Merit Pay Plans: Effects on teachers and their work.

Hide Document

Merit Pay Plans:
Effects on Teachers and Their Work
Susan Moore Johnson
Harvard Graduate School of Education
March 22, 1984
This speech was prepared for The 1984 National Conference on Merit Pay for Teachers, sponsored by The American Center for Management Development, Sarasota, Florida, March 22 and 23, 1984
Merit pay for teachers is a school reform drawn from ( the pragmaticr no-nonsense world of business. It is at ·once a simple and compelling idea. If teachers were paid competitively for their performance,th~ argument goes, all would work harder. The best, having 'been rewarded with higher salaries or bonuses, would remain in teaching. Average or poor teachers would strive to improve or leave. Ultimately, it is argued, schools would be better off. Despite its logic, simplicity, and congr~ence with American values, merit· pay has not won the enthusiastic support of American teachers. In fact, The press reports teachers vehemently opposing merit pay proposals in Tennnessee, Florida, and many local communities. The public, politicians, and many educational commentators are puzzled, sometimes angered by this opposition. Are teachers lazy, uncourageous, or simply unmoved by opportunities for advancement? Why would they reject a reform that piomises to benefit both them and their schools? I believe that we can better understand this opposition and the repeated failures of merit pay plans throughout the century by reviewing the priorities of teachers, the structure of their work, and the organization oftheir schools. We have learned from business that if pay incentives are to be effective, they must be adapted to fit the work, the workers, and the workplace of any particular enterprise. As they have been typically conceived, educational merit pay plans fit neither teachers, teaching, nor schools. Today as I explore these issues,. I will distinguish between merit pay, which competitively rewards teachers who
1
( have the same responsibilities, and career ladder or master teacher plans, which distinguish among teachers having different responsibilities. I will speak most directly about merit pay largely because we know most about how it works --or fails to work. However, much of this analysis is applicable to any performance incentive plan. I will begin by considering the fit between merit pay, reforms and what we know about the educational enterprise. I'd like to do this by briefly reviewing some important things we have learned through research about teachers, teaching, and schools and suggesting how they might influence the design of merit pay plans. Having noted what I consider to be the potential organizational hazards of merit pay, I will conclude with some specific recommendations for those who are currently drafting incentive pay plans for teachers. Teachers Merit pay reforms are based on a set of well-intentioned hunches about what teachers want, how teachers work, and how schools function. They assume that teachers seek competitive pay and that those who leave the profession do so because they are disenchanted with th~ standard salary scale. As one writer in the American School Board Journal recently argued: Imagine being' an effective, hardworking teacher condemned to receive exactly the same raise as the listless, barely adequate dolt down the hall. To anyone reared on tales of Henry Ford and Horatio Alger, this wrongheadness is so glaring that merely ~eeing it officially sanctioned must be grounds for despair.
Reformers assume that dissatisfied, unchallenged teachers
2
(
are a significant part of today's school problems, and that such teachers would respond enthusiastically to opportunities for competitive pay. Moreover,' the reformers assume that merit pay would attract new, more entrepreneurial individuals to teaching.
There are several important things 'that we have learned about teachers that are relevant to this discussion of merit pay. The first ,substantiated by the research of both Dan Lortie and John Goodlad, is that the majority of teachers enter teaching to teach. That is not as self-evident as it first sounds. Teachers, do not choose teaching over other occupations because of the money or the summer vacations, or even the long, leisurely lunches, but because of the kind of inherently rewarding work it offers.
Moreover, Goodlad's extensive research about teachers tells us that when teachers quit, they do so because they are "frustrated 'in what they wanted to do or disappointed in their own performance."
The seco,nd important fact is that those who teach are not well paid, and these low wages dissatisfy them. In 1981-82, the average salary of teachers with a BA was $12,769. Engineers with a comparable degree earned $22,300. Computer scientists earned $20,300. In order to demonstrate their dissatisfaction with low wages, 100 teachers in Summerville, S.c. recently lined up at the local Post Office to apply for a custodian's job paying $7000 more than starting teachers make in South Carolina.
Goodlad found that while money was not a major reason for teachers' entering the profession, it ranked second as a reason
3
for leaving. He speculated about this shift:
(
Anticipaing rewards intrinsic to the work,
teachers begin with a willingness to forego
high salaries. However, when confronted
with the frustration of these expectations,
the fact that they sometimes are paid less than
the bus drivers who bring their students to school
may become a considerable source of
dissatisfaction as well.
Repeated sociological studies have shown that teachers are a conservative lot--politically, personally, and professionally. They are not financial entrepreneurs, but ratner seek financial security from predictable salaries and retirement settlements. Their dissatisfaction with pay centers on low annual wages rather than on the standard salary scale.
