The research that changed how I lead schools

Why building collective teacher efficacy on purpose is a leadership responsibility. This post points to the research behind this claim and what it means in practice.

By Ben Marsh, Director, The American School of Kinshasa (TASOK), Democratic Republic of the Congo

For years, I led schools with commitment, care, and a genuine desire to improve student learning, and in many ways, I did. Then I encountered research that changed how I understood my role.

I had worked on systems, supported teachers, strengthened assessment practices, invested in professional learning, and paid close attention to results. I believed I was making sound leadership decisions, and often I was.

But when I saw John Hattie’s Visible Learning work identify collective teacher efficacy—teachers’ shared belief (backed by evidence) that they can improve student learning—as one of the strongest influences on student learning, I did not feel affirmed. I felt embarrassed.

I had been leading schools for years, but I had not intentionally designed schools around the condition the research suggested mattered most, nor around the leadership practices most likely to help it grow.

That realization changed more than my thinking. It changed my sense of responsibility.

I could no longer see school leadership as the management of programs, people, and operations around the edges of learning.

I began to see leadership as the intentional design of the conditions that produce the adult efficacy most likely to improve student learning.

What became clear to me is that schools do not improve because adults leave a conference inspired; they improve when leadership designs the conditions for better teaching to accumulate over time.

A lot of work in education is stimulating, but not cumulative. It creates energy, language, and activity, but not always durable improvement. Collective teacher efficacy challenged me to think differently. If it has the strongest influence on student learning, then my most impactful work as a leader is not to adopt more initiatives; it is to deliberately design the conditions that help teachers build confidence in their collective impact.

That shift changed how I viewed school improvement.

I became less interested in activity on its own and more interested in accumulation. What exactly changes because of this work? What leadership behavior changes? What teacher behavior changes? What student outcomes change? What evidence shows it is working? Those questions became more important to me than whether an idea was popular, energizing, or well presented.

I also came to see that collective teacher efficacy is often misunderstood.

It is sometimes described as a teacher mindset, or as a feature of school culture. I do not think that goes far enough. Collective teacher efficacy does not grow because leaders talk about it. It grows when leadership teams intentionally design the conditions that allow it to take root.

Jennifer Donohoo’s work helped make that responsibility more practical for me. She identifies four leadership practices that support the development of collective teacher efficacy: creating opportunities for meaningful collaboration, empowering teachers, establishing goals and high expectations, and helping teams interpret results.

What mattered to me was not only the framework itself, but what it clarified.

Collective teacher efficacy is not built through slogans or good intentions. It is built when leadership deliberately creates the conditions in which teachers can work together, act with confidence, and see evidence that their collective efforts are making a difference.

One moment brought this into sharp focus for me. A high school math team was frustrated that students were arriving underprepared for high school mathematics. That kind of complaint is common in schools. It is easy for one section of a school to believe the problem belongs to another.

But in this case, the middle school team looked closely at the data. Their students were actually performing strongly across almost every area except one: statistics.

That led to honest professional conversation. Middle school teachers admitted that statistics was the unit they least liked teaching, so it had become the final unit of the year, the one they often did not reach.

Once they saw that pattern clearly, they changed it. They moved the unit earlier. Over the next four years, results improved, and students began leaving middle school as strong performers in statistics.

What matters to me in that story is not only the improvement in results. It is the shift in behavior.

The conversation moved from blame to shared responsibility. Teachers used evidence to identify a real problem, examined their own practice honestly, and changed what they were doing together.

That is what collective teacher efficacy looks like in practice. It is not a slogan. It is the growing belief that, together, we can make changes that improve student learning because together we just did.

Experiences like that changed the kinds of questions I ask as a leader.

If we say collaboration matters, what does that collaboration actually produce? If we say teachers are empowered, what decisions are they trusted to make? If we say we have high expectations, how are those expectations made clear and shared? If we say we use evidence, how are teams supported to interpret results in ways that improve teaching and learning?

This is also why I have become more cautious about strategic vagueness in education.

Schools do not improve because we talk in broad terms about innovation, alignment, or culture. They improve when leadership creates the conditions for stronger teaching to spread, deepen, and endure across a school.

I no longer think the highest leverage work of leadership is managing initiatives; it is designing the conditions for collective teacher efficacy.

For me, that has brought greater clarity to leadership itself.

Leadership is not at its strongest when it manages many moving parts at once. It is at its strongest when it deliberately creates conditions that help teachers, with evidence, believe that their collective work improves student learning.

I still believe deeply in relationships, culture, and trust. But I no longer see collective teacher efficacy as something that sits alongside leadership. I see it as one of the clearest tests of leadership.

John Hattie’s Visible Learning work has long identified collective teacher efficacy as one of the strongest influences on student achievement, with an effect size often cited at 1.57. In plain language, that means this is not a small influence on learning. In Hattie’s work, 0.40 is a benchmark for meaningful impact, and collective teacher efficacy sits far above that at 1.57.

When teachers share a strong belief, grounded in evidence and experience, that they can help students succeed together, student learning is much more likely to improve.

If that is true, then this cannot remain a peripheral idea in school improvement. It has to move closer to the center of how we lead.

That research changed how I lead schools.

It changed what I pay attention to. It changed the questions I ask. It changed what I believe leadership is for. And I believe it should change something for us as a profession as well. If collective teacher efficacy has the strongest influence on student learning, then school leaders cannot leave it to chance.

We have to design for it, intentionally, consistently, and with the same seriousness we bring to every other priority we claim matters for students.