Here we go again. Remember school consolidation, team-teaching, flexible schedules, open classrooms, New Math, whole-language reading, community control, British Infant Schools, career ladders, education enterprise zones, individually prescribed instruction, zero-based budgeting, teacher-scripting, and self-esteem-enhancing curricula? The list of once fashionable but now faded education innovations could go on and on.
We once closed schools because they were too small. Now we are closing them because they are too large. We insulated schools from partisan politics to preserve effectiveness. Now elected mayors are overseeing schools to enhance effectiveness. Once we eschewed federal involvement in K-12 schooling. Now schools are addicted to federal funding, and there are those advocating a set of national performance standards for education.
Back and forth the pendulum goes.
So much of American education reform has the comforting but insubstantial feel of a soft summer breeze. There is a short, pleasant whiff of a budding idea. It is followed by politicians’ extensive touting of the idea as a cure-all for schools. And rapidly thereafter the program goes to scale. Then reality sets in. The innovative reforms with which well-meaning advocates hope to revolutionize schooling almost inevitably fade for lack of a scientific underpinning or a plausible research base.
Past reform efforts have tried to establish procedures for rigorously evaluating programs before taking them to scale. The joint dissemination-review panel at the U.S. Office of Education in the mid-1970s, for example, proposed to regulate programs’ dissemination to state and local education agencies until rigorous assessments were done. But these federal review initiatives were short-lived.
Present-day reform innovations are seldom challenged, and thus are often championed as though they were certain to be successful. The absence of an evidentiary orientation in the operation and management of public schools leaves education reform more susceptible to fads than comparable efforts in other disciplines.
It now appears that another school reform is about to bloom large: providing financial rewards to classroom teachers who significantly elevate students’ academic achievement. The state of Florida announced in late February that it would link teacher compensation to student scores on the state achievement test. And local and state officials in Arizona, Arkansas, California, Idaho, Iowa, Minnesota, New Mexico, North Carolina, North Dakota, Texas, and Wisconsin also are exploring pay-for-performance initiatives. The pay-for-performance plan pilot-tested in Denver was approved for funding by voters there last fall. And the federal government has appropriated $99 million for a Teacher Incentive Fund. All told, at least a third of the nation’s school districts appear to be poised to participate in teacher-performance incentive plans spurred by local, state, or federal initiatives.
This is not the first time the American education system has entertained the notion of providing performance incentives for teachers. Over the last two decades, school districts in Delaware, Kentucky, Missouri, North Carolina, Ohio, Tennessee, Texas, Virginia, and other states have tinkered with alternative-compensation plans such as merit pay and knowledge-and-skill-based pay.
These programs have typically ended in disappointment because of a host of factors: (1) insignificant dollar amounts awarded to successful teachers, (2) lack of agreed-upon and objective measures of teacher performance, (3) opposition to alternative-compensation systems by collective bargaining agents, (4) difficulty in creating an effective process for identifying high-performing teachers, and (5) lack of rigorous evaluation to assess, and possibly recalibrate, the program’s components to bring it to scale more effectively.
Viewed through one lens, the currently proposed second generation of teacher-pay-for-student-performance programs makes good sense. Approximately 96 percent of America’s public school teachers are placed mechanically on what is known as a single-salary schedule, wherein years of employment and post-bachelor’s-degree college course credits determine pay. These two conditions are correlated marginally, at best, with student achievement and teacher performance. The American taxpayer may be startled to learn that for more than a half-century, the ubiquitous single-salary schedule has resulted in about two-thirds of all education dollars being disbursed with little regard for a teacher’s instructional effectiveness—the very dimension of schooling most widely identified as influencing a child’s education.
In the absence of experimentation and independent appraisal, the idea of paying teachers for performance will remain a reform dominated by hyperbole and assertion.
Also, progress has been made in addressing many of the deficiencies associated with the failure of first-generation pay-for-performance programs. The federal No Child Left Behind law symbolizes a transition from preoccupation with the distribution of educational inputs to the expectation that every student will achieve at a level at least equal to state-specified minimum-competency standards. Within this context, evolving sophisticated testing and statistical procedures enable researchers and analysts to come closer to evaluating teachers on the basis of their students’ progress over the school year—and not their incoming or initial level of achievement.