Third, although teachers are isolated by the egg-carton structure of most schools, they regard themselves as team
c players rather than free agents. As teacher negotiations often demonstrate, egalitarian norms are strong among teachers. In my own research, I have been struck by the intensity of teachers' concern for equity and fair treatment. They are wary of administrative power and the potential for patronage and favoritism in assignments, promotions, and merit bonuses.
Given what we know about teachers' preferences, priorities, and group norms, there is little to indicate that individualized merit pay would motivate them to work harder or better. Teachers want appreciation and recognition for their work, but not at other teachers' expense. They seek better wages, but do not want to purposefully pursue them by teaching to the test or feigning compliance with administrative preferences.
(
Because most merit pay plans in education are directed toward a small number of outstanding teachers, there is little reason to expect that they will raise the general level of teaching in schools. Many argue that under merit pay, outstanding teachers will continue to succeed as they always have, while the average or below average teachers, who regard merit pay as unattainable, may even reduce their efforts in discouragement.
Merit pay is ostensibly an incentive for improved performance. But as it is being proposed by many school critics and legislators, merit pay is a stick disguised as a carrot. It will punish those who fail more than it will motivate and reward those who succeed.
( TEACHING Next, I want to mention a few things we have learned about teaching--the labor of this schooling enterprise--that have implications for the merit pay plans.
First, despite 80 years of effort, educational researchers have failed to identify the characteristics of effective teachers or effective teaching. What is clear is that there are many different styles of good ,teaching, and that it is difficult to separate teaching expertise from the individual. Although welders and salesmen undoubtedly encounter uncertainty and exercise discretion in their work, they do know what success is--a good weld, a final sale--and they know when they have achieved it. This is often not true for teachers, or for the evaluators who assess their work.
Second, teaching is what my colleague, David Cohen, calls
( an "indeterminate practice." Because schools do not define their goals carefully and often pursue many objectives simultaneously, expectations for teachers are often vague, muddled, or conflicting. They are expected to teach reading, handwriting, exposition, social adjustment, inferential reasoning, and more. There are no generally accepted strategies for achieving those outcomes that are made explicit. Moreover, teachers must constantly adjust to the uneven and unpredictable development of students, who ea~h bring a unique set of strengths, weaknesses, and past experiences.· There are no true raw materials in teaching. Third, although teaching is carried out in isolated classrooms, it is an interdependent process. Teachers must rely on those who instruct other grades or subject levels to do their work well so that the final product of this educational assembly line is a well-educated student. Moreover, teachers must constantly cooperate with other staff and adjust their teaching practices to larger program or school goals. Teaching, then, is an underdefined craft. In most schools, teachers must contend with unspecified, but all encompassing goals as they seek to teach students with varied capabilities, skills, and patterns of growth. Teachers work alone, but their work is successiver no single individual finishes the job. Merit pay plans assumes that there is general agreement about what good teaching is and what it should accomplish. There is in fact no such consensus. Moreover, merit pay assumes that the individual teacher's accomplishments can be isolated and assesJed. Yet because of the successive and interactive nature
6
of schooling it seems unlikely that various teachers'
(
effectiveness can be compared with any precision. Teachers know
that in their work they must tolerate considerable uncertainty
and ambiguity about their success. If the product of schooling is
a well-educated student, individual teachers control only a
piece of that product. However, merit pay plans assess each year
of the teacher's work as if it were the entire production process
or as if the pieces of that process were simply additive.
Teachers can neither control the quality or preparedness of the
students they teach; nor can they accurately predict or regulate
the uneven developmental rates of student learning. A second•
grade teacher's success in laying the groundwork of mathematical
reasoning may not become apparent until her students reach the
fifth grade; the change that the teacher will then be credited
with the success is slim. Conversely, students who perform
satisfactorily in first-year French may begin to fall behind in
the second year because they never really learned the irregular
verbs in the introductory course. Test scores will never reveal
the source of the probl~m. For these reasons, teachers recognize
that their work is interdependent, though often isolated. As
individuals, they can influence but not control the outcome.
Schools
Schools, too, have important features that reformers must take into account as they design merit pay plans. Performance•based pay systems assume a hierarchical relationship between supervisor and supervisee. The principal, assumedly having superior expertise, can assess teachers' performance from the
7
(

