Think Small, Start Small.

SPI-Photo-005_Student and teacher working 1 on 1_2014_lr.jpg

Note: Here’s part two in our Learning to Improve series which is the primary focus of the School Performance Institute blog this year. In it, we spotlight issues related to building the capacity to do improvement science in schools while working on an important problem of practice.

Across sectors, this message is antithetical to the typical ethos. In sports, it’s “go big or go home.” For tech startups, funding often hinges on a founder’s ability to show her plan for quickly scaling the business. In education, we often roll out rapid, large-scale reforms such as a new curriculum or technology platform in response to some perceived problem.

Unfortunately, more often than not these changes are made under the conditions least likely to produce success. In essence, education has a serious learning-to-improve problem. Too often we go fast, learn slow, and fail to appreciate what it actually takes to make some promising idea work in practice.

SPI-Photo-029_Teacher explaining 2_2017_lr.jpg

Why Think Small, Start Small?

Full disclosure, I got push back from a number of advisors when I chose to make think small, start small one of School Performance Institute’s core values. I think it was a fair push. One advisor asked, “What about think big, start small?” I really liked that suggestion and almost made the change. The reason I stuck with the core value as written was because I think that education leaders have such a tendency to “think big, start big” that we needed an explicit statement in the opposite direction. Education reform is littered with big ideas that were rolled out fast and big only to be abandoned a few years later. Left in the never-ending wake of good intentions are teachers with initiative fatigue. And, the worst part is that within many of those ideas was a kernel of promise. Too often though, the logic of action was underspecified, and the people on the ground didn’t have the tools and processes necessary to make them work.

Living Think Small, Start Small

That brings us to how we’re holding true to our “think small, start small” core value this year. I’m running an improvement science project at Columbus Collegiate Academy-Main St., a middle school within the United Schools Network. I’m serving as the improvement advisor for a team focused on improving the percentage of 8th grade students that are on-track for success in high school. By thinking small, I mean that we are approaching problem-solving from a user-centered perspective while attending to variability. Our team is less interested in average effects or average rates and more interested in why each student is either on-track or off-track for high school success. Starting small - we’re focused on two 8th grade homerooms - allows us to learn about root causes for off-track status at the student level and develop interventions accordingly.

SPI-Photo-022_Pair of students smiling_2014.jpg

Our Improvement Project

Our team kicked off this work in earnest this week. We are focused on two outcomes. First, we want to use this project to build organizational capacity to do improvement science within a team. Growing the number of staff with improvement science capabilities will in turn grow our network’s capacity to do disciplined improvement work that improves student achievement. The second outcome is the important problem of practice on which the team is focused. The overarching aim of the improvement team is to triple the percentage of 8th grade students that are on-track for high school success as determined by the key indicators of attendance and GPA (check out this research if you’re interested in knowing more about why we chose these key indicators).

We’ve written a draft charter for our project. This living document provides guidance to the team and is composed of the project aim’s, causal system analysis, and working theory of improvement. We’ll revisit this document frequently as we better understand the problem on which we are focused.


Improvement Science

Most of our meeting time is focused on the problem of practice. However, we are also spending 10-15 minutes at the start of each meeting building the team’s knowledge of improvement science methodology.  

As I introduced the team to the methodology at our first meeting, I made three key points. First, this science involves going on an improvement journey. It’s not a workshop or a professional development session. Instead, it’s an effort to increase the capacity of an organization to produce successful outcomes reliably for different groups of students, being educated by different teachers, and in varied contexts. This takes time. Second, a core ethos of improvement science is “probably wrong, definitely incomplete.” I first saw this phrase used during an improvement summit hosted at the Carnegie Foundation for the Advancement of Teaching. It can be a bit scary to admit this, but it’s key in creating a culture of improvement. Third, the methodology is built around three core improvement questions: (1) What specifically are we trying to accomplish?, (2) How will we know that a change is an improvement?, and (3) What changes can we make that will result in improvement?

Ultimately, by thinking small and starting small and working through the improvement science methodology, we hope to make big changes in the lives of our students.   

More on How We’re Learning to Improve

At the School Performance Institute, we are studying what works in schools both within the United Schools Network as well as at high-performing, high-poverty schools across the country. We’re opening our doors to learn from others and share what we’ve learned at our second Study the Network workshop of the year at United Preparatory Academy on October 18th.  And, be sure to follow our progress this year as we share what we learn during our improvement journey.

John A. Dues is the Director of the School Performance Institute and Chief Learning Officer for the United Schools Network in Columbus, Ohio. The School Performance Institute is the learning and improvement arm of the United Schools Network. Send feedback to

John A. Dues