The view through another, equally informative lens, however, can help us understand why additional experimentation is needed. We do not know, for example, how successfully statically determined estimations of teacher-performance effects can guide education practice and provide incentives for teachers to change practice. We do not know exactly how financial inducements alter teacher behavior and institutional and organizational dynamics, or whether such programs are even cost-effective in the long haul. Nor do we know much about negative, unintended consequences that may be associated with high-stakes teacher-pay programs. We do have a modicum of research suggesting that teachers can and will respond to incentives that are tied to gains in student achievement. But what we know about pay for performance in schools is far from conclusive.
From what we do know, with near assurance, each of the following components is sine qua non:
• Measures of performance must be aligned with what a teacher can reasonably be expected to accomplish, and student-performance targets triggering teacher bonuses must be realistically achievable and announced in advance.
• Award-calculation procedures must be replicable and transparent, and of a financial magnitude perceived by teachers as significant.
• Construction of pay-for-performance arrangements should not discourage teamwork among teachers and others in a school, but must discourage free-riding.
It is around these principles that carefully designed experiments and pilot projects should take place and be subjected to careful scrutiny.
What we know suggests that much of what is being proposed now by policymakers is almost assuredly going to doom present-day pay-for-performance initiatives to failure. There is an unjustified rush to policy implementation rather than experimentation.
For example, several of the most widely publicized new state proposals are suggesting teacher performance bonuses in the under- 5-percent range. The added work of attempting to elevate hard-to-educate students to ever-higher new academic standards is unlikely to be elicited by the promise of a $2,000 bonus. Why bother moving out of one’s instructional comfort zone when the reward stakes are so low and the effort so vast? Just wait a year, and conventional annual salary increases will kick in some 5 percent without the need to change one’s teaching effort at all.
The real issue needs to be whether education research, practice, and policy are aligned sufficiently to rigorously evaluate teacher pay-for-performance programs.
The tragedy in such a scenario is that pay-for-performance proponents will point the finger of fault at teachers’ unions, claiming that they have sabotaged the effort. This may, of course, have a grain of truth. In Florida, the teachers’ union went to court within hours of the unveiling of the state’s Effectiveness Compensation plan. But it is equally likely that pay-for-performance proponents will sabotage their own efforts by poor program design or an unwillingness to examine the program’s successes and failures critically before going to scale.
At present, the real issue needs to be whether education research, practice, and policy are aligned sufficiently to rigorously evaluate teacher pay-for-performance programs and to determine whether differentiated-compensation plans can be effective in a second incarnation.
There are fortunate exceptions to such an ominous characterization. In Texas, the governor’s Educator Excellence Award is a promising step in the right direction. It provides, in addition to conventional funding, anywhere between $60,000 and $180,000 to high-performing, high-poverty schools. Eligible schools must then distribute three-quarters of those funds to classroom teachers, based on an objective, quantifiable measure of student-performance gains. In addition to measures designed by school-level stakeholders and approved by the state, officials may also use a portion of the funds to reward teachers who work in hard-to-staff schools, teach in hard-to-staff subject areas, or demonstrate a sustained commitment to teaching and learning. High-performing teachers can receive from $3,000 to $10,000 in extra pay.
Ultimately, the promise of the Educator Excellence Award lies not only in the balance it strikes between state and local control and the well-crafted ingredients that guide its incentive-program designs, but also in the state’s willingness to have a third party independently observe, evaluate, and report on programmatic successes and failures before taking the program to scale. Indeed, in this instance, a Texas education reform initiative may once again prove the archetype for education policy reform.
One thing is sure: In the absence of experimentation and independent appraisal, the idea of paying teachers for performance will remain a reform dominated by hyperbole and assertion, tarnishing what may ultimately be an effective policy innovation for American schools, and making it yet another unworkable fad.
A version of this article appeared in the April 05, 2006 edition of Education Week as Teacher Pay for Performance: