Published Online:
Published in Print: April 10, 2007, as Not Performance Pay Alone


Not Performance Pay Alone

Teacher incentives must be matched by systemwide change.

Article Tools
  • PrintPrinter-Friendly
  • EmailEmail Article
  • ReprintReprints
  • CommentsComments

The ice is cracking. Teachers’ unions and the educational establishment in general are starting to move away from the single-salary schedule. In the face of global economic change and mounting pressure from the federal No Child Left Behind Act, governors in upwards of 20 states have called for differentiated teacher compensation, and state legislatures in Florida and Texas have allocated millions of dollars for this purpose. Most of these efforts have focused on “pay-for-performance,” a strategy to reward our best teachers with higher salaries.

While necessary, pay-for-performance is far from sufficient. Alone, it will not, and cannot, result in significantly higher achievement for all students, because it does not provide other critical system changes designed to improve the instructional effectiveness of all teachers. To think otherwise is to believe that all teachers know what to do and how to do it, and have been withholding their expertise, waiting for higher pay.

According to the National Commission on Teaching and America’s Future, of the current 3.4 million teachers, 2 million will leave the profession in the coming decade—three times as many through attrition as retirement.


Who will replace them? Pay-for-performance is the right prescription, but it is a long-term solution. It will take decades to attract enough talented people. In the meantime, much more must be done to help our current teachers improve their instructional effectiveness.

This is the human-capital-development challenge now facing public education. While we must provide additional compensation for our best performers, we must at the same time provide extensive support for all our teachers. But to invest in professional development without first creating the incentives for all educators to master what they are being taught is to permit the status quo to remain firmly in place. Many observers have noted that despite the widespread availability of “best practices,” surprisingly few are actually used. Until we align our reward structures so that it is in the direct interest of every adult working in public schools to seek out and master best practice, the goal of all students’ achieving at high levels will remain elusive.

What must be done to create a system in which educators can succeed in helping their students achieve the new and demanding standards required in the 21st century? We offer the following suggestions:

Do pay-for-performance right. With gender discrimination in the workforce greatly diminished, women of talent have been flocking to other professions. Teachers accounted for just under a third of all professional women in 1972, but only one-seventh in 2004. To attract new teachers from among our best college graduates, we must offer much higher salaries (not simply bonuses) to the best performers and speed the rate at which they can reach the top tiers of an expanded compensation system. Despite the belief promoted by the National Education Association’s Reg Weaver and other union leaders that pay-for-performance requires a “Darwinian competition” among teachers, the use of an external growth standard—a target based on how well a state’s students did at a point in the recent past—means that teachers would compete only with themselves, and not with each other in exceeding this common target.

Join the related discussion, “Not Performance Pay Alone.”

New evaluation procedures for all teachers. New compensation systems will require new evaluation procedures. These should have multiple measures, include appropriate safeguards, and be balanced between inputs (observation) and outputs (empirically measured student-learning results). Because student growth is best predicted by the quality of instruction (in contrast to student achievement, which is best predicted by family background), value-added models can and should be used for teacher evaluation. These models track the academic progress of each student over time and are effective in identifying the highest- and lowest-performing teachers. Roughly 70 percent of all teachers can be evaluated this way.

The balance of teachers are found in grades K-2, in which standardized exams are not administered; in subjects outside the academic core, including music, art, and physical education; and in specialist positions, such as librarians, guidance counselors, and social workers. New evaluation procedures should incorporate standards adopted by their respective national professional associations. They should be tied, in part, to aggregate student growth by school or district, and wherever possible should include an empirical component for individual instructors or specialists that attempts to capture change from the start of the school year when the students enter their classrooms or offices, to when they leave.

Mandatory remediation for struggling teachers. Teachers whose instruction is found wanting should be required to undergo remediation. Research shows that students with two consecutive teachers identified by value-added models as ineffective never attain the level of accomplishment they could have reached had they had better teaching. Remediation programs should be designed not as “gotcha” mechanisms, but as opportunities for teachers to improve their craft. Teachers should partner with administrators in designing and managing these programs.

