Tuesday, September 10, 2013

Impedance mismatch: are adjuncts better teachers?

I've long argued to my physics colleagues that adjunct professors, selected mainly on the basis of teaching ability, will do as well teaching introductory material as high powered researchers who (often) view their teaching assignment as a burdensome load (as in "teaching load"). It's also true that communication becomes more difficult as the brainpower gap between professor and student increases. (Some wags doubt whether two individuals who differ by, say, > 2 SD in intelligence can really understand each other ...)

Even if you find the results of the paper below unconvincing (perhaps the effect is too small), they nevertheless did not find that adjuncts are worse lecturers than tenure stream faculty. Yes, former and present colleagues, there is an arb to be had here: equal (or better!) quality teaching for students at lower cost, freeing up researchers in the department to produce more scientific knowledge.
The Atlantic: We all know the stereotype about tenured college professors: great researchers, lazy teachers. After all, you don't get tenure by dazzling 18-year-olds with PowerPoints. You do it by convincing other academics you're a genius in your field who's going to bring boatloads of grant money and prestige to campus. And nobody ever won a grant by grading papers.

A gross oversimplification? Of course. But there might also be a hint of truth in the caricature, at least judging by a new study from Northwestern University. The paper--co-authored by university president Morton Schapiro, professor David Figlio, and consultant Kevin Soter of The Greatest Good--finds that faculty who aren't on the tenure-track appear to do a better job than their tenured/tenure-track peers when it comes to teaching freshmen undergraduates.

Previous studies have suggested that colleges tend to hurt their graduation rates by hiring more part-time and non-tenure faculty. But Shapiro and his team wanted to measure the impact of tenure on "genuine student learning," a notoriously tricky task. So how'd they do it? Using the transcripts of Northwestern freshmen from 2001 through 2008, the research team focused on two factors: inspiration and preparation.

To start, the team asked if taking a class from a tenure or tenure-track professor in their first term later made students more likely to pursue additional courses in that field. So, to borrow their example, if an undergrad took economics 101 from an adjunct, and political science 101 from a tenured professor, were they any more likely to sign up for additional poli sci classes. That's the inspiration part. Second, the researchers wanted to know if students who took their first course in a field from a tenure or tenure-track professor got better grades when they pursued more advanced coursework. So, if our hypothetical student took more classes in both economics and poli sci, what did they fare better in? That's the preparation part.

Turns out, tenured and tenure-track professors underperformed on both the inspiration and preparation fronts. Controlling for certain student characteristics, freshmen were actually about 7 percent more likely to take a second course in a given field if their first class was taught by an adjunct or non-tenure professor. They also tended to get higher grades in those future courses. Taking an intro class with a non-tenure track instructor increased a student's mark in their second class by between .06* and .12 grade points, depending on controls. The freshmen who got the biggest boost tended to be less academically qualified students, judged by SAT scores and such, in the hardest subjects.   [Italics mine.]

As the study notes, these patterns held "for all subjects, regardless of grading standards or the qualifications of the students the subjects attracted..." In other words, the non-tenure-track faculty bested their more established colleagues every from English to Engineering.
The paper (NBER):
Are Tenure Track Professors Better Teachers?

This study makes use of detailed student-level data from eight cohorts of first-year students at Northwestern University to investigate the relative effects of tenure track/tenured versus non-tenure line faculty on student learning. We focus on classes taken during a student’s first term at Northwestern, and employ a unique identification strategy in which we control for both student-level fixed effects and next-class-taken fixed effects to measure the degree to which non-tenure line faculty contribute more or less to lasting student learning than do other faculty. We find consistent evidence that students learn relatively more from non-tenure line professors in their introductory courses. These differences are present across a wide variety of subject areas, and are particularly pronounced for Northwestern’s average students and less-qualified students.
This kind of study isn't really using Big Data by scientific standards, but it's an example of "evidence-based" research on higher education that is now made possible by electronic storage of student records and cheap computation. See also Data Mining the University: College GPA Predictions from SAT Scores and Nonlinear Psychometric Thresholds for Physics and Mathematics. The hiding places for "faith-based" theories of education are slowly disappearing -- at least for the numerate. When I discussed the use of adjuncts in past department meetings, I was confidently told by colleagues (who had not looked at any relevant data) that REAL! research physicists did a much better job in the classroom than mere adjuncts. I am dubious of this because many excellent researchers are lousy teachers and either lack the self-awareness to realize it, or just refuse to admit it.


rz said...

