Education policymakers are starting to act on what teachers have long known: No matter what policies, curricula, or governance structures are in place, they are only as effective as the teacher who translates them into the daily life of the classroom. Issues of teacher quality are absolutely critical, because research shows that teachers are the single most influential school-based factor in student success. As the education establishment increases its focus on excellent instruction, it’s imperative that the players in the teacher-quality arena don’t take the politically expedient but educationally dangerous shortcut of equating good teaching with high student test scores.
If the goal is to foster excellent teaching, a better alternative may be to finally invest in a dramatically improved approach to teacher evaluation. It’s time to adopt an approach that creates a common language of what good teaching looks like, helps everyone in a school system learn that language, and provides a clear means of supporting them as they walk the talk.
The task is difficult, but the time is right. There are signs that the standards movement that has dominated public discussion on school reform for the past decade is finally giving way to a teacher-quality movement. At every level of governance, a consensus is emerging that teacher-quality policies are something that we have to get right. Both major-party candidates in the presidential race are tackling the issue of performance pay for educators. On Capitol Hill, a Democratic plan for revising the federal No Child Left Behind Act includes devoting major federal funding to research into performance assessments for teachers. In the high-profile local school reform effort under way in the nation’s capital, schools Chancellor Michelle Rhee is looking for ways to link teacher effectiveness to compensation. In this, the District of Columbia school system would be following in the footsteps of districts such as Denver, Minneapolis, and Toledo, Ohio—all of which have adopted innovative approaches to improving teacher evaluations.
By not paying upfront for meaningful and fair teacher evaluation, we pay later in costs to the profession.
Since supporting effective instruction is the school reform that matters most for student achievement, districts placing a high priority on direct observations of classroom practice appear to be on the right track. In Toledo, for example, teachers become eligible for performance bonuses based on a multiple-method evaluation that includes peer ratings, up to six classroom observations, assessment of written-communication skills, and a standards-based portfolio. Another mixed-methods approach is the Teacher Advancement Program, or TAP, which is being implemented in more than 200 schools nationwide. In many of the TAP schools, teachers can earn bonuses based on a combination of schoolwide test scores and five classroom evaluations performed by trained observers.
The observations used in these systems are far from the hastily scribbled, ill-defined checklists that are the most common form of teacher evaluations—and which often aren’t really evaluations at all, but instead are a tedious chore that neither administrators nor teachers enjoy, value, or trust. Rigorous classroom evaluations aren’t just someone’s opinion; they are carefully conducted qualitative assessments that meet research standards for clarity and transparency. They guide trained observers to look for specific indicators of excellent instruction: effective questioning that pushes students to think critically; the ability to foster a positive, productive classroom environment; the use of engaging lessons that draw students into the learning process.
These are among the indicators of great instruction that cut across teaching situations—they’re applicable in every grade, in every subject, in every school. Far from being a “warm and fuzzy” approach to teacher evaluation that merely distracts from the “real” issue of student achievement, substantive teacher-performance assessments focus on nationally recognized standards for quality teaching. And where there is quality teaching, there is student learning.
The idea of direct performance assessments for teachers isn’t new. It just hasn’t been done widely, or well, because it takes more money and more expertise than cheaper, easier alternatives such as the “drive by” principal evaluations that are ubiquitous in most public school systems. But by not paying upfront for meaningful and fair teacher evaluation, we pay later in costs to the profession. Effective teachers never get the recognition they deserve, and they leave the profession at higher rates than their less effective peers; schools have no way to hold underperforming teachers accountable, short of firing them; and the teachers in the middle rarely receive the type of constructive feedback that could inspire them to refine their craft.
The glaring inadequacies in low-stakes, low-quality administrator evaluations are one factor that has added to the popularity of using student test scores as a more reliable alternative. It’s unclear, however, that replacing quick-and-dirty classroom observations with quick-and-dirty test scores represents progress toward the goal of collecting reliable data that would help all concerned recognize and foster excellent teaching. The problem with using achievement tests for teacher evaluations is that they simply aren’t designed for that purpose, and so they are too blunt an instrument.
Instructional leaders who have been hip-deep in test data for years have already found that reading and math scores alone don’t offer the level of detail they need to make wise decisions about teacher professional development. For that, they typically rely on what they’ve seen in the classroom. Why? While test results allow only for a discussion of the “what” of student achievement, looking directly at classroom instruction can tell teachers and administrators the “why” and “how.” With a substantive assessment of which aspects of instruction are working and which aren’t, teachers can be more purposefully directed on the path to excellence.
Although proponents of standardized tests argue that they are the most practical method of getting reliable data, there are significant logistical roadblocks to using them for measuring teacher effectiveness. The biggest is that fewer than half the teachers in any school district teach a tested grade or subject, which begs the question of what the increasingly popular “value added” approach will do with the Spanish teacher, the history teacher, or the kindergarten teacher.
Worse, linking high-stakes employment decisions to children’s performance on a single test could create a truly dangerous dynamic in classrooms. We’ve already seen the unintended but depressing effects of the testing craze on a national scale: Art, music, and even recess are being abandoned in an effort to make more time for test prep. Increasing the pressure by upping the stakes will only increase the negative backlash. A struggling child threatens the bottom line on a bonus check; a high-scoring child becomes the goose that will lay the golden egg. Students could retaliate against an unpopular teacher by collectively bombing a test. The temptation to fudge the numbers would certainly increase.
As the national and local debates over teacher merit pay continue, decisionmakers should instead support evaluation measures that recognize and reward teachers who create rich, rigorous, and relevant learning experiences for children. Tests simply can’t capture the information we need. If what we really want is quality teaching, shouldn’t we take the time to look for it?
A version of this article appeared in the October 22, 2008 edition of Education Week as Seeking–and Finding– Good Teaching