Until we align our reward structures so that seeking out and mastering best practice is in the direct interest of every adult working in public schools, the goal of all students’ achieving at high levels will remain elusive.

Dismissal of teachers who fail to improve. The companion reform to mandatory remediation is a fair and timely process to dismiss teachers who fail to improve. Teachers must play the pivotal role in this process, just as in other professions where colleagues render consequential judgments on their peers. The current system, which places the decision solely in the hands of administrators, is not working. The difficulty of winning a dismissal in due-process hearings has led to the widespread belief that it is not worth the time, money, or aggravation to fire a teacher whose instructional practices deprive children of the learning to which they are entitled, thus allowing poor educators to do irreparable harm to children.

Expand the role of teachers. Since the success of teachers will be determined in their classrooms, rather than solely at the bargaining table, they should be given equal say in key decisions that affect their instructional practice, such as professional development, curricula, and assessment not mandated by the state. In evaluations, teachers should assume responsibility for the observation of their colleagues through a districtwide peer-review process.

New evaluation procedures for all administrators. The reward structures for every administrator must be tied to the system’s goal of increased student learning. Administrators must provide their teachers with an appropriate working environment and the professional development necessary to promote the academic progress of their students. Administrator evaluations should be balanced with observation protocols and empirically derived student-learning results through the use of value-added models, and should include multiple measures and appropriate safeguards.

School-based focus for professional development. Current compensation systems reward educators for courses beyond bachelor’s and master’s degrees, leaving the choice of study largely up to the individual. Professional-development programs in each school should be designed based on the academic needs of students and the capacity of educators to address these needs.

Multiyear mentoring for new teachers. Beginning teachers, on average, do not provide effective instruction. Yet the induction process of “sink or swim” used by most schools provides very little support to novice educators. Mentors and coaches, recruited from the ranks of highly effective teachers, should assist new teachers for a minimum of two years.

Coaches for all teachers. It is time to create a new category of “teacher coach,” whose sole responsibility is to help colleagues improve instruction. Among 46 interventions studied by one New Zealand researcher, the most effective at improving student learning were “direct instruction” and “feedback” from a competent colleague (0.81 effect size). Instead of focusing resources on the politically popular but less efficient policy of reducing class size (0.12 effect size), these dollars should be targeted at the training and deployment of high-quality teacher coaches.

Changes in the school calendar. Teachers need more time regularly throughout the school year to interact with each other as colleagues. Using powerful diagnostic data provided by value-added and more frequent formative assessment, teachers will collaborate to improve their instructional effectiveness. This change is necessary to create and sustain “learning communities,” which in turn will make the profession more intellectually stimulating and rewarding, and will reduce burnout.

Strategic reviews of professional-development spending. Expanded professional development will be expensive, but before the net new costs of these investments can be known, it is necessary to determine the current spending level. Districts often expend two or three times more on professional development than they thought they were spending. Strategic reviews will tell districts what their gross outlays have been, so that calculations can be made on whether new dollars are required after current investments are redirected.

New state investment in return for more comprehensive reform. These systemic reforms will require a substantial infusion of new funds—higher compensation for top-performing educators and very likely increased spending on professional development—to be sustained over time. But initial outlays would be limited if state legislatures committed the funds only in a few select districts that volunteered to embrace these reforms. If they yielded the strong positive results we believe they are capable of producing, voters could decide then whether they want to invest in statewide implementation.


One of the key lessons learned in Denver’s experience with its ProComp pay plan is that because proposals to reform compensation command everyone’s attention, their implementation is a propitious moment to undertake systemwide changes. If schools are to succeed in raising student achievement, pay-for-performance should be accompanied by critical changes in other rules and incentives that bring new support to all educators.

Vol. 26, Issue 32, Pages 35,48

Related Stories
Web Resources
You must be logged in to leave a comment. Login | Register
Ground Rules for Posting
We encourage lively debate, but please be respectful of others. Profanity and personal attacks are prohibited. By commenting, you are agreeing to abide by our user agreement.
All comments are public.

Back to Top Back to Top

Most Popular Stories