Vision / Society and communities / Covid-19 and lessons for teaching and learning

Society and communities

Covid-19 and lessons for teaching and learning

The Covid-19 crisis has offered us the long-overdue opportunity to radically re-imagine the university education experience.

It would be a great shame if we emerged the other side of lockdown/social distancing with nothing more for our efforts than a slightly higher definition webcam and a marginally better quality microphone to record our traditionally delivered lectures.

With a grateful acknowledgement to inspiration from the Institute of Physics network on physics teaching and learning led by Helen Vaughan at the University of Liverpool, here are six key changes I’d like to try see implemented beyond the standard model:

Replace lectures with tailored online tutorials

I am firmly of the opinion that recording/streaming traditional lectures online is a massive waste of time, effort, and opportunity. There is a universe of difference in the ‘projection’ and enunciation required to reach those semi-detached (at best) students at the back of a packed 300-seater ‘real world’ lecture theatre as compared to speaking directly to a learner on the other side of a laptop screen. The idea of recording oneself speaking to an empty lecture theatre or seminar room - as I’ve heard suggested on more than one occasion over the last few weeks - and then uploading that footage for the class seems to me to be utterly self-defeating.

We have the opportunity to present the material directly to the student on the other side of the internet connection. So let’s embrace that opportunity. Instead of projecting to a class, focus on talking to that individual student. (This is one key reason why the Sixty Symbols series of videos - a collaboration between Nottingham Physics and Astronomy and Brady Haran - is successful. We talk directly to the viewer; many of us look straight down the barrel of the camera lens when presenting so as to better connect with the audience.)

In other words, think of each video as a one-to-one tutorial rather than a lecture. (We also have to carefully consider our video production values. Our students are paying a great deal of money for the education we’re providing and footage recorded in murky, underlit and under-financed WobbleVision with audio of 1 kHz bandwidth is not going to cut it anymore.) There is, of course, room for a small number of real-world lectures to enthuse and ‘orient’ the class but let’s leave behind the outdated notion that course material needs to be delivered at a fixed time, in a fixed place, with a fixed mindset.

Engage and interact

Having freed up the traditional lecture slots so as to ‘deliver’ the material via tutorial videos, we can then directly engage with students through problems classes and interactive sessions. There is no easier mode of teaching than traditional lecturing ( - if you don’t believe me, try talking to a secondary school teacher) and I realise that any increase in interactivity is going to mean we have to work harder. But we can be smart about this. We can make extensive use of multiple choice questions, online quizzes, simulations, and peer-to-peer instruction and discussion. Will some students dislike this mode of learning as compared to traditional lecturing? Yes, of course, because they, like us, will have to work harder. But then deep learning is never easy.

There’s nothing special about a 50-minute ‘quantum’

We are too often locked into the idea that our course material has to be delivered in chunks that are just shy of an hour in length. Why not 20 minutes, or 10 minutes, or, indeed, 5 minutes, where necessary? After all, a great deal of the education literature highlights that we’re on a losing wicket with regard to keeping students’ attention after the first 15 to 20 minutes of a lecture in any case. (And that’s even if we’ve managed to avoid being timetabled for the graveyard shift on a Friday evening…)


So many research, teaching, and/or management strategy meetings/ ‘away days’ I’ve attended have involved breakout sessions, where a sub-group nominates a representative to feed back their thoughts and ideas to the main meeting. We have incorporated this type of strategy in the final year of our MSci Physics degrees at Nottingham, which are assessed via groupwork, coursework, presentations - the students give the bulk of the lectures in many of the modules - and projects (with, for the majority of the class, no examinations.) But there is a great deal of scope for embedding this strategy in earlier years of the degree and via the type of online breakout sessions

After all, there is nothing that quite focuses the mind as realising that at the end of the breakout discussion you will have to accurately represent the ideas and solutions of your colleagues/classmates.

Stop assessment by the book

Having traditional bookwork (BW) exam questions in an online, socially-distanced setting in which students can access textbook answers (from their own lecture notes or other sources) is, to put it mildly, very problematic. That type of BW question is often nothing more than an exercise in memorisation and regurgitation (especially if it involves the reproduction of a derivation that the student barely understands but which they can reproduce line for line). Covid-19 means that we have got to tip the assessment balance towards the application, rather than the demonstration, of knowledge.

And, yes, I know full well that moving to problem-based assessments involving no bookwork element will almost certainly cause the average module mark to plummet. But that’s still no excuse.

Building lab skills, virtually

My first major teaching role was to revamp our second year undergraduate course so as to incorporate a much higher level of computer-based measurement and interfacing. We didn’t go far enough with the revamp for my liking. The vast majority of experimental physics (indeed, experimental science) now involves analysing, visualising, and fitting digitally acquired data from a variety of sources. There is thus immense scope to have students work with simulated data or, indeed, real world data acquired from a variety of sources without the necessity of their ‘turning the dials’ themselves.

I realise that this may already sound like the worst type of heresy to some of my experimentalist colleagues - how could we think of having a lab course where the students don’t get their ‘hands dirty’ with real world kit right from the start? - but I’ll go even further. I think that we should start by teaching experimental techniques via simulation - including, in particular, getting the students to code virtual instruments - before they’re let anywhere near the real-world lab.

Experimental science is messy. Much better - and much less frustrating - for students to work with simulations whose parameters they can tweak, and learn what to expect when everything is working correctly, than to sit them in the lab in front of a piece of kit with a multitude of dials that they dare not breathe on in case they screw up the experiment.

The same is true of all the various measurement techniques used in 21st century labs. A student’s insight and intuition can be developed much more systematically in the virtual world before we let them loose on the real thing.

After the virus

Read more from our researchers as we reflect on the challenges we face after the coronavirus crisis, as well as opportunities to rebuild a more resilient, fairer society.

Philip Moriarty

Philip Moriarty is a Professor of Physics. This is an edited version of a blog that first appeared on his website Symptoms of the Universe.

More impact case studies

A year on: working class women and work during the Covid-19 pandemic

Read More

An academic working in government: reflections from a Chief Scientific Adviser

Read More

Cities and communities after the pandemic

Read More