Learning with peers by asking questions
Action Learning
A small group of colleagues helps each person think through a real problem from their own work by asking questions, not giving advice
What it says, in plain language
The idea in one sentence: a few colleagues meet regularly, each brings a real problem from their own work, and the others help by asking questions rather than giving advice. The person with the problem usually leaves without “the answer”, but with a clearer view of what the problem really is and what they’ll try next.
You’ve seen this already
Think about the last time you were stuck on a decision, like whether to move house or change jobs. One friend tells you straight away what they would do. Another asks, “What’s really bothering you about it?” and “What have you already tried?”
The second friend’s questions make you do the thinking. Quite often, you realise the problem wasn’t the one you started with. And because you worked it out, you’re more likely to act on it.
Why it matters for your teaching
Most training gives people expert knowledge: the guideline, the policy, the right way to do it. That matters. But it can’t answer a problem that is new, messy or specific to your team.
Picture an action learning group (a “set”) of four people who all teach as part of their job: a pharmacist, a healthcare assistant team leader, a physio and a midwife. They meet monthly. This time the team leader brings a problem: “New healthcare assistants come to my observations (obs) teaching, but they still don’t tell anyone when a score is high.”
The others don’t jump in with “do a quiz” or “make a poster”. They ask:
- “What exactly do you want them to do differently?”
- “What happens on a night shift when they spot a high score?”
- “Who do they feel able to tell?”
By the end, she sees that the new starters know the rule. What they lack is confidence about who to tell on nights. So she’ll practise the actual call with them, and talk to the nurse in charge.
That is the heart of Teaching That Lands: start with what people need to do differently, then build the teaching from there. Good questions from peers often get you to that starting point faster than advice does.
Try this
- Ask before you advise. When a colleague brings you a problem, ask three questions before offering ideas. “Have you tried just being firmer?” is advice in disguise.
- Bring a real problem. Pick something you’re actually stuck on, not a made-up case.
- Try it in threes. One person describes a teaching problem in two minutes. The other two ask only questions for the next eight. Then they can offer ideas. The person with the problem finishes by saying one thing they’ll do.
- Meet again. Agree a date a month away and report back on what you tried. Action learning is meant to run over weeks or months, not as a one-off.
How sure are we?
Action learning is a widely used method, not a single tested finding. Most of the evidence comes from people who run sets describing what happened, and from case studies. Few studies compare it fairly with other kinds of training, so we can’t say how much better (or worse) it is. It has not been well tested as a way to develop people who teach in healthcare.
A common misreading is that experts are no use. Its founder argued that expert knowledge alone isn’t enough for problems nobody has solved before. He didn’t say experts had nothing to offer. Another trap: without someone keeping to the rules, sets often drift into swapping advice, especially among clinicians used to fixing problems fast.
Teaching That Lands doesn’t run action learning in its three sessions. It is a good way to keep peer support going once the programme ends.
Where it comes from
Reg Revans trained as a physicist in Cambridge, then worked in education. From the late 1940s he encouraged coal-pit managers to meet in small groups and question each other about their problems. In the 1960s he worked with London hospitals. He summed up the idea as L = P + Q: learning needs both programmed knowledge (P, what experts already know) and questioning insight (Q, fresh questions about a problem you don’t yet understand). His books set out the method (see his 1982 collection). Later writers, such as Marquardt, McGill and Beaty, turned it into the structured sets used today, often with a rule that the group asks only questions at first.
The evidence