The Mighty Middle

Improving teaching through continuous improvement means concentrating on the majority of teachers, not just the ‘bad apples’ and high achievers.

April, 2015 Kirsten Kainz, Lora Cohen-Vogel, Christopher Harrison

In 1989, Donald Berwick published a seminal article that sparked an improvement revolution in healthcare. In his article, “Continuous Quality Improvement: An Ideal for Healthcare,” Berwick contrasted two very different reform strategies.

The first reform strategy, widely used in healthcare at the time, focused on inspection as a means to quality improvement. Managers were responsible for inspecting quality in the workplace. Results of inspections were made public for the purpose of coercing improvement. Reform leaders dedicated themselves to developing better and more precise instruments for inspecting performance quality. The ultimate goal of inspection was to identify and eliminate workers in the bottom tail of the quality distribution, otherwise known as bad apples. According to Berwick, bad apple strategies aimed at identifying and eliminating the bottom tail of the quality distribution were common in business and industry until overwhelming evidence suggested they were not effective at improving quality. Bad apple strategies were limited because they focused on the low-performing minority of organizations.

More effective methods, called continuous quality improvement, aimed to foster improvement throughout the organization by engaging managers and the majority of workers in frequent and collaborative examination of work processes and outcomes. Continuous improvement methods, that truly did improve quality in business and industry, were a natural fit for healthcare as well.

In contrast to bad apple strategies, the continuous quality improvement approach recognized that meaningful and lasting improvement would emerge when the majority of workers became better at producing desired outcomes. Therefore, continuous quality improvement managers established collaborative learning opportunities for managers and workers so that high leverage aspects of the work could be chosen for significant improvement over time. Managers and workers in partnership predicted aspects of the work that could be changed to result in improvement, and then examined whether enacted changes resulted in desired outcomes. The continuous quality improvement approach fostered learning communities that used evidence of practice and change to drive improvement throughout the organization.

Since launching the Institute for Healthcare Improvement in 1991, Berwick and his colleagues have created a nationwide healthcare improvement agenda with a track record of success. The amount of high quality research evidence indicating effective practices for hospitals has increased. Thousands of hospitals have formed continuous quality improvement teams comprised of doctors, nurses, and healthcare administrators who implement and study the impact of effective practices. And, most important, mortality and infection rates in hospitals have decreased substantially due to continuous quality improvement efforts based on evidence of effective practice. Imagine all that could be accomplished if we were to move the continuous quality improvement approach into America’s public schools.

The Education Context

Like healthcare reform efforts of the past, education reform movements intended to improve teaching quality often focus on the extreme minorities of the performance distribution. On one end, they focus on identifying and eliminating low-performing teachers. On the other end, they focus on rewarding high-performing teachers by paying for performance.

Rarely does a reform movement set its sights explicitly on the center of the distribution. This is unfortunate because the distributional center of teaching quality reflects the vast majority of teachers in the nation, what we call the mighty middle. By directing our attention to the mighty middle, we act upon an understanding that even small shifts in teaching quality across the majority of teachers will have greater impacts on the majority of students than elimination of the lowest performing teachers possibly can.

Let’s assume that the teaching quality in School A is distributed such that most of the teachers are located somewhere around average quality, a small number are far above average, and a small number are far below average. If we identify those far below average, eliminate them, and replace them with average teachers, then the resulting numeric mean for teaching quality in School A will be higher, but in reality not a single teacher has increased his or her teaching quality. School A is not improving by eliminating low-performing teachers, but the mean of teacher quality in School A is shifting, somewhat meaninglessly, to the right of the distribution. From this example we conclude that, although identifying low-performing teachers is an essential step in the process of ensuring that all students have adequate educational opportunities, removing low-performing teachers from the teaching pool is not a true improvement strategy. Maintaining simultaneous focus on improving teaching quality at all levels of the quality distribution is the surest way to improve learning for all students in a school district.

Continuous Teaching Quality Improvement

Efforts to continuously improve teaching quality will not only affect the greatest number of students, but such efforts also hold promise for redirecting teacher evaluation away from “identify and punish” tactics toward collaborative studies of improvement grounded in evidence of student learning, thus revitalizing schools as effective learning organizations. Recommendations for creating coherent teaching and learning systems where teaching practice is studied through multiple lenses can be achieved through a continuous quality improvement approach. And, new research indicates that such practice-based teaching evaluation systems can produce improvement in student learning district wide. As new teacher evaluation systems take hold across the U.S., we have an opportunity to adopt what has been learned about improvement from the field of healthcare and use this knowledge to transform the quality of teaching and learning in our schools.

Berwick launched a revolution with seven recommendations to make continuous improvement an ideal for healthcare. We adapt his recommendations here for the context of improving teaching and teaching evaluation.

  1. Educators and education leaders must in collaboration craft a shared vision for quality teaching in their school districts beginning with the statement that “Our schools in general are good today, and we intend to make them better.”
  2. Significant investments must be made toward improving teaching quality, including careful and sufficient study of the actions that result in improved teaching and learning.
  3. Respect for teachers and the act of teaching must be affirmed by declaring that teachers are working hard and doing their best within the systems we have designed.
  4. Efforts must be made to support positive and productive dialogue among teachers, principals, central office leaders, school board members, parents, and students, recognizing that it is through relationships among these parties that improvement in teaching and learning will be achieved.
  5. The continuous quality improvement methods that have proven successful in healthcare should be put into practice throughout schools, so that we can capitalize on the learning of those who have come before us and begin to produce good evidence about the practices that lead to improvement in teaching and learning.
  6. We must be committed to organizing schools and instruction according to the evidence we produce about practices that lead to improved teaching and learning.
  7. We must leave behind the notion that inspection and coercion will help us to improve teaching quality, and instead through our teaching evaluation systems create a rich and sufficient evidence base to improve teaching and learning throughout the nation.

About the Authors

Kirsten Kainz ( and Lora Cohen-Vogel ( are professors and Christopher Harrison ( is a research associate at the University of North Carolina at Chapel Hill.

Go to top