The Education Endowment Fund (EEF) recently found that teachers do not engage with research. The question is why? The EEF claim that a lack of time and lack of support from senior leadership are the main culprits. But if we think about it a little that seems like a rather implausible explanation: if research can inform better teaching practice, then the EEF have concluded that teachers aren’t willing to invest the time to make themselves better teachers, and that senior teachers don’t want their more junior colleagues to become better teachers. 

That sounds absurd. I have met very few people who do not want to improve the way they do things in any job, and I would suggest that teachers, as a group, are more motivated to become better at their jobs than most.

Rather, I suspect that the real issue is that teachers don’t feel as if engaging with research is worth their time, a big difference.

So the question then becomes, why do teachers feel that engaging in research isn’t worth their time? Is it a problem with teachers or is it a problem with the research?

I want to argue that the problem is with the research, and not with the teachers, and I want to do that based on my experience as an academic physicist: I have in-depth, first-hand experience of the career structures and incentives of academic research.

1) Education research is not written by or for teachers.

Go and look at any field that is heavily evidence based and relies on research and you will find something striking: the people who do the research are the people who use the research.

For example, in medicine, it is mainly doctors or researchers working closely with doctors – often physically located in a hospital or university hospital – that carry out research. In the NHS, even the “academic medicine” research career paths require you to become a fully qualified doctor and to see patients once you have become a research doctor. That means the main motivation is to do useful research that can inform practice. This changes both the research they choose to do in the first place and the way they disseminate that research.

This separation of teaching and education research creates huge misaligned incentives: the goal of a teacher is to find out what works to help students learn best; the goal of an education researcher is to play the game of academia in order to climb up the academic career ladder. Informing and changing teacher practice is only a secondary goal for education researchers: it’s a nicety rather than a necessity. 

You might not think this is a big issue, but when you look at what academic careers reward and what “playing the academic game” consists of you’ll quickly notice that this separation of teaching and research is responsible for most of the barriers between teachers and research.

Academia rewards:

  1. Publishing lots of papers. Ever felt overwhelmed at the sheer volume of research out there? Here’s your answer.
  2. Getting those papers cited a lot. What’s one way or getting a lot of citations for your research? Publish controversial findings and submit a press release. Sound familiar?
  3. Publish papers in “high-impact” journals. These journals largely publish results in “fashionable” areas and also those that fit with their own particular “world-view”, which explains why research on stuff we know is rubbish still gets done and still gets published. Why does so much education research revolve around progressive ideologies? Here’s your answer.

Now let’s say you’re not deterred by that: you’re an extremely determined teacher who wants to get to the truth. You’re willing to wade through the crap and put in the time necessary. What are the next barriers you face?

2) Open access.

Have you ever tried to read a paper only to find that it’ll cost you the princely sum of £30? Academic pay walls make it impossible for anyone outside of a university to access research. There are of course some papers that you can still access. But if you really want to understand the total evidence base in a particular area you need to read all of the relevant research. If you want to know what the best way of marking students’ work is but you can only access a handful of papers and not the other 40 that exist, it is impossible to come to an accurate conclusion of the evidence.

In physics, we basically solved this problem 20 years ago with something called the arXiv: most researchers (certainly those who want their research to get read) post a version of their papers onto the arXiv that can be freely accessed.

It is high time education researchers did something similar. I’m not getting my hopes up though, since the entire field of biology only just got onboard with the biorxiv and it’s still far from popular or main-stream.

3) Education research papers are too long and badly written.

What’s the average length of a research paper in education? I don’t know the answer but I have read many that are 30 pages or more. I’m not sure I’ve ever seen one with fewer than 5 pages.

In physics, we typically publish papers 4-10 pages long. Nearly every “high-impact”,  “highly-cited”, or “highly-interesting” paper is 4 pages long, because those papers tend to have been submitted to the so-called glamour journals, which demand short papers.

The reason for this is that a paper is an announcement of a discovery or new piece of knowledge. The goal is to explain the significance of your discovery and back it up with evidence as quickly as possible. These journals know that nobody except those who are working directly in that field or who are very interested  in the study will read anything over 5 pages, and since they want to get the research to as wide an audience as possible, they demand that you think hard about the message of your research and condense it.

I hope nobody wants to tell me that a typical education research paper is more complicated or sophisticated than a typical physics paper. So if physicists can do it, why can’t education researchers? For example, is there really any need to publish tables full of data in the middle of your paper? Stick it at the end or in supplementary information. Better yet, plot a graph!

Education research papers tend to also be very boring and full of jargon. Jargon serves only as a barrier-to-entry for those who are not on the “inside” of your little crowd. Many fields do this on purpose (some parts of philosophy, for example) so that only the indoctrinated read the papers, lest their entire field be found out as a con.