distance of his or her formal position. However, schools have both a formal and informal structure. The formal structure, which is apparent on the organization chart, is hierarchical, bureaucratic, and rulebound. The principal is in charge; the teachers are subordinate. The informal structure is flat, reciprocal, and highly dependent on personal relationships. There is extensive overlap between teachers' and administrators' responsibi1ities,interests and concerns.
Although principals are technically the instr'll;ctional leaders of the school, many readily admit that their teachers are the experts at the craft of teaching. Many principals who have never taught at the level where they administer turn their attention to school management rather than classroom supervision. They leave the teaching to. the teachers. They are different from the industrial foreman or the unit head in business who con~inua1ly oversee and inspect their supervisees' work.
Although principals have formal authority for the school, its staff, and program, they must rely on the cooperation of teachers to make their schools run well. As one principal I interviewed ~xp1ained:
A principal can't be strictly management and run a good school. I need the teachers' help and advice. I can't make decisions in
. a vacuum. I have to maintain a certain level of trust if I expect the school to function well. (HER 1980,p. 224)
In their effort to' foster reciprocal relationships and collegiality with staff, principals o.ften assume positions somewhere between teachers and managers, facilitating effective instruction. They support and protect their teachers,
8
( they moderate conflict, and they minimize competition. They respect, or at least acknowledge, teachers' concern for equity and take steps to eliminate, or at least conceal, favoritism. They intuitively recognize that if they were to formalize their relationship with their staff, they would pay an organizational
. price. Relationships might well become strained; administration would certainly become more difficult; the school might run less well.
Effective schools--those schools where students consistently outperform predicted levels of success--are notable for their consistent focus and purp.ose. In such schools, administrators and staff acknowledge their interdependence in setting schoolwide goals. Teachers plan cooperatively, and work
( collegially. As Stewart Purkey and Marshall Smith observe, "An academically effective school is distinguished by its culture: a structure, process, and climate of values and norms that channel staff and students. in the direction of successful teaching and learning." Our limited understanding of effective schools suggests that serious efforts should be made by school reformers to promote reciprocity between teachers and administrators, and to increase teachers' interdependence and cooperation. Most merit pay plans, however, would have the opposite effect. They would likely isolate rather than integrate the individual teacher. If teachers were competing for scarce rewards, they would be less likely to share their ideas and materials with colleagues. If merit pay incentives encouraged teachers to work on their own
behalf rather than on behalf of the entire school, students of
(
exemplary teachers might profit at the expense of students in other classrooms. The quality of curriculum development, building supervision, and inservice training might all be diminished by merit pay plans that stress the accomplishments of the individual over the accomplishments of the group.
The Hazards of Merit Pay
From the perspective of teachers, teaching, and schools, merit pay plans, as they are typically designed, have little promise for motivating teachers, improving schools, and creating working conditions that would attract talented new staff. Moreover, there are a number of oiganizational hazards to beware of if you are innstituting merit pay for teachers, and I would
( like to summarize those now before moving on to some specific recommendations. 1 •. Merit pay may generate divisive competition among staff and ultimately divert teachers' attention from teaching.The line between healthy and unhealthy competition is a fine one that warrants careful watching.
2.
If merit pay bonuses are restricted to . only a few teachers, the ent·ire staff may reject the initiative as an empty symbolic gesture or ostracize the individuals who are selected for recognition.

3.
Merit pay may actually serve as a disincentive for average and below-average teachers who regard performance bonuses as unattainable .


10
4. Administrators who have unrealistic expecations about the
( efficacy of merit pay to improve teaching may neglect other
\
important means of motivating and improving staff.
5.
Merit pay may promote "gaming" among teachers by redirecting their attention from their instruction to their ratings. This has proven to be an ~nintended consequence of merit pay reforms in the civil service. Those teachers who are intent on achieving high merit ratings will focus their efforts on that challenge. In some cases, that will lead to better teaching; in many it will not.

6.
By promoting independence rather than interdependence, merit pay may increase teacher isolation i~ schools. Distrust about evaluations and suspicions about patronage may reduce cooperation among teachers as well as between teachers and principals. Such organizational disruption may well interfere with efforts to improve schools.


