The NFER blog

Evidence for excellence in education

Teacher research: what’s the point?


“One of the most depressing experiences I’ve had is talking to teachers who describe a research project they have poured their heart and soul into that is methodologically crap”.

This was one of Ben Goldacre’s more provocative lines from his presentation at the Research Ed conference last month. If by ‘methodologically crap’ he means ‘not an RCT’, then – judging by the level of interest in sessions at the conference that discussed teachers undertaking their own small-scale research projects – there are an awful lot of teachers wasting their time.  And NFER’s efforts to create a suite of resources to support teachers doing their own research will also have been wasted. But surely we can’t all be wrong?

There’s much talk at the moment about ‘teachers engaging in research’. Indeed, this week’s TES carries a poll suggesting that 80 per cent of respondents believe educational research can be used in the classroom in the same way as doctors apply research to their work. But, like the shouted conversation in a crowded conference room, I fear some misunderstandings and crossed wires are occurring about what this really means. I’m going to try to unpick some of these here.

Teachers engaging in research

I would firstly like to differentiate between two distinct uses of the term ‘teachers engaging in research’.

  • Teachers accessing evidence produced and synthesised by research organisations, and using the findings to inform their practice (teachers as ‘consumers’ of research).
  • Teachers undertaking their own research, either in the form of small-scale action research, or as participants in larger projects coordinated with other schools (teachers as ‘producers’ of research).

The twist (plot spoiler alert) is that while being distinct, for reasons that will become clear, the two also overlap.

The purpose of research

Our main focus for now then, and the subject of Ben Goldacre’s comment, is on the second of these – teachers doing their own research. NFER will shortly be publishing a literature review exploring what is already known about the challenges and opportunities for all forms of teacher engagement in research. The key message emerging from this work for teachers doing their own research is to have a clear purpose in mind. We have identified three possibilities:

  1. To inform a national knowledge base. In other words, to generate research findings that can supplement evidence produced by research organisations and inform national practice decisions and spending choices.
  2. To support school improvement. Research is a whole-school (or consortium) exercise in critical reflection, by which priorities are identified and new solutions are developed and tested. It is the means to so-called ‘disciplined innovation’, and the audience for the findings is primarily internal to the individual school or consortium.
  3. To support individual professional development. Similar to school improvement, but conducted by individuals or small groups of teachers to support their own specific professional development needs or interests.

Informing a national knowledge base

Research intended to inform a national knowledge base obviously needs to satisfy high standards of methodological rigour (which, Ben Goldacre perhaps envisages in the form of groups of schools collaborating on an RCT). It’s the only way to justify the findings being used to inform decisions by other schools around the country. If this is the purpose of teacher’s research project, then I agree with the suggestion that a weak methodology implies wasted effort.

An end in its own right

However, in the case of the second and third purposes, the precise research methodology adopted becomes less important. It’s the particular way of thinking and the types of processes that research introduces to school or individual improvement that matter most. Reflecting on what is currently unknown, isn’t working, or needs improvement. Encouraging an openness to enquiry and innovation: something new that hasn’t been tried here before. Implementing new ideas in a conscious, deliberate, and critical fashion that leads to continuous evaluation and refinement. The asking of further questions about how we might do things better.

Don’t get me wrong, I still strongly advocate teachers adopting the best research methodology possible given the constraints they’re under (and NFER will be contributing further here soon by publishing our own evaluation policy). However, even in the absence of a gold standard approach, the process – and the culture it encourages – has great potential to bring about positive change. Indeed, while the first purpose for research above can be thought of as a means to and end (better pupil outcomes resulting from subsequent implementation of findings), the second and third purposes can be seen as ends in their own right, leading directly to improvement in professional practice and pupil outcomes.

(In the spirit of evidence informing practice, I’m conscious that I’ve not provided any evidence to back up this assertion. The deeply ironic paucity of evidence in this whole area is another emerging message from our literature review, which we’ll blog about another day, and are working to address).

Use of evidence

So, where do we find the overlap with the researcher-produced evidence that I alluded to earlier? Any one of the three types of teachers’ production of research described above will be most effective when undertaken in parallel with consumption of existing evidence. Teacher-led research intended to add to the national knowledge base should first consider what is already known, and organisations such as NFER have a responsibility to ensure this is as accessible as possible to teachers. In my view one of the greatest benefits of school and individual professional development-led research, is the opportunity, means and motivation it creates to draw on the wealth of evidence that others have already produced.

Looking for ideas more widely than your immediate experiences and peers can lead to innovations that no one in your school would otherwise have thought of. And looking more widely for ideas supported by evidence (rather than, say, taking on trust the claims from a commercial provider’s brochure) will lead to innovations that many other teachers have already tested, giving confidence that successes can be replicated in your school.

There are two possible ‘research systems’ we might nurture – research undertaken by research professionals, and research undertaken by teachers. Each has the potential to improve our schools. But they will be the most effective when they function in parallel, successfully engaging with one another like cogs in a well-oiled machine.

Author: Ben Durbin

Ben Durbin is Head of Impact at the National Foundation for Educational Research

5 thoughts on “Teacher research: what’s the point?

  1. Whilst, you make a good point, research of any kind seems likely to foster a culture of questioning, intolerance of the status quo and innovation, surely the whole danger of doing any old thing and calling it research (Bad Science) is that is muddies the waters between the kinds of research we should listen to and kinds we should not.

    For example did you ever try to argue with a Daily Mail reader that MMR was not likely to cause autism as Wakefield’s study only looked at 18 individuals and was not likely to be as valid as dozens of studies that tested thousands. If you did, you as I did, you will have noticed that such lines of reasoning cut no ice whatsoever.

