This article comes with resources. The HTH GSE Center for Research on Equity and Innovation developed these interactive self-reflection guides to better understand their own applications of the practices described in this article. Download these and try them with your team today!
Not everything that is faced can be changed, but nothing can be changed until it is faced.
In the midst of a national reckoning with centuries of racial injustice, where Covid-19 has laid bare and exacerbated systemic inequities in health, wealth, and education, schools and districts across our country are facing into their systems and asking themselves a fundamental question: How do we get better at getting better?
Many educational leaders are embracing continuous improvement as a methodology for tackling our most intractable problems (Bryk, 2020). At the core of continuous improvement are three simple, yet profound questions: What are we trying to accomplish? How will we know if a change is an improvement? What changes might we introduce, and why? (Langley et al., 2009). At its best, continuous improvement supports educators to better understand where and how our systems are failing Black and Brown children, set ambitious goals that press us toward more equitable outcomes, redesign our systems to achieve these goals, and learn our way into more just practices grounded in evidence. By pursuing improvement in the context of a network, schools are able to learn from variation and from each other, accelerating our collective learning and spreading good ideas across diverse systems. Yet, just because we are engaged in continuous improvement does not mean we will improve, or achieve greater equity.
Indeed, if you talk to leaders of this work, the same questions often emerge: How do we help people rally behind a common problem and shared purpose? How do we create networks where people are learning from each other, so we are better than the sum of our parts? How do we support people to see their systems from multiple perspectives—as they are and as they could be—so we can work together toward liberation? How do we design for distributed leadership and sustained change, not just temporary improvement?
Over the past five years, as we have participated in multiple improvement efforts and led many of our own, we have grappled with these same questions. We have facilitated—or served as the “hub”—for improvement networks focused on abolishing the phrase “I’m not a math person,” improving literacy through culturally responsive pedagogy, and increasing college access for students who are BIPOC or from low-income backgrounds. We have supported improvement efforts in individual schools ranging from reducing chronic absenteeism to improving equitable group work. Along the way, we have learned some important lessons (often the hard way) about how to support equity-focused improvement in schools. We offer them below in the hope that they will support others to get going faster, and avoid some common pitfalls.
People bring different definitions of equity to this work (Becerra & Weissglass, 2004). For some, equity is about access and outcomes; if we are working to address inequitable access or outcomes, we are engaged in equity work. Others argue that real, systemic change will only occur if we meaningfully address people’s beliefs, bias, and values and attend to the interpersonal dynamics within teams and organizations. And still others observe that since the institutions and systems within which we work were themselves designed to bolster white supremacy, we must dismantle and redesign them through policy and pedagogy (Kendi, 2019; Hammond, 2015; Love, 2019, Muhammad, 2020).
Like so many things in education (and life) it’s not an either/or. We’ve learned from Lindsay Hill, Victor Cary and many other amazing educators of color that we need to be weaversattending to outcomes, identity, relationships and systems–and be explicit about why each aspect matters. We need to support people to do the inside work and interrogate their systems. We need to provide time and scaffolds for people to engage across difference in reflective, honest dialogue about inequities, and work together to design more liberatory systems. We need to face into our disaggregated data, stop setting goals for “all students,” and get clear about the students—the groups and the individuals—we need to serve better. We need to relentlessly ask “What is working for whom, under what conditions?” (Bryk et. al., 2011) to ensure our learning and next steps are focused on those students and families. We need to find ways in our networks to decentralize whiteness and elevate the voices and experiences of people of color. And white friends, we need to help each other manage our white fragility (DiAngelo, 2018) and avoid white savior complex, so that we can respond with curiosity, compassion, and courage, rather than defensiveness or distancing, and be true co-conspirators in the work (Love, 2019).
