How do you actually get research-based knowledge into the classroom? This is a question my colleagues at Leerpunt and I keep coming back to. Leerpunt is a knowledge broker for education in Flanders, Belgium. Publishing a report or a set of guidelines is one thing, but ensuring that the knowledge from it is actually put to use in schools is a completely different challenge.

That is why the abstract of a new study by Mareike Ehlert and Elmar Souvignier triggered my curiosity. The two researchers investigated what happens when you turn teachers not only into users of research but also into disseminators of it. I deliberately wrote “the abstract” because it was in English. The actual study is in… German. And I have to admit two things: 1) I genuinely do not speak German, despite many attempts to learn it; and 2) I had to rely on ChatGPT and Google Translate to understand the study.

The study I deciphered in this way took place in German primary schools and focused entirely on evidence-informed reading instruction. First, ten experienced teachers underwent six months of training to become “multipliers.” These ten teachers then trained another 43 colleagues from eleven schools. A group of 33 other teachers from eleven different schools served as a waitlist control group. The professional development was relatively intensive: about thirty hours spread over six months, comprising theory and exercises, e-learning modules, practical teaching materials, and opportunities to try out new methods first.

Afterward, all the teachers in the experimental group implemented the methods. For the first six months, they could still call on their “multiplier” for support, but after that, they were on their own. I think this is an important and relevant point. After all, educational projects can work like a charm as long as researchers and facilitators keep investing energy in them. But what happens when they’re gone?

That is exactly what Ehlert and Souvignier examined. A year and a half later, they still found differences. Teachers in the intervention group reported greater knowledge and higher confidence in their own abilities than those in the control group. Collaboration appeared to get off to a relatively slow start; in fact, it was only after a year and a half that a significant difference in teacher collaboration emerged.

But something else happened. The kind of thing people like me dream of. The approach began to spread further throughout the schools. After eighteen months, 71 percent of participating teachers had discussed the content informally with colleagues, nearly 30 percent had done so during school meetings, and 63 percent had shared the actual teaching materials with colleagues.

This suggests that research-based knowledge may not spread simply because someone tells their colleagues what the research says. Something concrete, something you can put to use the very next day, seems to find its way to other teachers more easily. The researchers themselves point to the immediate usability of the teaching materials as a possible explanation. We already know this from looking at Leerpunt’s own statistics.

Incidentally, the researchers expected the teachers to use the reading approach for 60 minutes a week. In reality, they spent considerably less time on it. During the guided phase, the average was 36 minutes, followed by 29 minutes. Yet, the observed differences persisted. The authors therefore suggest that an educational innovation needs to withstand a certain degree of adaptation to be robust enough, so to speak. After all, in real-world schools, an intervention is rarely implemented exactly as its creators designed it.

This research offers hope. I don’t want to dash that hope now, but we have reached the nuance section. The study’s sample size is quite small, and the key findings are based on self-reported data. The study didn’t even measure the  ‘knowledge’ acquired using a knowledge test. Nor do we know which component of the overall package was responsible for the observed differences.

Nevertheless, reading this brought to mind the various ‘Research School’ models that have emerged in different countries, although they go by different names. Naturally, this study does not provide direct evidence that such models work by definition. However, it does illustrate a mechanism that can make an underlying concept compelling.

You could try to get research-based knowledge directly to every single teacher. Alternatively, you could invest in teachers and schools that not only learn to work in an evidence-informed way themselves but also help others to do so. That leads to a model like this:

research -> intensively supported teachers -> colleagues -> concrete practice -> further dissemination.

At the same time, there is a note of caution here. An ‘Inspiration School’ cannot simply be a good school that others visit for just an afternoon. The approach in this study involved providing people with time, intensive professional development, concrete materials, and long-term support.