    I would add to Ben Goldacre list of the most depressing experiences when talking to educators about research and you site a study you think is conclusive on an issue and an educator fires back ‘well, the Institute of Well Meaning but Under-qualified did a study and found the opposite’. Then when you try to engage them in a discussion of the relative merits of how the two studies were carried out you see the discussion is already over as the educators’ eyes have narrowed contemptuously as if their point could not be more fully made and they wonder off their work done. Without a discussion of methods all such dialogue achieves is the undermining of research in general, tarring the good stuff with the bad.

    Surely, research ultimately comes down to discussions about the strength of evidence from different methodologies. For example see discussion about what skulls of early man tell us about our origins, it all comes down to the quantity of data (quantity of skulls) By not sticking up for how research is carried out, surely we help to perpetuate the seemingly prevalent notion in education that all things called research are equal and all equally valid, when the business of research is the very opposite, questioning the validity of the evidence presented.

  2. Thanks for your comment, I absolutely agree that there are challenges around ensuring different types of research are used appropriately. Given that even professional researchers continue to debate competing evidence on various aspects of educational policy and practice, I don’t think this is a challenge that will go away anytime soon!

    As for the strength of different research methodologies teachers might adopt or read about, perhaps it’s helpful to differentiate between several different steps in a school enquiry process (assuming we’re not talking here about schools intending to add to the national knowledge base). Firstly, gaining a better understanding of ‘where we are now’: reflecting on your own practice and experience, opening yourself up to challenge and new ideas. Secondly, ‘what will make things better’: identifying the appropriate actions to take or strategies to adopt. Thirdly, asking yourself ‘did it work’? See my article in Sec Ed for some more thoughts on this ( Anyhow, whilst clearly we would encourage best methods to be adopted at all stages, I’d suggest that that the standards of evidence that are appropriate/realistic for each will differ. Specifically, the decision about what to do or how to intervene can and should be based on much more robust evidence than the school’s own measurement of where they are now or the level of progress they make.

  3. As a teacher of ESOL who has recently begun learning about Action Research, I found this article encouraging. My reasons to want to do research relate to self-evaluating, understanding my environment, and engaging in professional development. While I plan to use the most sound methodology, I might still make mistakes and have a lot of learning ahead. Regardless, I believe I still need to jump in and try to ask effective questions, observe well, and analyze the data I’ll collect. I certainly don’t pretend to get findings that represent all teachers and learners in my field, but I don’t believe this takes away from the research itself.

  4. I agree with Laura’s point above. Action research is empowering for teachers, providing support for reflection and leading to a better understanding of oneself in relation to one’s work. As I teacher educator, I want teachers to do exactly what Laura describes, for the reasons she gives. And I want them to do it well.

    Ben Goldacre is not an educationist. I much admire him when he’s tearing into the pseudoscience of Brain Gym or something similar, but I think he’s wrong to dismiss teacher research as ‘crap’ (is he guilty of just going for the headline, maybe, or unable to see beyond his own scientistic view of knowledge?), so long as teachers bear in mind Laura’s caveat that ‘I certainly don’t pretend to get findings that represent all teachers and learners in my field’. Is Goldacre, albeit by implication, seriously suggesting that medical practitioners (his own field) do not – and should not – learn about themselves as individuals in relation to their work through local investigation and reflective practice?

    However, like Ben Durbin’s blog article, I am not to saying that RCTs do not perhaps have a potential place in education. It will be useful to try to build a professional knowledge base in this way, and the emerging field of educational neuroscience – which the average teacher cannot research alone – may also provide some interesting insights. But education is not medicine, and contextual factors are, dare I suggest, more significant. Cognition can be both situated and social, yet neither of these figure – to the best of my knowledge – in medical decisions, at least not in the same way as we need to consider them in education. So, whilst RCTs may add to a general bank of knowledge, it will be important, as Ben Durbin suggests, to check the findings out in one’s own classroom – does this work with these kids, here in these circumstances, and if so/not, then why? In other words through action research!

    It may be found that what is ‘true’ in an RCT, may not be true in the individual case. Will this mean that the teacher will be judged as somehow deficient because they ‘couldn’t make it work’? We live, after all, increasingly in a culture of performativity, yet classrooms and schools are ‘messy’ places in terms of interacting variables. A ‘superhead’ in one school may fall flat on her/his face in another, since what worked in School A may not transfer to School B. We accept risks and side effects with medicines, and we accept a less-than 100% success rate. What will be the parallels in education if we take our view of ‘what works’ from RCTs?

    I do not want to see the day when teachers are expected only slavishly to implement the procedures which reflect the outcomes of RCTs – what a dystopian future the medicalisation of education would be!

  5. Thanks Laura and Tony for your comments – it sounds like we’re all on the same page regarding Action Research. That’s a great summary Laura of how it should be, and I suspect Ben Goldacre might even agree with the value of such teacher produced research, given its purpose and clear focus. If possible, I’d encourage you as part of the process of development to look for what other research has been conducted in your area of interest, and to factor this into your thinking (if nothing else, it could save you a lot of time not having to reinvent the wheel!)

    Tony – the challenges of applying findings from RCTs in a diverse range of settings are of course significant. I would argue that this is especially the case now when there are so few studies to choose from. However, I would love to see a future whereby there are many many studies covering given types of practice or intervention, conducted across different settings, which allow teachers to make better-informed judgements about what will work best with their class. I agree that no teacher should ever be required to slavishly implement procedures based on the findings from RCTs – not now, not ever. I do wonder whether even in the medical profession if this is how doctors would describe their professional experience though?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s