Several of the improvement networks we have participated in (or led) have been a seemingly random assortment of schools, rather than a strategic slice of a system. And few have effectively engaged system leaders. As a result, pockets of improvement emerge, but it’s hard for the work to gain traction and make a large impact. We’ve learned the importance of being strategic about spread from the beginning—having a clear aim and participation expectations, recruiting diverse teams with these in mind, engaging vertical wedges (teachers, school and district leaders, county offices, higher ed, nonprofits from a particular district/region) where possible—so we can learn and expand across a system, tap into existing networks, and work with leaders and policymakers to remove obstacles. We believe this intentionality helps ensure the work doesn’t stop where it starts, and heightens our collective impact. It’s also really hard to do. Teams change. Leaders leave. It can be tough to get system leaders and policy makers in the room, even when the work aligns with their priorities.
We’ve learned to think differently about the ways we engage and communicate with leaders, and design for their unique needs. For example, we’ve found it more effective to invite systems leaders to shared learning experiences (i.e. visiting Cincinnati Children’s Hospital, Strive Partnership, or Georgia State so we can learn together from organizations using continuous improvement and collective impact for systems change) rather than expecting them to join school teams during network convenings. During convenings, we design special sessions or role-a-like groupings where school leaders can share dilemmas and learn from each other. We’ve also found it essential to cultivate multiple champions for the work so that it is sustained if a key person departs. This has required us to understand each school/district/system well enough to make sure the right people are in the room, on the team, and in the know. And it has continuously reminded us that while some of this work may feel technical and strategic, it is at its heart deeply relational.
We’ve participated in networks that functioned more like spokes in a wheel, where each team was strongly connected to the hub, but not to each other. This is a natural place to start; teams need to trust the network leaders to get going. But every time that we were a “spoke” we found ourselves hungry to learn from the other schools and frustrated by the pace of our collective learning and progress. When we started launching our own networks, we wanted to create less hierarchical and more liberatory relationships across the network. We worked to consciously build relationships between the various teams/participants so that they saw each other as resources in their learning. In our networks, we pair schools together in “buddy school” partnerships during convenings; we coach them in pairs on biweekly calls and they provide feedback to each other on their ongoing work. We work toward shared leadership by spotlighting the expertise of diverse network members and engaging them to co-design and facilitate portions of our convenings. By building trust among teams and developing their capacity to lead for improvement, we accelerate our collective learning and ensure that schools/districts are not reliant on the hub alone to continue the work.
We learned this one the hard way. We launched our first college access network with an aim (or goal) that was deliberately broad and unfortunately imprecise. Then we supported each team to pursue their own unique aim, many of which were only tangentially related to our broader aim. We soon discovered that this breadth is problematic if you want to function as a network and actually improve something for a specific group of students. For example, it was difficult to establish a common set of measures to assess our overall progress and impact, and few schools had the data structure in place to track disaggregated data relevant to their aim. Furthermore, there was no authentic reason for teams to share their work or learning.
In contrast, when we launched our second college access network three years later, we had a clear aim: increasing the percent of students who are African American, Latinx, Indigenous or from low-income backgrounds who enroll in colleges from which they are likely to graduate. The network would focus on four critical drivers of college matriculation: sense of belonging, financial access, the college application and enrollment process, and reducing summer melt (the number of students who are admitted, but fail to enroll in college after graduation from high school). The first year, we all focused on increasing FAFSA/Dream Act completion and Cal Grant awardance (which provides up to $50,000 in tuition relief for qualified students). With teams working toward shared, focused, and measurable aims—and using common data to assess their progress—there were authentic opportunities for sharing and adapting successful practices across teams. Most important, there was evidence that our collective efforts were leading to improved financial access for the students who needed it most. In the last two years, we improved FAFSA and Dream Act Completion by 11 percentage points. In our first year alone, 103 more students received a Cal Grant, which meant up to $5 million more dollars in aid that our students didn’t have to pay back. We have learned that if you choose a compelling and clearly defined problem of practice from the beginning, you will attract the teams/people who want to take on that challenge, and you will get further faster.
