Imagine a leadership team that reads a respected summary on retrieval practice over the summer. In September, every lesson in the school starts with a low-stakes quiz. By February the routine is everywhere. But ask whether it has improved learning, and the honest answer is that nobody has checked. The practice came from research. The reason for keeping it is now the same thing it was meant to replace: it feels right.
That example is hypothetical, but most school leaders will recognize it. It shows where evidence-based teaching practices usually get stuck. Finding credible research is rarely the problem. The hard part comes after that, when a finding has to travel from a leadership meeting into thirty different classrooms, and when someone has to show that it made a difference for these students.
So how does a school move from intuition to evidence-informed practice in every classroom? In short:
- Treat research as a starting hypothesis.
- Help teachers translate it with peers rather than by memo.
- Use your own assessment evidence to check whether the change worked.
The rest of this article explains why each step matters.
Finding the Research Is No Longer the Hard Part
In February 2024, the Education Endowment Foundation (EEF) used the survey app Teacher Tapp to ask teachers and school leaders in England six questions about how they use research evidence. 2,312 senior leaders and headteachers were asked whether they look for research before asking teachers to change their practice. 94% said they do this at least sometimes: 30% always, 40% often and 24% sometimes. The EEF reads this as a sign that consulting research is now a normal expectation in English schools.
There are limits to this evidence. The answers are self-reported, they come from one country, and they come from people who use one app. The data are weighted to match the teacher workforce in England, but the EEF itself notes that app users may not fully represent the profession. Self-report tells us what leaders say they do. It does not tell us how well they do it.
The Gap Opens at the Classroom Door
The same survey asked whether respondents find it “hard to apply educational research to my practice.” Across 9,313 respondents, 54% agreed to some degree. The split by role is the more useful finding.
| Role | Find it hard to apply research (agree to any degree) | Regularly apply published research |
| Classroom teachers | 60% | 30% |
| Middle leaders | 52% | 41% |
| Senior leaders (excl. heads) | 42% | 50% |
| Headteachers | 36% | 49% |
Read the two columns together. The people least likely to struggle with research are usually the people who decide which practice the school adopts. The people most likely to struggle are the ones asked to deliver it every day. Experience matters too: 60% of people with under five years in teaching found research hard to apply, compared with 49% of those with more than twenty years.
The survey also shows where teachers actually get their advice. 68% of respondents named immediate colleagues as a source of advice they regularly apply. Only 37% named published research. When asked how they prefer to engage with research, only 14% chose material sent out by senior leadership, and only 11% chose email newsletters.

For school leaders, one implication is uncomfortable but practical. A well-sourced slide in a staff briefing is unlikely, on its own, to change what happens in classrooms. This does not mean teachers reject research. It suggests research needs a human route into practice.
If the people who choose a practice find research easy to apply, and the people who deliver it find it hard, whose practice has actually become evidence-based?
“What Feels Right” Is Not the Enemy
It is easy to frame this as research versus instinct. That is the wrong fight. When a teacher feels something is right, that feeling often contains real information: knowledge of the class, the timetable, or the student who shuts down when called on. Evidence-informed practice has always meant combining research with professional expertise and local knowledge.
The real problem is untested intuition, whether it belongs to a teacher or to a leadership team. A practice adopted because it appeared in a well-known study can become just as unexamined as one adopted because a respected colleague swears by it. Both can harden into habit. Neither is evidence-based until someone checks what it is doing for students in this school.
Research Tells You What Is Likely to Work. Your Own Evidence Tells You Whether It Did
External research answers a general question: across many settings, what tends to help? A school-wide decision also needs a local question answered: is it working here, for these students, in the way we expected?
That second question needs internal evidence, and assessment is usually the most direct source. Suppose a department adopts a new approach to teaching fractions. The test is not whether staff like it. The test is whether the specific skills it targeted moved between one assessment cycle and the next, and for which students.
That requires two things. The assessment must be designed to show change at skill level, and teachers must be confident enough to read it. Teachers also need support in turning assessment reports into testable teaching decisions. Without that support, the evidence sits unread.
Five Questions Before Any Practice Goes School-Wide
| Question to ask | Why it matters | What to look for |
| What learning problem is this meant to fix? | A practice with no defined problem cannot be judged | A named skill or gap, ideally visible in current assessment data |
| What does the research actually show, and for whom? | Findings rarely transfer unchanged across ages and contexts | The population, setting and size of effect in the original source |
| What will teachers do differently on Monday? | A principle is not yet a practice | Concrete routines, worked examples and time to rehearse them |
| How will teachers learn it from each other? | Peer routes are more trusted than top-down ones | Department time, coaching, lesson study or PLC cycles |
| What evidence will tell us to keep, adapt or stop it? | Without a check, “evidence-based” becomes habit | A planned reassessment point and a measure agreed before roll-out |
The last question is the one schools most often skip. It needs agreeing in advance, because once a practice is everywhere it becomes politically hard to stop. These five questions fit naturally inside a teaching and learning effectiveness framework. That way, new practices are judged by the same standard as existing ones.
Teachers Adopt Practice Through People, Not Memos
The EEF findings match what many leaders see in practice: research moves through conversation. That means structures for collaboration matter more than the amount of research shared.
Professional learning communities give teachers a place to test a practice and compare results, especially when they start from student work rather than a slide deck. Training design matters too. Much of why most teacher training fails comes down to a single input session with no rehearsal and no follow-up.
If a practice has been in place for a year and no one can show what changed for students, is it still evidence-based, or just familiar?
Answering that question honestly needs a culture of continuous improvement. In that kind of culture, stopping something that did not work counts as good practice, not failure.
The Real Shift Is From Adoption to Checking
Moving away from “what feels right” does not mean removing judgement from teaching. It means asking judgement to show its working. Leaders in England already report looking for research before making changes, and that habit is worth keeping. The next step is to make sure research reaches teachers through routes they trust, and that every school-wide practice comes with a plan to find out whether it helped.
FAQs
Q: What are evidence-based teaching practices?
A: They are teaching approaches chosen because credible research suggests they are likely to improve learning. They are then applied with professional judgement and checked against local evidence of student progress.
Q: Is evidence-based teaching the same as evidence-informed teaching?
A: The terms are often used interchangeably. “Evidence-informed” usually stresses that research is combined with teacher expertise and school context rather than followed as a fixed rule. In practice, most schools need that combined approach.
Q: Why do teachers find it hard to apply education research?
A: In the EEF’s 2024 survey in England, 60% of classroom teachers said they find research hard to apply, compared with 36% of headteachers. The EEF suggests that classroom teachers may have fewer chances to attend events or less protected time to read research.
Q: How can school leaders tell if a new teaching practice is working?
A: Agree before roll-out which skills the practice should improve. Then compare assessment results on those skills across cycles, by class and by student group. Staff enjoyment and visible adoption are not evidence of impact.
Q: How long should a school trial a practice before deciding to keep it?
A: There is no single rule. A reasonable minimum is one full assessment cycle, so the targeted skills can be measured before and after. This is a practical judgement, not a research finding.