Editors’ Note: This is the seventh blog post in our series featuring reflections on and learnings from the 2016 Carnegie Foundation Summit on Improvement in Education. Each post provides recaps of sessions you may have missed and further insight into presentations you may want to revisit.

The use of improvement science in education involves careful study of problems of practice as well as the systems, processes, and activities that give rise to those problems. It seeks to refine those elements of processes prone to problems in order to produce more efficient, effective, and reliable outcomes. Interest in applying Carnegie’s six core principles of improvement — not only to obtain better outcomes but to rapidly and consistently improve over time — is spreading.

In a Summit session called “Leading the Transformation of Large Complex Systems,” Andrés Alonso (Superintendent Emeritus of the Baltimore City Schools), Eric Gordon (Chief Executive Officer for the Cleveland Metropolitan School District), and Christopher Steinhauser (Superintendent, Long Beach Unified School District School) provided a glimpse into how they profoundly shaped improvement in their school districts by adopting strategies that resonate with three of the six core principles of improvement. The three principles are: 1) Seeing the system that produces the current outcomes, 2) Using disciplined inquiry to drive improvement, and 3) Accelerating improvements through networked communities.

In 2004, prior to Alonso’s term as superintendent of Baltimore City Schools, the student suspension rate was at its highest — reaching 26,000 suspensions in a district of 88,000 students. Alonso’s administration noticed that the main problem was a set of procedures that allowed teachers to suspend students for any reason — even minor infractions. After clearly identifying the problem in the system, the district office revamped the code of conduct, giving Baltimore principals a range of alternatives to suspension, including parent conferences, mediation, referral to a student-support team, development of behavioral intervention plans, and the like. One of these alternative initiatives — Success Academy — helped the most troubled students stay in school and stay motivated by engaging them in a full day of instruction and counseling and by providing a place where they would be less prone to get into trouble. By 2010, suspension rates had fallen below 10,000 annually, affirming that the shift to the alternatives offered an effective approach to this problem. The Baltimore City School’s approach to alleviate suspension rates helps us understand how major change in school performance can be accomplished.

Gordon and Alonso explained how they engaged in improvement by decentralizing their administrative offices and shifting both accountability and more autonomy to individual schools. In embarking on decentralization, Gordon facilitated the practice of improvement through disciplined inquiry by giving principals and teachers the ability to determine the scope of school hours, programs, and curricula. The initiative empowered schools to run tests of change, learn from their failures, and continuously improve upon them. This, in part, gave each school the opportunity to address their individual institutional setbacks as opposed to the district office imposing a “one policy fits all” approach.

On a similar note, Alonso expressed his concern with any “one policy fits all” approach, because if schools do not have the resources and the autonomy to respond to problems (as they are defined locally), then widespread and effective change might not be forthcoming. In other words, it is not possible to achieve effective system-wide change by imposing any singular solution upon all. Both Gordon and Alonso went beyond imposing some general set of reforms that is supposed to work for all schools, instead considering what actually works for schools: for which local schools does a particular change work, and under what set of conditions? This in turn becomes a learning opportunity for the districts as they can study how certain schools were able to make powerful ideas work in their settings in ways that can be mirrored by others, thus maximizing greater improvement. Alonso and Gordon set a path for schools not only to have greater autonomy to use iterative testing to shape local implementation of common practices, but also to learn from each other on how to improve performance.

While there are often serendipitous opportunities to learn about improving from one another, Steinhauser takes this idea a step further by initiating and supporting formal networked communities. He does so along with leaders from the Long Beach City College and California State University, Long Beach. Together, they are committed to building a network called the Long Beach College Promise. The mission of this College Promise is to help Long Beach Unified School District students prepare for, enter, and succeed in college while offering guaranteed college admission at California State University, Long Beach for those who have completed minimum college preparatory or community college requirements. The College Promise partnership has been influential in having forty-nine percent of LBUSD students meet University admission requirements, which is an increase of seven percent over the past four years. This successful network prioritizes a common problem to solve together while recognizing that each participant in this partnership holds unique expertise and therefore is able to make valuable contributions to the collective effort. Each member provides a different perspective and resources to solve problems that might be identified in each step of a student’s educational journey. The College Promise is also data-driven; student outcomes are expressed as targeted goals, which allows for ongoing assessment of existing and new programs and services. This encourages the network to identify needed changes in order to continuously improve.

In a similar vein, Alonso stresses the need for collaboration amongst professionals who share in a common local problem, thus fostering community building. He emphasizes that by partnering with the community, charter schools, and government agencies, we can reach more effective improvement. While a problem may take considerable time for an individual or a single institution to solve, networked communities that pursue a common aim and follow a set of norms and beliefs about improvement can accomplish more together at an accelerated pace. While Baltimore City schools implemented new procedures for and alternatives to suspension, the impact of these alternatives depended on how effectively the district, government agencies, and the community worked together as a deliberate network. Both Alonso and Steinhauser have identified the need to develop these network partnerships as a means to help produce long-term continuous improvement.

These three district leaders have championed seeing the system that produces the current outcomes, using disciplined inquiry to drive improvement, and accelerating learning and improvement through networks (such as the College Promise). The leaders have helped us understand how school districts can get better at improving. In embracing the core principles of improvement science, the districts can realize greater accomplishment and secure more effective schools reliably and at scale.