Tuesday, December 04, 2018

Well, Yeah...

I was surprised to see the headline “Why Teaching Engineering Costs More than Teaching English,” but not because the content was surprising.  I was surprised that it was news.

The recent piece summarizes a report from the National Bureau of Economic Research.  It makes the point that classes in some fields are more expensive to run than others, with the more STEM or vocational classes generally running more expensive (with the notable exception of math).  

Faithful readers may remember this paragraph from a post this summer entitled “Things That Seem Obvious:”

“Hard” vocational programs are more expensive to run than “soft” academic ones.  The least expensive classes to run are the ones that can run well with thirty students per section, and without any specialized equipment.  That tends to describe the Intro to Psychs of the world. Hands-on classes in vocational areas require more equipment, more people to tend the equipment, and more instructors per student.  In practice, we engage in cross-subsidy, with the profits generated by, say, History offsetting some of the losses generated by, say, Nursing. This matters because many outsiders assume that if we could just drop the “ivory tower” stuff and focus entirely on job readiness, the budget would balance.  In fact, we’d go bankrupt. If you want to remake community colleges as entirely vocational, be prepared to pony up more money. A lot more.

That was in response to a question on Twitter about things that are obvious to people in a field, but not obvious to those outside it.  

Context matters.  The NBER paper refers to different salary levels by field, which really doesn’t apply here.  I’m in a collective bargaining setting in which the computer science faculty and the philosophy faculty get the same starting salaries.  (They vary with seniority and rank, but those are also independent of discipline.) So it isn’t a matter of engineers coming in at six figures while humanists come in at half that.  That’s not it. It’s subtler than that.

In addition to the factors I listed last summer -- class size, equipment, staff to maintain the equipment -- I’d add relative availability of adjuncts.  Generally, it’s easier to find adjuncts in history or sociology than in computer science or engineering, particularly during the day. That matters when we have to allocate new full-time positions.  The enrollment crunch has made new hires scarce; we need to deploy them where they make the most difference. All else being equal, that means allocating them to the fields in which substitutes are hardest to find.

Over time, that leads to higher labor costs in the specialized fields relative to the gen eds, even when starting salaries are the same.  A department in which 75% of sections are taught by full-time faculty will cost more than one in which only 40% are. Compound that with smaller classes in hands-on areas, and the cost gap gets even worse.

This may all seem wonky, but it has implications externally.  Angry calls for colleges to tie themselves more closely to the job market are often based on the false premise that doing so would lower costs.  In fact, it would increase them significantly. Sages on stages are cost-effective; guides on sides in well-equipped labs cost a lot more. There’s a valid argument to be had about how much we should increase higher ed funding, and to what end.  But to have that debate, we have to know some basics.

Lectures are cheap; labs are expensive.  I don’t consider that news, exactly, but if putting it in headlines helps people connect the dots, I’m all for it.  Now, about that funding...