Just rewards for teachers


Two decades of empirical research in education have confirmed at least one fact that just about everyone already knew: There are good teachers and bad teachers. The difference between your child being assigned to Mrs. Smith’s class or to Ms. Johnson’s down the hall can be as much as a grade level’s worth of learning by the end of the school year.

The wide variation in teacher quality suggests that some teachers deserve higher salaries than others, and indeed today’s public school systems have a tiered system of rigid salary ladders in which teachers are given extra compensation for factors commonly thought to be related to effectiveness. Pay differences are based primarily on, first, years of classroom experience and second, additional academic work toward an advanced degree.

There’s nothing inherently wrong with a compensation system that rewards experience and credentials. Business professionals pursue MBAs to garner higher salaries, and actuaries get salary bumps as they move toward becoming fellows of the Society of Actuaries.


But a compensation system for teachers based on additional academic credit and experience makes sense only if those factors are actually related to classroom effectiveness. They aren’t.

In a new study soon to be published in the peer-reviewed journal Economics of Education Review, my coauthors and I sought to measure the relationship between student achievement and those factors typically used to determine teacher compensation. Using data from all test-taking students in Florida over a period of five years, we found no discernible relationship between a teacher’s experience and credentials and the academic gains made by their students during the school year.

Our results confirm the findings of a wide body of existing research: A master’s degree is unrelated to a teacher’s effectiveness in the classroom. Further, our research dovetailed with an expansive body of research showing that while teachers get better at their jobs during their first several years in the classroom, these improvements level off after three to five years.

That’s not to say that no teacher has gotten better because of additional experience or studying for a master’s degree. But overwhelming evidence shows that such attributes are not generally related to teaching quality.

In fact, just about nothing that can be found on a resume tells us very much about how well a teacher will perform in the classroom. According to research by economist Dan Goldhaber, easily observed characteristics such as experience, education level and certification status only explain about 3% of a teacher’s measurable influence on student achievement.

It’s easy to see how the system developed to compensate teachers for credentials and experience. Those things are tangible achievements, and it wasn’t illogical to suppose that more experienced and better-credentialed teachers would be more effective. But modern research findings have made that supposition indefensible.


The vast majority of what makes one teacher better than another comes from attributes that are not easily quantified. That’s not so shocking, actually. All teachers need the skills that they acquire with experience and effective training. But great teachers also have innate characteristics such as patience, kindness, indefatigable dedication and the knack for getting reluctant students excited about learning.

If the goal is to reward great teachers so they stay in the classroom, we won’t find out who they are by looking at their college transcripts. We would do far better to identify effective teachers by evaluating their actual performance in the classroom. The ubiquity of standardized testing in public schools, coupled with modern statistical techniques, has provided us the necessary tools to do this.

Currently, public schools make no meaningful effort to identify effective teachers. Even in the worst-performing public school systems, it is common for 98% or more of teachers to receive a “satisfactory” or higher designation on their evaluations. Everyone understands that such results are highly inflated.

Over the last several years, researchers have been working hard to develop ways to identify the effect that individual teachers have on their student’s test scores. Such “value-added” measures of teacher quality are far from perfect and thus should not be used in isolation to make employment decisions. But they are much better indicators of a teacher’s effectiveness than are attributes such as credentials and experience. Public schools should utilize such quantitative measures of teacher quality along with qualitative observations of their performance to identify their most effective teachers and compensate them accordingly.

Effective teachers deserve to be rewarded for their achievements. Targeting higher salaries to the best teachers, rather than to the most experienced and best credentialed, would also help schools to retain those teachers who make the biggest difference for kids, while sending a signal to the least effective teachers that they might want to think about other careers.

Rewarding teachers for attributes that are unrelated to how well they perform in the classroom makes no sense. We need instead to focus on identifying the system’s most effective (and least effective) teachers and using that information to decide how much we should pay them.


Marcus A. Winters is a senior fellow at the Manhattan Institute and an assistant professor at the University of Colorado at Colorado Springs. His book, “Teachers Matter,” comes out early next year.