow measures that do not support learning or improvement. The best hope for significantly improving education at all levels is for educators to take charge of accountability and ensure it reflects our highest aspirations for ourselves and our profession.  


  1. National Board for Professional Teaching Standards (NBPTS). The Impact of National Board Certification on Teachers: A Survey of National Board Certified Teachers and Assessors. Arlington, Va., 2001.
  2. Stephen P. Newton. Predictive Validity of the Performance Assessment for California Teachers. Stanford, Calif.: Stanford Center for Opportunity Policy in Education, 2010; and Linda Darling-Hammond, Stephen P. Newton, and Ruth C. Wei. Developing and Assessing Beginning Teacher Effectiveness: The Potential of Performance Assessments. Stanford, Calif.: Stanford Center for Opportunity Policy in Education, 2012.
  3. For more information, see
  4. Donna Hanby. “Pioneer Reflections from Ohio’s 2010–2011 Journey: Pilot Year Insights from Leg I and Leg II Travelers.” Presentation at the Ohio Council of Teacher Educators Conference, Columbus, October 2011.
  5. Nicole Renner. Presentation to the deans and directors of the SUNY Colleges of Education. Albany, N.Y., June 18, 2012.
  6. Stephanie Wittenbrink. “Rigor and Results: How edTPA Accelerated My Preparation as a Pre-service Teacher,” in Tomorrow’s Teacher, 2013. p. 29.
  7. Maria E. Hyler, L. S. Yee, S. R. Barnes, and R. L. Carey. “Teacher Performance Assessment and Culturally Relevant Pedagogy” (forthcoming).

Linda Darling-Hammond ( is Charles E. Ducommun Professor of Teaching and Teacher Education at Stanford University, founding director of the Stanford Center for Opportunity Policy in Education (SCOPE), and faculty sponsor for the Stanford Teacher Education Program.

Maria E. Hyler is an assistant professor in the Department of Teaching and Learning, Policy and Leadership at the University of Maryland and faculty director of edTPA.

David McLimans’ editorial illustrations can be found at