We love design thinking. We appreciate the focus on empathizing with users, and the energy and enthusiasm generated through the design thinking process. Having said that, early on we realized that people didn’t always want to dream it up themselves, or have time to cull through the research for promising practices. As network leaders, we took responsibility for identifying high-leverage practices that were grounded in research, addressed key root causes, and had already been adapted successfully across diverse contexts. We launched our math network and our current college access network with a concrete aim, key drivers, and a preliminary change package (i.e. a set of research/evidence-based practices for the network to iterate on). We then supported teams to engage in their own root cause analysis and better understand their systems so that they could adapt these practices for their own contexts. By providing a clear direction and a place to start, we could get moving quickly and accelerate the learning.
To be clear, we are not advocating for launching a network with a bunch of top-down directives that rob educators of their agency and creativity. And this didn’t mean that novel ideas were off the table. It just meant that every team was also testing, adapting, and refining some common high-leverage practices. This facilitated authentic opportunities for sharing and learning from variation. It also helped to generate early wins, which are empowering and crucial to building a high-functioning network. Publicly celebrating these wins helped us spread effective practices/adaptations and maintain momentum when the work was tough.
We love empathy interviews, fishbones, interrelationship digraphs, process maps, and driver diagrams. However, if we are focused on the tools without grounding them in our collective purpose, they rarely generate the moments of insight we are after. Equitable access to college, abolishing the phrase “I’m not a math person,” ensuring kids learn to read and love to read, building belonging and reducing the disproportionality of suspensions. These are the issues educators care about. It’s what brought them to the work. Every member of our networks can identify particular students who they hope their work impacts. Ideally, when we are digging into root causes and generating change ideas we will have students in the room with us. If not, we have people pair-share about them, bring pictures of them to put in the center of their tables, interview them to surface their felt needs, analyze samples of their work to surface moments of brilliance and confusion—anything to keep students at the center. When we introduce an improvement tool, we do so in service of doing better by those students. As the hub, we continually return participants to the “why” behind the work, while also revisiting our “what” (i.e. our aim and theory of action) and equipping them with the “how” to achieve it (i.e. tools, protocols, change packages). And we’ve found that there is no better way to get grounded in the “why” than having people conduct empathy interviews or shadow the students on their minds.
As noted above, there are lots of tools people can use in continuous improvement work. The tools are important, but not enough on their own. For this work to be liberatory and transformative, we need to attend to relational dynamics and help people construct their thinking together. We also need to consciously return to the ways in which our systems are inequitable by design, and hone our abilities (and willingness) to ask tough questions of ourselves and others, seek diverse perspectives, and engage in crucial conversations about race and systemic oppression. This requires more than good intentions. We’ve found that protocols designed for equity are essential for helping teams engage with the tools of continuous improvement, while also reflecting on their own assumptions and engaging with each other in more inclusive, purposeful ways. Without the intentionality that structured protocols and norms provide, the tools of continuous improvement can feel like a hollow exercise and fall flat. Or worse, they can reproduce the inequities we aim to address.
Here’s an example: picture a cause-and-effect diagram (or fishbone diagram) unpacking the root causes contributing to a problem that focuses primarily on the ways students and families are perceived to be deficient. Now, by contrast, picture a fishbone diagram that highlights the ways in which our own beliefs, practices, and processes as educators are contributing to the problem (many of which are actionable and within our locus of control).
Here’s another example: imagine a team mapping their school’s process for supporting SAT registration where the two white men in the group do most of the talking, versus a team where everyone’s perspectives are included and authentically explored (and perhaps where the people of color and those who were/are first generation college students themselves are viewed as experts with the most to contribute). Not only would this team’s understanding of the process be richer and more nuanced, they would have generated some shared understanding and ownership to make change. In this way, protocols designed for equity help build teams’ conversational capacity and improvement capacity, so the work can be sustained by the team, not borne by individuals alone.
