Teachers in Ontario are currently paid based on experience and qualifications. Recent debate in the United States about the desirability of merit pay has led to similar discussions in Canada. In this article, merit pay is defined as a scheme that links some portion of teachers’ pay to their students’ academic achievement. An extensive review of research in education and other sectors supports my argument that this is not a desirable policy. Here are eight reasons why.
1. Few people in the labour force are paid on the basis of measured outcomes
Paying people based on the outcomes of their work is quite a rare practice. In many fields, earnings have much to do with reputation, but reputation is not the same as a measured outcome.
2. No other professional group is paid based on a measured outcome
Nurses, lawyers, engineers, architects, or even aircraft pilots are not paid based on outcomes, though their pay may be based on volume of work. Even where there is pay for performance, the performance measures are not primarily related to measures of client outcomes.
3. Most teachers oppose such schemes
Surveys consistently show that teachers strongly oppose merit pay schemes. (70 percent or more are against.) Since improvement in education depends critically on teachers’ commitment, anything reducing that commitment is likely to be unhelpful to achieving better outcomes.
4. Pay based on student achievement is highly likely to lead to displacement of other important education purposes and goals
When people have a financial incentive to achieve a score, that incentive may displace other, more desirable goals. Research by psychologists shows that extrinsic rewards can act to displace intrinsic motivation; thus, for some teachers merit pay schemes could reduce their desire to do the job well simply because that is their professional responsibility and wish. As well, the outcomes that are measured to determine pay are likely to get more attention than those that are not, such as achievement in subject areas not tested, interpersonal skills, or the ability to motivate students.
If merit pay is individual and competitive (as is often the case), teachers will have incentives not to cooperate with colleagues. They may even have reasons to be glad that colleagues are ineffective.
5. There is no consensus on what the measures of merit should be
Merit pay seeks to link teachers’ pay to student outcomes; however, deciding which outcomes will determine merit is highly problematic. In addition to academic achievement, we also value such things as students’ ongoing ability and desire to learn, ability to work with others, creativity, and citizenship skills. Factoring in these goals would make merit pay schemes too complicated.
6. The measurement of merit in teaching inevitably involves a degree of error
In merit pay schemes, teachers’ merit is being inferred from student performance, but any measure of student performance, whether a classroom assessment or a standardized test, will have some error in it. The total measurement error in any merit pay scheme is likely to be quite large, making the results unreliable.
7. The details of merit pay schemes vary widely, yet these details have great impact on how suchplans are received and their effects on teachers and schools