Going from teaching in a physically co-present mode to teaching online is EXTREMELY DIFFICULT FOR THE BEST TEACHERS


This is a short article written to help people in thinking through their choice of techniques for teaching online. It is aimed at people who have extensive experience of in-class teaching. Although I had primary schools in mind when I decided to write this, it applies equally to any context.

This advice is based on research carried out within the “ethnography of expertise and practice” paradigm (Polanyi, 1958; Schön, 1990; Eraut, 1995).

The first point to start with is, as we have observed many times, excellent classroom teachers often don’t quite realise what it is about their practice that makes it so good. Teaching quality depends on what we call “micro-actions” – fast, fleeting and often unconscious habitual patterns of thinking and acting. I find that it’s often the case that the better a teacher is in the classroom (their natural habitat) the harder it is for them to teach online. Their habitual micro-actions don’t straightforwardly transfer to a very different environment. This might be accentuated in primary education, where teachers spend most of their time in their own classrooms.

So we start with that, and ask the question: is excellent classroom teaching easily replicated online? (the answer is currently NO). This is why…

In the physical classroom we can do highly interactive, responsive, personalised, physically present teaching. The teacher leads, and has a working-plan of how the lesson will proceed, which is broken down (often tacitly) into a series of moves. The teacher has what we can call an “expected engagement profile” – that is to say, an often implicit set of expectations about how the students should respond as the lesson proceeds. Are there points at which the students will struggle? This will also be varied between different children. The teacher will have a sense of how each different child will respond and when a child might need extra attention, or when they might respond in a different way. Each move aims to get a response from the students, allowing the teacher to modify their actions based on what they observe as well as explicit responses from the class (such as asking a question). Many of the important cues happen on a fast, small and almost unconscious scale. The teacher is continually looking and listening for non-verbal cues to check engagement and to quickly select their next move. They are often much more mobile than they think, making small or large moves around the classroom so as to get a better view of individuals. This also helps to show support for individuals, giving confidence, and dealing with behavioural issues.

Let’s be completely clear about this: replicating this kind of teaching online is currently impossible. The online environment is nowhere near rich enough to support this kind of interaction (see my recent article in the psychology/philosophy journal Aeon for more on this topic). If the students and the teacher are at the high-end of capability in using a collaboration platform with a rich set of interactions (e..g. Teams plus Vevox) they do an adequate job. But this is rarely the case, and it does depend as much on student skills as teacher skills. Any student without those skills will be significantly disadvantaged.

In my university teaching (at Warwick), I can get closer to this ideal, because I’m teaching some of the most highly qualified, intelligent and capable people in the world. Yet we still find it difficult.

So, our advice is always: don’t try to replicate in-class teaching online. A different approach is required. It’s not perfect. But it’s the best we can do in that environment. It may also have additional benefits – allowing students to develop their own independent study abilities, with less dependency on the teacher.

The core technique for this is to separate out the modes of input that the teacher needs to provide and the types of student interaction necessary, and put each of those modes into the best possible format, removing as far as possible a dependency on immediate hi-res teacher-student interaction (of the kind described above).

So for example, schedule very short sessions in which everyone will be online at the same time, and use simple interactive methods to do instantaneous but meaningful interaction. You might use a set of cards that the students can hold up to instantly give a response. Or small whiteboards. This is faster and more flexible than using chat or emojis (although those methods are improving). Take advantage of the new ways of viewing a whole class on screen, such as Large Gallery or Together Mode in Teams. Don’t try to do too much through these short live sessions, and do record them so that students who miss the session can catch up later (Teams is perfect for this). The main aim is social, motivational and emotional.

Another important technique is to use narrated screen recordings for demonstrating complicated processes (ideal in Maths). That allows the students to watch and rewatch. This might even be better than when teaching live. Combine it, if you have time, with simple online interactive tests – but make sure you can easily see the results, give feedback, and work through them in short videoconference sessions if needed.

There are many ways in which you can present content and activities. But perhaps the key revelation is that, using a good collaboration platform (and Teams is ideal), you can all keep connected while working independently or in small groups. This is how, in the real world, many people work in the digital and creative industries – working on their own work, but at the same time connected in an ongoing “meeting” to support each other practically and emotionally.

Get the students working live on shared documents that you can access. You can add feedback straight away, and also start up audio chat with them whenever needed. We do this a lot in university teaching. Again Teams is ideal for this. You can give each child their own version, or get them working in groups. The new breakout room features are especially good.

Encourage the students to communicate with each other as they work on things independently. They can use text chat, audio or video. Working in breakout rooms, means they have a smaller group to communicate with. You can pop-in to the rooms to see how they are getting on.

That’s just a few ideas to illustrate the possibilities.


Eraut, Michael (1995) “Schön Shock: a case for reframing reflection-in-action?”.

O’Toole, Robert (2021) “Zoom and gloom: how empathy and creativity can re-humanise videoconferencing”.

Polanyi, Michael (1958) Personal Knowledge.

Schön, Donal (1990) Educating the Reflective Practitioner.

Dr Robert O'Toole NTF

Senior Teaching Fellow, Arts Faculty, University of Warwick. Fellow of the Higher Education Academy, National Teaching Fellow, Warwick Award for Teaching Excellence.

You may also like...