If we agree that the current situation is so exceptional that we need to fully monitor all students taking exams (and that's still a big if, I also am a university teacher, we don't use any monitoring system. There are other solutions as well such as taking oral exams...), one could simply hire a bunch of proctors who watch the video feed. Considering the savings that universities have made they could send one to everyone's home even.
> I see it as a currently-necessary annoyance, as the least bad option. The alternatives have greater deficiencies: human invigilators using Zoom etc. don't scale (institution experience); oral exams don't scale (my experience moderating such assessments); no invigilation leads to cheating (by few students but enough to be a real problem, especially for professional qualifications - my direct experience and institution experience); shutting down education until the pandemic's over is unfeasible.
Last year, we tried not monitoring, we tried Zoom invigilation, we tried orals. Lockdown rules prevented us from sending in-person invigilators in one semester (and we couldn't have hired even 10% of what that would take) and the other we invigilated in person like normal. We are spending more than ever before, so there are no savings. All of these methods failed. This year, grudgingly, we have moved to proctoring software.
Regarding scaling of oral exams, there's actually some interesting research/calculations (I try to find the reference later) and the cross-over is somewhere around 150 students when oral exams become slower (I do think this is quite teacher and subject dependent though).
Also about budgets, I believe that your department is spending more than ever, the issue is property services departments of the universities should be saving large amounts. Considering lock-downs and staff and students working from home, maintenance cost should be way down. I suspect though that money lands in completely different buckets (don't get me started on the business of university property services, we pay rent in our overhead cost when we get grants that are higher than renting office space on the main shopping street in our city).
The oral exams I moderated didn't seem like they could scale past 20 students. The reason our online students are online is almost always because they are overseas, and the majority of them have limited English. In that context, orals are discriminatory, stressful, and very slow.
I'm in NZ. Our universities have mostly been operating in person as normal (i.e. costing just as much) with far fewer international students (who bring in money), except in addition we have comparatively small numbers of resource-intensive online students as well, and are attempting to bring parts of entire degree programs online for the first time.
Just because this solves your problem (it clearly doesn't) doesn't mean it's a good idea. Killing half the students also solves your problem.
You need to live in a world where this is not an option, period. What's the next solution then? People forget there are always more manual solutions even if they take more time...
Could you explain further why this did not scale?
I'm imagining an invigilator watching video feeds using an interface similar to, say, that used by security guards to monitor surveillance feeds. I would think that a single human invigilator can monitor more students using this system than an in-person setting. What am I missing?
On top of that, there's always a bunch of annoying mucking about getting set up for Zoom invigilation: angle of camera, light, checking environment, etc. All that needs communication to and fro, and it can take up to 5 minutes for a single student. Now multiply that by say 100 (a bit of parallelism is possible, but individual communication is needed with each student).
We have designed entirely new fully-online interactive courses in response to this.
In other courses, where we have e.g. 640 local students and 10 online, they need to sit the same test for it to be fair.
