The controversy over tying teacher pay to performance, rather than seniority and credentials, has escalated in the past year. Teachers unions, which have mostly come out against such “performance” or “merit” pay, have been bolstering their arguments with research suggesting that it isn’t linked to increased student achievement. But education reformers have, for the most part, stood firm in their support of teacher bonuses based on student test-scores, noting shortcomings in the design of the many programs studied. (Some programs have awarded bonuses based on school-wide performance, while others have paid out relatively small amounts to individual teachers.)
From 2007 until last year, the New York City Department of Education and the United Federation of Teachers offered bonuses in a random sample of the city’s “high-needs” public schools. The RAND Corporation was hired to study the program’s results. According to RAND, the extra pay “did not, by itself, improve student achievement, perhaps in part because conditions needed to motivate staff were not achieved (e.g., understanding, buy-in for the bonus criteria) and because of the high level of accountability pressure all the schools already faced.”
Proponents of merit pay must be feeling rather ambivalent these days. On the one hand, the federal government—led by Democrat Barack Obama—has spent considerable political capital supporting performance pay for teachers. Legislation in Florida now mandates that teacher pay statewide be based in part on teacher-effectiveness ratings. On the other hand, merit pay has been found wanting in two recent high-profile studies. How can we make sense of this?
Merit-pay programs are intended to positively impact the teaching corps via two distinct avenues: one involves motivation for current teachers, and the second involves a potential improvement in the composition of the teacher workforce. First, proponents contend that the possibility of earning financial rewards will motivate current teachers to focus their efforts on student achievement through innovation and additional effort.
The results of the NCPI Nashville study (released last year) and of the New York City RAND study (released this week) certainly call into question the motivation theory. I would caution, however, due to a variety of weaknesses in the designs of these plans, that it is still too early to close the book on the possibility that a well-designed merit pay scheme could result in enhanced teacher motivation in the short term. Two studies do not constitute an evidence base.
More importantly, however, these sobering findings from two short-term programs provide no insight into the potential benefits that the broader adoption of merit pay might foster for the overall composition of the teaching force. In our current non-merit-pay world, newly entering teachers can expect very high levels of job security, but very few rewards for high performance. This recipe may not be appealing to talented young people confident that they would flourish in a more differentiated system. The data agree: unfortunately, our most talented college graduates do not aspire to be teachers (Teach For America is one notable exception). However, if teacher salaries were related to effectiveness, talented and self-assured individuals might be more likely to enter the profession and turn into excellent classroom teachers.
Indeed, the widespread use of merit pay has the potential to enhance the composition of the teaching corps at the front end and beyond. Over time, a well-designed merit pay system would send the right signals and foster a sort of “natural selection” whereby effective teachers, encouraged by annual recognition and rewards, would eagerly return to the classroom each year. At the same time, their less-effective peers would find teaching to be less financially rewarding and would thus work to improve their skills or seek out other career options.
Perhaps the greatest hope of merit-pay advocates is that the introduction of a meaningful merit-based component to teacher compensation might foster a systematic change by which more talented individuals are attracted to teaching, excellent teachers remain in the classroom, and those who are not suited to the profession seek a new career. On this point, we have only logic and intuition—but no studies—to guide us.
Gary Ritter is a professor in the College of Education and Health Professions at the University of Arkansas.
We live in an era when educational policy is far too often shaped by ideological dogma. Our challenge is to engage in educational experimentation and innovation, and yet remain grounded in what research tells us works in real classrooms and real schools.
There is a well-established, substantial body of educational research that has found individual merit pay for teachers fails to produce meaningful gains in student achievement. What is more, individual merit pay has negative consequences, as the culture of trust and collaboration that is at the heart of a good school is undermined when educators are set in invidious competition with one another. In recognition of this reality, the United Federation of Teachers has consistently opposed individual merit pay for New York City public-school educators.
Until the UFT and the New York City Department of Education entered into an agreement to do a pilot program, there was no research as to the efficacy of school-wide bonuses as a tool of educational improvement. Since a school-wide bonus would not have the negative effects of setting educator against educator, and could conceivably contribute to collaboration within the school, the UFT decided that a pilot program was an experiment worth having, provided that it was subject to a rigorous evaluation by independent researchers.
With the publication of RAND’s “A Big Apple for Educators,” the results of that evaluation are now in: the school-wide bonuses have not produced meaningful gains in student achievement. While one might object that the standardized New York State exams used to evaluate the bonuses were a poor and unreliable measure of student achievement, the report’s other findings—most importantly, that the bonuses were seen as a weak motivation that did not change educator behavior and practice—leave little reason to think that a more robust measure of student achievement would produce substantially different results. Indeed, one of the significant findings of the RAND study was that the heavy reliance of the program on benchmarks drawn from the standardized state exams was a factor in diminishing its legitimacy with teachers. The evidence tells us that it is time for the UFT and the NYC DoE to return to the negotiating table to find new tools for improving student achievement, such as the development of a rich and powerful curriculum.
If one lesson is to be taken from this study and from the literature on individual merit pay, it is that teachers do not answer to the economic calculus of stockbrokers and hedge-fund managers. This observation may not sit well with those for whom the rule of the market and individual financial incentives are an ideological first principle, established prior to logical argument and evidence, but it is the reality of our lives and our schools, and it is affirmed again and again by educational research on performance incentives. While we believe that our challenging and exhausting professional work should provide us with middle-class lifestyles, our primary motivation in entering the field of education is not economic gain, but to make a difference in the lives of the young people we teach. Educational policy must recognize this motivation to produce lasting, constructive change.
Leo Casey is vice president for academic high schools at the United Federation of Teachers.