Much rejoicing last Thursday when Science Minister Kirsty Duncan announced that the federal government was re-instating the funding for the Universities and Colleges Academic Staff System (UCASS), which was last run in 2011. But what caught most people’s attention was the coda to the announcement, which said that Statistics Canada was going to “test the feasibility” of expanding the survey to include “part-time and public college staff” (the “C” in UCASS stands for colleges in the Trinity College sense, not the community college sense, so despite the name public colleges have never been in the survey).
What to make of this? It seems that by “part-time” Statscan meant sessionals/adjuncts/ contract faculty. That’s a bit off because every university I know of makes a very sharp distinction between “part-time” (many of whom are tenured) and “sessionals”. It make one worry that Statistics Canada doesn’t understand universities well enough to use the correct terminology, which in turn bodes ill for their future negotiations with universities around definitions.
Because let’s be clear about this: universities will do almost anything they can to throw sand in the gears on this. They do not want data on sessionals in the public eye, period. Oh sure, in public the Presidents will welcome transparency, evidence-based decision-making, etc. But institutional research shops – the ones who will actually be dealing with Statscan on this file – are Olympic champions in shutting down government attempts to liberate information. In fact, that’s arguably their main purpose. They won’t actually say no to anything – they’ll just argue relentlessly about definitions until Statscan agrees to a reduced program of data collection. Statscan knows this is coming – they have apparently allocated four years (!!!) for negotiations with institutions, but the safest guess is that this simply isn’t going to happen.
And to be fair to universities, the kind of data UCASS would provide about sessionals would be pretty useless – a lot of work for almost nothing. UCASS can count individuals, and track their average salaries. But average salary data would be useless: it would conflate people teaching one course with people teaching five. And since UCASS had no way to track workload (you’d actually need to blow up the survey and start again if you wanted to get at workload, and as interesting as that might be, good luck getting universities to green-light it), the data is meaningless. Knowing the number of sessionals tells you nothing about what proportion of undergraduates are being taught by sessionals. Are 200 sessionals teaching one course each worse than 100 teaching two courses apiece? Of course not. But if raw numbers are the only thing on offer then we’ll ascribe meanings to them where they arguably shouldn’t exist.
You see, “sessionals” are not really a single phenomenon. Many are professionals who have full-time jobs and like teaching a class on a side, and they’re usually a huge boon to a department (especially in professional faculties like law, nursing an business) because they help expose students to a life beyond academia. Others are PhD students teaching a class while another professor is away – and thus learning valuable skills. The “bad” sessionals – the one people claim to want to stamp out – are the ones who have a PhD, are teaching multiple classes the way professors do. I suspect this is a pretty small percentage of total sessionals, but we don’t know for sure. And adding sessionals to UCASS won’t get us any closer to finding out because even if they wanted to, universities couldn’t collect data on which of their employees have other full-time jobs outside the institution.
Despite all the kumbayahs on Tuesday about how this UCASS expansion is going to promote “evidence-based decision-making”, I’m genuinely having trouble imagining a single policy problem where data from UCASS would make a difference. Universities already know how many sessionals they employ and whether numbers are going up or down; UCASS might let them know how many sessionals other universities employ but frankly who cares? It’s not going to make a difference to policy at an institutional level.
If you really wanted to know something about sessionals, you’d probably start with requiring institutions simply to provide contact information for every individual with teaching responsibilities who is not tenure-track, along with the amount paid to them in the previous academic year (note: Statscan couldn’t do this because it would never use the Stats Act to compel data this way. Provincial governments could do so, however). Then you’d do a survey of the instructors themselves – number of classes taught, other jobs they have, income from other jobs, etc. Now I know some of you are going to say; didn’t some folks at OISE do that just recently? Well, almost. Yes, they administered almost this kind of survey, but because they weren’t drawing their sample from an administrative database, there’s no way to tell how representative their sample is and hence how accurate their results are. Which is kind of important.
So, anyways, two cheers for the return of UCASS. More data is better than less data. But the effort to get data on sessionals seems like a lot of work for very little practical return even if universities can be brought round to co-operate.