Schools should team up with academics to research the effectiveness of their teaching and work out how best to improve standards, according to a researcher from Harvard University in the US.
Too much academic research is done in isolation from schools, believes Christina Hinton from the university's Graduate School of Education. To maximise its impact, more research should be driven by teachers.
Dr Hinton leads the International Research Schools network, which is attempting to change the relationship between schools and universities. It is working with seven schools around the world, including leading English independents Wellington College and Bedales.
"One of the things which is very different about this research model is that we are working continually in collaboration with schools to make sure research is grounded in their needs and interests," Dr Hinton said. "A lot of academic research is disconnected from what is needed on the ground, so we are involving teachers in collecting evidence-based practice rather than telling them what to do."
Unlike privately conducted research, the results would be published in academic journals and could be distributed more widely, she added.
Alistair McConville, director of teaching and learning at Bedales in Hampshire, said he wanted to work with researchers directly, rather than having research "filtered through third parties".
Bedales asked the Harvard academics for help in encouraging pupils to become "inquisitive thinkers", he said. The researchers conducted a literature review that explored which factors contributed to inquisitiveness. Then, with the help of students, they developed a survey to measure inquisitiveness, which provided a baseline for the project.
"The research says it's really important, if you want to support inquisitiveness, that there are good underlying relationships between students and teachers and between the students themselves," Mr McConville said. "So, for Year 9 [when students first enter the school] we have restructured the tutoring system so the tutor becomes more central to the experience."
Carl Hendrick, the newly appointed head of research at Wellington College in Berkshire, said that through its partnership with Harvard the school would be investigating areas such as resilience and growth mindsets.
He added that part of his job was to share useful findings not just within Wellington but also with other schools. "My vision for the research centre at Wellington is to be a sort of hub where people can work together, collaborate and find out what works," he said.
This is one of a burgeoning number of initiatives that aim to embed the principles of research in schools, by assessing the evidence behind initiatives, disseminating good ideas and then evaluating them in context.
Tom Bennett, a teacher at Raine's Foundation School in East London and founder of the ResearchED conference, said that classroom teachers were tired of being "steamrollered" by bad ideas. "I was annoyed with dodgy education research," he said. "[Physician and science writer] Ben Goldacre was important because he catalysed people to think science was important and there was something you could do about bad science.
"There is a wave of people who are sceptical about what is presented to them as dogma, and because of social media people can now talk to each other really, really quickly."
But although the idea of research-rich schools is gaining traction, thus far there is little evidence to support it. The Education Endowment Foundation, which was set up by the UK government to fund educational research, has just given the go-ahead to several trials that will examine the best ways of taking evidence from research in academic journals and putting it into practice.
The studies, which will cost pound;1.8 million and involve 8,000 schools across England, will test different methods of communicating research findings to school staff and explore whether teachers' knowledge of research has any effect on pupils.
John Tomsett, headteacher of Huntington School in York, which is leading one of the trials, said: "We are interested to know if research really does make a difference or if it is the new fad - the new thinking hats or Vak [visual, audio, kinaesthetic learning]. Does it make any difference?"