Popular Posts

Caveat Emptor

The opinions expressed on this page are mine alone. Any similarities to the views of my employer are completely coincidental.

Monday 16 March 2020

Making an online lecture

I'm very lucky to be on sabbatical leave and living until September (hopefully) in a country that has taken a different approach to the UK to dealing with the Corona virus epidemic. That means that I've not been forced suddenly into teaching online, for which I'm very grateful. A couple of years ago though I did decide to create some online video content for one of my courses and I'll share here what I learned about what to do and what not to do.

I should start by saying that my ambitions were very limited. I wanted to produce something that was serviceable, but I had (and have) neither the skill nor the patience to uphold glossy production values. 

A little context. The original stimulus was a larger than average cohort of MSc students taking a compulsory Research Design course timetabled to take 3 hours. The format of the course was normally a 1 hour - one to many lecture - followed by two seminar sections of 12-15 students. However with 40 students each seminar section would have had 20 students which was a bit too large so I was faced with a dilemma. I didn't want to add an extra hour to my teaching load so I decided to do away with the live lecture & replace it with a video lecture  thus freeing up an hour for an extra seminar.

I started off with a few fancy but either impractical or unrealizable (by me) ideas about what I wanted to produce. The first thing I dropped was the idea of  doing a talking head to camera lecture. After a short trial I found that my webcam and the lighting in my office were not good enough to produce decent results. Particularly trying was finding a decent background to film myself against. Stripy curtains, which was basically all I could use without radically rearranging my office furniture, do not do a lot for a foregrounded talking head. Eventually I went for something much simpler - basically my power-point slides with a recorded commentary saved as a video and uploaded to my YouTube channel. 

My first efforts were made using the mic from a Microsoft headset. The sound quality was passable (just) but not brilliant. Later I  bought a decent desktop microphone and I have to say it made a big difference to the sound quality (compare the last 4 recordings with all the rest). After doing what everyone does -  read the Amazon reviews - I bought a Blue Yeti USB mic and I've been very satisfied with it.

Mastering the recording features of Powerpoint is pretty straightforward. An hour or two of trial and error is sufficient to learn all you need to know to do a basic job. What took me more time was figuring out the best workflow process to produce 50 minutes of content.

It's easiest to say what didn't work (for me). My naive first thought was that I would just "talk through" my slides much as I would do if I were giving a live lecture. Very quickly I discovered that what I  achieved by doing this was...awful. Ums, errs, verbal ticks, extended pauses while  I composed my thoughts, not to mention weird lip smacking noises and loud booms when I scratched my nose made the whole thing sound even more horribly amateurish than it actually was. 

I don't know whether this will work for everyone but what worked for me was to ditch the ideal of sparkling spontaneity and write a script. Of course this makes the whole thing a bit more theatrical (reading it out in a monotone is not going to improve things much) but it helped me to impose a bit more structure & discipline on the production process.

I normally give extempore lectures  so my first challenge was to write everything out in a style that wouldn't sound wooden when spoken to the mic. Having already produced the slides helped a lot as they functioned as a kind of story board. Early on I decided to chop up each lecture into 10-15 sections for uploading. That also helped with the structuring of the material and the delivery. Once I had my script I then did a few tests to get an impression of timing and the right pace of delivery. My normal pace of delivery is quite slow. I can get away with that when doing a live performance, but in a recording a slow pace very quickly becomes  boring. I listened carefully to the pace at which talk is delivered on Radio 4 and concluded it was at least twice as quick as I normally talk. 

Even after producing the script I still screwed up. I wasted a bit of time experimenting with  an autocue program that scrolled through the text at the bottom of the screen. This really didn't work for me and I found that I needed to see what was coming next in order to make transitions between slides relatively seamless.

In the end I went for a low tech solution: a printed script, double spaced, in 20 point Garamond with slide transitions noted in bold red. Each slide corresponded to a 2-3 minute gobbet of audio. I tended to record in short sections and rehearse each section before recording in order to iron out anything that seemed awkward in the transition from writing to the spoken word.

The rest is just perspiration rather than inspiration. I frequently rerecorded sections I had fluffed and made all the usual beginners mistakes - recording level too loud, recording level too low, recording when there is bound to be a lot of ambient noise etc.

It's a time consuming job to get even passable results. After I had written the script - which could easily take me 8-12 hours (and I already had the slides prepared) it would take 3 hours to record 50 minutes of material plus the time to turn the Powerpoints into a videos & upload them to YouTube. 

Once done though I actually thought it was worthwhile. Wasting time on 1 to many live lectures is silly. Putting that part of teaching online makes a lot of sense (as long as the material remains in the control of the producer). In the second iteration of my course with video lectures, and a reduced cohort size, I used the time that I formerly had devoted to delivering the lecture to answering student questions about the lecture material.  That means more face-to-face time focusing on the content that students actually want to talk about and more time for them to think about what they want to ask.