They tend to be boring partly because of their length and partly because no effort has been made into crafting a succinct, well-written, memorable message for the paper. It’s not uncommon for me to get 10 pages into an education research paper and think: Why have they done this? What’s the point in this paper? What are they trying to tell me?

4) Lack of expertise.

Much of the most interesting and informative educational research is highly quantitative. I have a PhD in theoretical physics, yet I spend a lot of time learning about the statistical techniques and methods used in education research, because you can’t properly evaluate the validity of research unless you can interrogate the methods. I do it because I’m nerdy and because I care about this stuff, and I know from experience that I have the ability to understand it provided I put in the effort.

But is that true of most teachers? What about teachers whose last contact with mathematics was GCSE Maths 10+ years ago? Would they have the same confidence or even necessary background skills to learn the statistics necessary to evaluate the research? Some have done it, so it is clearly possible, but it requires a massive investment of time, and faith that the huge investment of time (probably several years) will actually lead to the reward of becoming a better teacher. Given the dubious quality of education research, that risk-reward pay-off diagram doesn’t look too great to me.  

You can see why lack of expertise is a big problem when people parrot off studies they’ve read in defence of their arguments or belief systems. “This study showed that…”, “I read a study that showed that…”

Have you properly read those studies? Did you notice that their methodology was crap? Did you realise that their statistical model can’t do what they claimed it could? Did you notice that they compared two things in which many other factors varied and therefore couldn’t be sure the differences were due to the things they compared? A case in point, Jo Boaler’s paper on discovery vs traditional maths learning, a paper that has been cited over 500 times. Clearly, 500 people have thought this research has value (in physics, some papers with fewer than 500 citations have won Nobel prizes), yet if I were a referee, I would have flat-out rejected that paper for publication.  

This also links back to 1) in that we cannot trust education researchers to do the right research or to think about the implications of classroom practice. The onus then falls onto teachers themselves to properly evaluate and critique the research and this requires teachers to have essentially the same skills as the researchers themselves.

How can we fix it?

This is the really depressing part, because I can’t really see how things are going to change.

A large part of the problem is the academic system and the bad incentives inherent in it. I have not seen any leadership on this issue in the 5 years I’ve been in academia. Academics may be clever but they’re still human, and they respond to incentives like the rest of us.

As long as universities still have a publish or perish culture we’ll continue to be subjected to piles of bad research; as long as teachers and education researchers still remain separate people, education research will continue to be less relevant than it could be; and as long as only those who work at a university can access papers, teachers will remain unable to evaluate properly the research that already exists. 

And finally, short of turning every teacher into a statistician, we are going to struggle to critically evaluate research and to get close to the “truth” about what works and what doesn’t.

One potential way of fixing this is to do what we have done in the sciences for a long time, and write comprehensive review papers. Review papers are written by an expert who has read (and hopefully understood) all of the relevant research on a particular topic and then tries to summarise the results and present it in a manner that is understandable to a non-expert. But the quality of review papers varies a lot, and we are not going to get teachers to read this stuff unless they are consistently high quality.

Crucially, the person who writes the review papers must be a teacher, or someone who is genuinely interested in informing classroom practice, rather than an education research academic, for all of the reasons enumerated above: if we leave it to education researchers, they will just try to write review articles that impress their education-researcher friends and colleagues.

Another potential way is to go down the route of having a central body that tries to evaluate and summarise research for teachers to use. I believe the EEF is trying to fulfil this role, but you only have to look at their recent summary of the evidence on marking to see that they’re not fully realising their potential or being as useful to teachers as they could be.

And then back to the EEF’s original claim: that lack of time and lack of support from senior leadership are the reasons for teachers’ lack of research engagement.

If those were the real reasons, the solutions are simple:

  1. Reduce teacher workload to free up time to engage in research. Does anyone honestly think that the majority of teachers would spend this time engaging in research? Or is it more likely that they would use it to have a better (rightly deserved) work-life balance?
  2. Increase support from senior leadership. SLT: “We support you to engage in education research!” Teacher: “Great, can you tell me why the authors of this paper used a multi-level model rather than just a multi-factor linear regression model?” SLT: “Errrm”.

Of course, if support from senior leadership actually means that they’re willing to pay for staff to do extra degrees, give them time off to do it, and they’re willing to accept that it will take several years before teachers have developed the necessary skills to evaluate research fully, then I’m all for it, but I highly doubt that’s what they had in mind.

In summary, if academics want their research to inform teaching practice, and they want teaching to become an evidence-based profession, then they need to engage in a large degree of introspection and improve the way they do research.