Recommendations for Merit Pay Design
Lest you think that my entire perspective is 'one of gloom and doom, I would like to conclude with some specific, hopefully constructive recommendations for those of you who will be instituting merit pay in your states or local districts.
1. Do thorough plarining and organizational preparation for any reform of teachers' pay and responsibilities. That means meeting and talking wi~h teachers, soliciting their ideas, and
. taking those ideas seriously. It means listening to principals' concerns and involving them in the design of what they must administer. It means negotiating an incentive system that all
11

· segments of the school community believe can work and are willing to
make work.
2.
Recognize that merit pay is not a cost-saving reform. It cannot substitute for good salaries and it should only be instituted as an extra incentive. Merit pay will not yield a big return on a small investment, particularly if teachers regard it as a way for school officials to avoid paying adequate wages.

3.
Design your plan so that its emphasis is positive, not punitive, so that the focus is on recognizing good teaching rather than eradicating bad teaching. If merit pay is really a stick rather than a carrot, it will not serve as an incentive for


improvement~
4. As counterintuitive as it may seem, consider deemphasizing the monetary component of merit pay, especially if
~
there is little funding available for the program. Experiemnt
with symbolic rewards. Or recognize outstanding teachers not with
bonuses in their pay envelopes, but with credit lines to be used
for ordering special curricular materials, buying consultants'
time, or attending conference. Honor teachers by granting them
discretion over instructional purchases and you may find that
you have a genuine performance incentive.
S.If you institute an individualized merit pay plan, spend the money and time necessary to have a thorough, creditable evaluation system. We have learned from industry that if employees believe that pay and performance are not tig~tly linked, merit pay will not work. Therefore, you should commit
12
(

~
/
r
. yourselves to defining clear performance criteria, training evaluators to achieve rater reliability, requiring multiple observations, and, if possible, multiple evaluators. Short cuts will, in the end, prove costly.
6.Couple any performance appraisal system with ongoing inservice training and individual remediation. It's possible that merit pay might motivate some teachers to work harder, but it won't improve the performance of mediocre teachers who don't know how to change and probably will not choose to leave.
7.
Use test scores cautiously. There will be great pressure for objective measures and increased accountability. Test scores are currently the only comparative measure of instructional outcomes readily available in most school districts. But test scores are problematic. They measure only part of what is taught They do not assess creative expression, scientific reasoning, or citizenship. Moreover, some students do not test well, while thers test far beyond their routine performance. In determining merit awards, use test scores only with reference to an entire school's progress, never in reference to the success of an individual teacher. When tests they are used for schoolwide comparisons, be sure to include qualitative measures of effectiveness as well.

8.
Structure merit pay plans so that there is frequent opportunity for success by many teachers. They need not be annual awards conveyed on a small number of people. Set districtwide or schoolwide goals and reward all who achieve them, not simply the select few who do the very best. In In Search of


13

Excellence, Peters and Waterman describe excellent companies as
( those that "turn the average Joe and the average Jane into
winners. They let, even insist that, people stick out. They
accentuate the positive." We need more winners and would-be
winners in schools.
9. Design your merit pay plans so that they promote teacher
interdependence rather th"an isolation. Reward teams of teachers,
departments, or schools rather than singling out individuals. If
there is peer pressure, let it be on behalf of the group.
10. Finally, recognize that incentive plans require ongoing
tending, realigning, and refurbishing. Ip schools, as in all
organizations, entropy reigns. Even the best-designed merit pay
system or career ladder will become stale and ineffective unless
/" \ it is flexible, regularly reviewed, and revived.
In closing, I would like to remind you that we live, as Jay
Featherstone says, in the United States of Amnesia. Merit pay
for teachers is not a new idea. It has been tried at many times,
in many places, and with few exceptions, it has failed. If
/• school reformers dismiss or disregard those failures, they will
likely repeat them. In the process, money will be lost, teachers
will be disaffected, and children will pay the price.
Many of you are under pressure to institute merit pay as a
condition of increased financial support by business and
government. If you decide to institute merit pay reforms, I urge
you to design plans that are consistent with who teachers are and
what matters to them, plans that acknowledge the complexity of
teaching and the difficulty of assessing it, plans that support
14

more coherent, cooperative school organizations, plans that, in the end, will have been worth your time and energy_
c.

Hide Document