A key principle of improvement is to be problem-focused. However, if we are always focused on the problems, people can become paralyzed. Seeing the problem more clearly doesn’t necessarily lead to action. Or as my colleague Dr. Michelle Pledger likes to say, “You can be woke, but still laying in bed.” In Switch (2010), Chip & Dan Heath suggest asking, “‘What’s working, and how can we do more of it?’ Sounds simple, doesn’t it? Yet, in the real world, this obvious question is almost never asked. Instead, the question we ask is more problem focused: ‘What’s broken, and how do we fix it?’” (2010, p. 45). And yet, in every system, there are people who are working differently, challenging the status quo, and achieving results that push back against the common excuses—often by tapping into the brilliance and indigenous knowledge of the students and families they serve.
As a hub, our goal is to be reality facers and systematic bright spotters, and to cultivate these dispositions across our network. We have developed routines for regularly assessing where teams are and getting into the weeds with them, so we are learning together what is working for whom under what conditions, and what is not. When we learn about a high-leverage practice or even a small detail that has made a big difference—like holding the first FAFSA parent night a month earlier, providing think time before a pair-share, or responding to students’ questions with a question instead of an answer—we call it out, give them a mic, and share it with the network. We want people to know that solving tough problems is about addressing root causes and about getting inspired and building on what is working (somewhere).
Don Berwick (1996), President Emeritus of the Institute for Healthcare Improvement, has argued that one of the main factors preventing systems change is that people operate as if improved performance was an effort issue rather than a design issue. There is a strong tendency to do what we’ve always done, only more of it. We’ve seen many schools muscle their way to improvement, but it’s hard to sustain the work over time and teams end up exhausted. A saner path (that may feel less comfortable initially) is to support teams in mapping and unpacking their current processes so they can identify potential breakdowns as well as under-utilized resources. They can then redesign their process to integrate possible solutions. This is tough. As Heifetz and Linsky (2002) have noted, change is loss. People are often reluctant to let go of the practices and processes they have in place, even if they are burdensome, ineffective, and inequitable.
The questions we return to, as a hub and with our teams, are: Are our aims ambitious enough that they require us to redesign our processes and work in new ways? How can we reengineer our systems to better serve those students our schools have traditionally failed? For example, let’s say we want to improve FAFSA/Cal Grant rates by 10% in one year for our Latinx students. Hosting more parent workshops is not going to get us there. We may also need to improve the parent workshops we offer, develop routines for looking at data to determine who has completed and who has not, and find ways to support those students who still need to complete. We may need to start earlier in the year, push into Senior classes, pull small groups during the school day, and leverage additional faculty and their relationships to provide personalized support to students with especially complex situations. This would represent a shift not just in what the work looks like and how it gets done, but in our beliefs about students, families, and our role in supporting them. If we see financial aid completion as our responsibility, we are not just offering information and seeing who comes to us. We are reaching out to provide targeted support and to ensure completion.
For this work to be sustainable, however, we cannot improve through addition alone. Sutton and Rao, in their excellent book Scaling Up Excellence (2014), discuss the dangers of breaching people’s “cognitive load”—of adding so many new tasks, responsibilities and priorities that people lose the capacity to devote sufficient attention to what matters most and as a result, end up doing many things poorly rather than a few things well. When I was teaching in schools, I experienced this as “innovation overload.” The result was not only individual fatigue, but also system incoherence and confusion. To combat this, Sutton and Rao encourage organizations to play the “subtraction game” and to ask “What can we stop doing?” By letting go, we can create space and energy for leaning into new processes (and mindsets) that work better for those we serve, and ultimately for us too. Letting go can also take the form of broadening the team and engaging underutilized people in the system—such as students, parents, coaches, tutors, etc.—as collaborators. The above questions encourage us to think differently, to mind the gaps, and to be more focused and strategic in our next steps.
We used to think that the action happened at network convenings. We would put the bulk of our energy into designing memorable, thought-provoking experiences when we convened the network together. Then we would cross our fingers and hope that we had inspired each other enough that the work (learning from users, conducting inquiry cycles, looking at data to guide next steps, etc.) would continue after teams left us. In some cases it did. In many other cases, the work got overtaken by other priorities, and the myriad daily demands of educating young people in schools. Of course, as Paul Batalden and other brilliant thinkers have noted, every system is perfectly designed to get the results it gets. So we started to ask ourselves what real supports or scaffolds we were providing to ensure that teams were able to sustain the work and reflect on their progress between convenings. The answer: very little.
