In August 2010, the Los Angeles Times sparked a heated education debate when it published a series of articles about using value-added assessment techniques to evaluate teachers’ performance. The paper also published scores for individual teachers that attempted to demonstrate whether a teacher had advanced the performance of students as measured against the students’ previous performance in school over a six-year period. In September, the LAUSD Board of Education asked the schools superintendent to expedite negotiations with teachers to develop a performance evaluation process that uses multiple measures. We asked two leaders in education — Magaly Lavadenz and J.E. Stones — to address the question, “Should value-added assessments be included in teacher evaluations?”—The Editor
There is vast agreement among policy makers that teacher effectiveness should be assessed — and that the elements of the assessments to measure and analyze teacher effectiveness are unreliable. As a democratic nation, we value justice and fairness. Thus, we must reject the notion that a single-point assessment, such as a standardized test score, should be the determinant of teachers’ effectiveness, performance or merit pay. This would be tantamount to rating dentists’ effectiveness on whether their patients develop cavities within a year.
In teaching, as in other professions, expertise develops through a process of apprenticeship, mentorship, evaluation and support. Thus, the purposes of teacher assessment and evaluation mandate clarity of terminology and intentionality. How will we differentiate teacher assessment results in order to provide support for future learning and professional development needs vs. determining pay/merit increases? How will evaluation results be weighed in discussions about tenure, renewal and compensation?
There is general consensus in the education community that multiple measures that are fair, valuable and reliable should be used in determining teacher effectiveness. If we apply the current argument that solely student achievement data be used for evaluating teachers, we have, to a large degree, violated some of the psychometric principles of fair testing. Most standardized measures of student achievement have been found to have cultural and linguistic biases against certain populations, including students with special learning needs and English learners. Thus, using this single point to determine teacher effectiveness is disastrous, because the measure that is used is flawed to begin with.
We know that California’s public schools are basically unequal and inequitable. For example, legal findings from the settlement of the 2000 class action suit by San Francisco County students against the state of California (the Williams Settlement) produced evidence that schools remain inherently unequal in resources and facilities, as well as adequate preparation, experience and distribution of teachers and administrators. Factoring these context variables into the value-added formula will be necessary if California is to establish a truly equitable teacher assessment system.
Policy makers should consider that mathematical models, such as value-added teacher assessment, are faulty and technically flawed. These models are receiving a great deal of public attention, along with discussions about school accountability, comparisons of U.S. and international students on test performance, and federal education policy. While it seems like a simple fix to develop a formula to determine teacher effectiveness, the educator professional community must define effectiveness like other professions do, including the legal and medical professions.
Promising practices have emerged to define systems of teacher evaluation. Piloted by the California Charter Schools Association, one example applies a teacher development/growth approach and multiple points of evidence of student performance. This system approximates a “professional learning community” model and is a fairer system that parallels the legal and medical professions by including supports in developing expertise. A fair system of teacher assessment and evaluation should include:
• Multiple measures of student performance in addition to standardized tests, and consideration of the validity of assessment measures for use with English learners and students with special needs.
• Contextual factors that take into account certain “opportunity factors” or “risk factors.” Privilege and school inequalities are institutional realities that impact student achievement.
• Recognition that administrators and/or others who have the decision-making authority and power to observe and evaluate teachers should have the requisite skill set to conduct teacher observations and provide assistance to teachers based on research-informed protocols and processes.
• Appropriate representation and preparation of stakeholders, such as teachers, parents, labor and others, to collaborate and have genuine dialogue about the career of a teacher.
National, state and local formulas for teacher effectiveness should be developed; however, just as “one size fits all” has been proven not to work for literacy instruction for all learners, neither should teacher assessment and evaluation systems fall into the “one size fits all” trap.
Finally, an essential question that should frame our discussion of value-added teacher assessment and evaluation must center on our vision for educating the next generation of learners. Will we have a value-added curriculum? How do we map the complexities of the evaluation of schooling (not just teachers) onto a much more expansive vision of global citizenry, imagination, creativity and democracy?
Magaly Lavadenz is professor and director of Bilingual/TESL Education and the Center for Equity and Excellence for English Learners at the LMU School of Education. She also is president of the California Council on Teacher Education (www.ccte.org), located in San Francisco.
Value-added assessment is a sophisticated statistical methodology that permits non-educators to make fair and accurate comparisons among teachers with respect to how much they improve the achievement levels of their students. From the standpoint of all who want schooling that emphasizes student preparation for academic and career success, value-added assessment should be at the heart of teacher evaluations.
Why? Because evaluations based on value-added assessment ensure that teachers and schools maintain a sharp focus on the knowledge and skills that are necessary to students becoming productive and self-supporting members of society, and not the wider set of academic, developmental and social concerns that can preoccupy today’s teachers. Value-added assessment is focused not on teaching inputs, but teaching results, i.e., the increases in student achievement that the public thinks of as an indispensible marker of good teaching.
Haven’t schools always focused on academics and regularly tested student academic progress? Yes, but until the emergence of value-added assessment, many educators viewed student achievement as only one of many outcomes by which student progress is measured. Teacher effectiveness was subjectively evaluated and focused mainly on teacher attributes and classroom practices. By contrast, value-added assessment is objective and focused on results. It makes student academic progress a top, although not exclusive, educational priority.
Before the development of value-added assessment, there was no fair and objective way to separate the teacher’s impact from the many other factors that influence student test results. For example, researchers have found that more than 70 percent of the achievement test differences among schools are attributable to the poverty level of the students. Plainly, a teacher with a classroom full of economically disadvantaged students cannot be assessed on the basis of standardized test scores alone.
How does value-added assessment solve this problem? In the case of disadvantaged students, standardized achievement tests often show students to be a grade or more behind their advantaged peers. The usual reason for this gap is that they were a year or two behind their peers when they entered school and have never been given the intensive instruction they need to catch up.
Here is where value-added teacher evaluation makes a critical difference. Instead of attempting to compare average test scores among teachers with similar students, value-added assessment analyzes the year-to-year increases in achievement of each student taught by a given teacher. It identifies those teachers who do the best job of helping students gain in achievement — regardless of the students’ starting level or other advantages or disadvantages.
To put it differently, standardized test scores are like mile markers on the highway: They can tell you where you are, but not how fast you’re moving. By contrast, value-added assessment is like a speedometer. Teachers are compared in terms of how effective they are in moving their students along on a level playing field. Within certain statistical limitations, parents and school officials can see which teachers are adding the most (or least) to their students’ academic progress.
Regardless of whether students are disadvantaged, disabled, advantaged or talented, their teachers can be compared. Having this kind of information can make an enormous difference in the ongoing efforts to improve public education.
Simply giving teachers who are doing the best job the recognition they deserve would be a real advancement. And using such information in determining retention, tenure and advancement among teachers would have a revolutionary impact. Teachers would have an incentive to use proven teaching practices, not just those that are familiar or convenient. Teacher preparation programs would be encouraged to focus less on social and philosophic concerns and more on effective teaching.
Value-added assessment of teacher performance would refocus schooling on the public’s priorities by encouraging the work of those teachers who are focused on equipping students for adult success and by encouraging the others to improve or find a new line of work.
J.E. Stone is president of Education Consumers Foundation in Arlington, Va. He is also professor in the Department of Human Development and Learning in the College of Education at East Tennessee State University.