Way back in late 2025 I wrote a post on practitioner action research, and mentioned that I would later dive further into the issue (see update #131, “Promise and Pitfalls of Practitioner Action Research’).

It’s taken me a while, but I’m finally back with part 2 of that topic! I think that sharing this in the late summer is actually a pretty good time, given that many projects will be undertaken across the coming academic year. Perhaps you already have research in mind – if so, I hope this provides food for thought.

A brief recap – what is action research?

By action research, I mean research that reflects on and attempts to inform one’s own practice. It is carried out by the practitioner, rather than someone external, on their own students and in their own context. Overlapping terms that are widely used include practitioner research, practitioner enquiry, or teacher research.

Projects tend to be narrower and more classroom-focused than a lot of other education research. There are many things that could be researched, but examples might include:

• Checking to see whether a change in classroom organisation impacts on student behaviour or motivation, for example by giving questionnaires to students;

• Trying out a new form of group work and analysing the outcome based on test scores;

• Interviewing parents about their children’s experiences with primary-to-secondary transitions.

It’s easy to see the potential of doing very grounded research in real contexts. How much better does it sound to find out about behaviour in my own class than from the broader research literature, for example? This is the promise of practitioner action research.

However, as I raised in my earlier post, there are real risks of bias. Many interventions are studied by practitioners that already think that they are a good idea.

Consider, for example, the idea of flipped classrooms or technology use – they tend to be researched by enthusiasts rather than sceptics. Coupled with rather wobbly research methodology, and there is a lot of potential for educators to see what they want to see. Learning myths may persist because people find this pseudo-evidence from action research. These are the perils that I referred to before.

Why Errors Occur: A Scenario

Let’s consider a scenario to further illustrate how flawed conclusions can be drawn from action research:

Bob decides to explore the spacing effect (see update #39). The class does one task with spaced practice and one without, and test results are then compared using a statistical test. Bob concludes that spacing has not made a difference.

How should Bob respond? According to the logic of action research, they have a finding that is relevant to their unique context, and should decide not to apply the spacing effect in future because it (apparently) didn’t work. They may also be encouraged to share this with others as an example of good practice (🤯).

That would actually be a major mistake. We know from the broader research literature that the spacing effect does work. This leads two possibilities:

1. There is something about Bob’s context that is so unique that it overturns decades of research on learning and memory.

2. The experiment was flawed, leading to a false negative. For example, the sample size (limited to the numbers in a class) could be too small to detect a statistical difference, the tests may have differed in their difficulty level, or a lack of experimental control might have distorted the findings in any number of ways.

I would be very strongly inclined to believe possibility #2! Of course, the same logic holds if the teacher does find an effect. This is perhaps even more misleading and hard to avoid – people expect intervention X to work, their results show a difference of some kind, and so they conclude that their research has provided evidence for intervention X.

The problems are exacerbated by the fact that most teachers are not experts in research methodology and statistics, and may lack experience, too. That is before we consider issues about research ethics, which are also riskier with inexperienced researchers.

In defence…

But wait, you might ask—didn’t I actually write a book about research for teachers? Well, yes!

It might sound like I have really changed my mind about teachers engaging with research, but that’s not so. My 2019 book The Teacher’s Guide to Research (more here) does not suggest that teachers should blunder into research projects without ethical scrutiny, make changes on the basis of tiny samples, or share findings that lack sound evidence to support them. It also doesn’t suggest ignoring the existing theory and evidence on an issue!

In fact, The Teacher’s Guide to Research does not focus just on teachers doing research, but also engaging with it. There will often be more to be gained by engaging with existing research evidence on an issue than by carrying out your own project.

Take reading, for example. I think that a primary school teacher would be much better off reading reputable systematic reviews of research into reading than trying to figure out how reading develops via action research with their own class.

However, once a teacher has begun to understand the theory, the existing evidence as well as key areas of uncertainty, they may start to generate important questions that can help the education community as a whole understand how ideas are best implemented.

On balance, is action research a good use of time?

There no doubt are cases where good projects are done and where teachers have learned something useful. But we have to balance the time spent on this with other things that they could have been doing, such as:

• Reading up on learning and memory;

• Working on their classroom management skills;

• Developing better explanations and questioning;

• Upgrading their curricular knowledge;

The points are arguably more core to successful pedagogy than the majority of action research topics.

However, I still believe that engaging with research is highly valuable and motivating for teachers – at times even transformative. Even a poorly conducted study can be a good learning opportunity, and act as a stimulus to read evidence and to engage with the science of learning community.

And if we want to start engaging with research, then trying it out ourselves is definitely one of the best ways to understand the methodology of the research that we read.

So, I’m perfectly happy for all educators to engage in action research. I’d just warn against reading too much into findings – at least while one’s understanding of methodology is still developing.

This leads me on to the final part of this series of posts – where I talk about guidelines for successful projects. You can check it out next week!

Please consider sharing this post with relevant colleagues! Thank you!

Share

All the best for the week,

Jonathan

Last week: Episodic and Semantic Memory

I hope you are enjoying Memory and Metacognition Updates. Subscribe for free to receive new posts and support my work – and consider sharing it with colleagues!! Thank you!

Please note that my slides and similar materials are shared under a CC BY-NC 4.0 license. This means you can use or adapt them with attribution for non-commercial purposes. If you wish to use my materials for other purposes, feel free to get in touch.