Now we design the arc of the year with the action periods in mind. When we sit down to plan convenings we first figure out what we want teams to do in the next action period and we backwards plan. This question, initially posed by my colleague Ryan Gallagher, has been particularly helpful: When teams get back to their sites, what do we hope they are doing differently next week? This has helped us better prioritize what needs to happen at convenings (i.e. What is going to set teams up best to engage in the action period work?) and articulate more clear work streams/tasks, as well as more clear roles and expectations for team members, during these action periods. It’s reminded us to build substantial team planning time into convenings, giving teams the “chew” and “do” time they need to execute well on their ideas and next steps. And it’s encouraged us to think strategically about our coaching calls and site visits, so that we are anticipating where teams may get stuck or lose momentum, and providing the needed support to keep the work moving. Which brings us to a bonus lesson…
We dismiss the power of story at our peril. When people share their stories—their histories, their identities, their purpose and passions—as well as the learnings and successes they are experiencing in the work, the network shifts from “theirs” to “ours.” Stories bring learning to life in vivid detail, and help good ideas spread across a network better than any change package, statistic, or powerpoint presentation. And it puts the participants at the center and on the stage, rather than the hub (see lesson #3). Sharing our personal stories builds a sense of deep belonging, where people feel seen and known. When we build networks where all participants feel they can bring their whole selves to the work, systemic and sustained improvement for equity is possible.
A couple years ago, my team had the good fortune of meeting Dr. Uma Kotagal from Cincinnati Children’s Hospital. She is an improvement force to be reckoned with and is fond of saying, with loving impatience, “You have to start before you’re ready. Because you’re never ready.”
We all have much to learn. And the only way to keep learning is to get going, before we have it all figured out. Our students are waiting.
1 The Social Change Ecosystem Map developed by Deepa Iyer with the Building Movement Project is a helpful way of envisioning the various roles people can play in an ecosystem focused on social change.
Return to article
Special thanks to Alicia Grunow and Sandra Park for pushing us on #8 & #9, to Becky Margiotta and Joe McCannon for inspiring this piece with their own synthesis of lessons learned, and to Ben Daley and the whole CREI team for learning and churning alongside me.
Becerra, A. & Weissglass, J. (2004). Take It Up: Leading for Educational Equity. The National Coalition for Equity in Education.
Berwick, D. (1996). A primer on leading the improvement of systems. The BMJ, 312, pp. 619-622.
Bryk, T. (2020). Improvement in action: Advancing quality in America’s schools. Harvard Education Press.
DiAngelo, R. (2018). White fragility: Why it’s so hard for white people to talk about racism. Beacon Press.
Hammond, Z. (2015). Culturally relevant teaching and the brain: Promoting authentic engagement and rigor for culturally and linguistically diverse students. Corwin.
Heath, C., & Heath, D. (2010). Switch: How to change things when change is hard. Random House Canada.
Heifetz, R. and Linsky, M. (2002). Leadership on the line: Staying alive through the dangers of leading. Harvard Business Review Press.
Kendi, Ibram X. (2019). How to be an antiracist. One World.
Langley, G.L., Moen, R., Nolan, K.M., Norman, C.L. & Provost, L.P. (2009). The improvement guide: A practical approach to enhancing organizational performance. Jossey-Bass Publishers.
Love, B. (2019). We want to do more than survive: Abolitionist teaching and the pursuit of educational freedom. Beacon Press.
Muhammad, Gholdy (2020). Cultivating Genius: An Equity Framework for Culturally and Historically Responsive Literacy. Scholastic Teaching Resources.
Sutton, R. & Rao, H. (2014). Scaling up excellence: Getting to more without settling for less. Crown Business.