CSBA outlines district scaling methods

School improvement sounds like a classroom problem. But this new California guidance says the real bottleneck is usually the district itself. Not teacher effort. Not one more training day. Basically, the claim is that instructional change lasts only when the district’s governance, central office, and routines are built to carry it. That was the point of a May 5 post from the California School Boards Association, which highlighted a February 2026 practice brief from Policy Analysis for California Education, or PACE. The brief draws on work with California Education Partners and argues that districts scale improvement by engaging the whole system — especially the superintendent and central office — rather than hoping good practice spreads informally from school to school. ### What changed here? What changed is not a new curriculum or a new state mandate. CSBA elevated a specific district-improvement playbook: if leaders want instructional gains to stick across many schools, they need structures and processes that support those gains everywhere. The brief frames district systems as the lever. That means schedules, coaching, decision rights, monitoring, and resource allocation — the boring machinery that usually gets ignored until reform stalls. ### Why isn’t “copy the best school” enough? Because scale is not just spread. The PACE brief leans on a long-running idea in school reform: a change has not really scaled if more people are doing it superficially, without local ownership, depth, or staying power. A district can point to lots of schools using the same language and still get weak results if the underlying routines never changed. what does the brief say districts get wrong? Three mistakes stand out. First, leaders delegate the work downward — often to principals or a central-office designee — and then senior leadership disengages. Second, districts assume teachers’ informal networks will carry the practice across campuses. Third, they try to do too many top priorities at once. The common thread is thin support from the system above the school. ### Why is the superintendent such a big deal? Because only district leaders can move the pieces that schools cannot. They can repurpose existing resources, create new structures where gaps exist, and watch districtwide progress closely enough to adjust when implementation starts drifting. Turns out that is the core argument here: improvement spreads when authority, money, time, and monitoring all line up behind it. ### What does “system support” look like in practice? PACE’s recent Chino Valley case study makes this concrete. The district spent 8 years building a Professional Learning Community model across 37 schools serving more than 25,000 students. What mattered was not just the model itself. It was the infrastructure around it — scheduled collaboration time, sustained coaching, consistent leadership, shared rubrics, guiding coalitions, and feedback loops that let the district keep adjusting. ### Where do school boards fit? Boards are not running the instructional program day to day, but they set the conditions. CSBA’s own governance guidance puts that in plain terms: boards set direction, establish structure through policy, support implementation, and ensure accountability through oversight. It also stresses a balance between systemwide coherence and site-level autonomy — which is exactly the tension this scaling debate is about. ### Why does this matter now? Because districts keep cycling through reforms that never harden into routine practice. A pilot works in one corner, enthusiasm fades, leadership turns over, and the district moves to the next thing. This brief is pushing against that pattern. The message is simple but demanding: durable instructional improvement is an organizational design problem before it is a workshop problem. ### Bottom line? If a district wants better teaching in every school, it has to build a system that makes better teaching easier to do and harder to drop. That is the real scaling method here — not inspiration, but infrastructure.