Based on personal experience I'd even suggest that up to a very high level (say, first-year grad, QM, E&M and Mechanics) instructors who are specialized in teaching do better at it. Of course, by that stage a student that hasn't learned how to self-teach is in trouble.

Douglas Knight said...

arb...freeing up researchers in the department to produce more scientific knowledge

But can the department actually capture the money from this arb? The department doesn't want to displace teacher-researchers by adjuncts. Maybe in the short term the rules are set and the department benefits, but do they trust the administration not to change the accounting? Now that you're an administrator, maybe you can either capture the arb yourself or make sure there are long term incentives so the department definitely benefits.

steve hsu said...

The incentive landscape is complex and varies from university to university. But, just to take a simple example, a chair could convert one or a few tenure stream FTEs into adjunct salaries and meaningfully reduce the teaching "load" on active researchers in the department. In my experience, reduced teaching is very attractive to good researchers, and would aid recruitment. If, for example, you could give every active researcher a semester off every other year that would be very enticing.

Douglas Knight said...

Yes, that's exactly the kind of deal that would be good if the department could trust the administrators in the long term.

To put it another way, say the deal is to trade in the tenure lines for adjunct positions plus a slush fund equal to the saved salary (maybe to be used to buy out teaching). The department doesn't have much in the way of property rights to that slush fund and is afraid it will be seized, particularly when the dean changes.

But if it were not a slush fund, but rather a rewritten contract with each member of the department specifying the reduced teaching load, it would be easier to make the Pareto improvement permanent (though by no means certain).

Of course, the dean may not trust the chair with a slush fund and may prefer
to barter reduced teaching for adjuncts, so it may all work out, but there are other things that the department could do with the money, avenues that probably won't be considered because of lack of property rights.

Some departments may not trust the administration not to expropriate them even in the short term. They may be afraid that if they try to negotiate such a change or even admit the possibility that adjuncts are good teachers, that the positions will be changed and the saved money sent elsewhere. Knee-jerk reactions sound to me like fear that factual claims can be used against them.

steve hsu said...

Yes, the department has to trust that the administration wants to improve research productivity in the department, attract better faculty, etc. and is not unduly sensitive (for whatever reason) to the use of adjuncts.

The reactions of my colleagues to my adjunct proposals were sincere -- not based on meta-strategic considerations. I think they honestly had the wrong beliefs regarding relative quality of instruction.

Diogenes said...

"there is an arb to be had here"

really? no!

my best instructors i always went to bed with. they were text books.

DK said...


"Teaching" in universities is researchers' con of the unsuspecting general public to gain cushy permanent jobs for themselves. Bait and switch, basically.

Diogenes said...

and it's bad for everyone. bad for students, bad for researchers, bad for the economy. the model is left over from a time when books were chained to their shelves. with video and sound there's no excuse. steve and his colleagues are criminals pure and simple.

Jess Riedel said...

> When I discussed the use of adjuncts in past department meetings, I was confidently told by colleagues (who had not looked at any relevant data) that REAL! research physicists did a much better job in the classroom than mere adjuncts. I am dubious of this because many excellent researchers are lousy teachers and either lack the self-awareness to realize it, or just refuse to admit it.

I had the same beliefs as your colleagues, and my personal experience is that my best (and most inspirational) teachers were always researchers. But upon reflection, my worst teachers were researchers too.

Two competing hypotheses: (A) adjuncts are all-around better teachers (presumably with the caveat that we aren't talking about graduate-level classes that strain their expertise). (B) The dominant issue is "IQ-differential", as Steve alludes to at the beginning, and that what course is being taught or what the teacher's career track is are not as important when you control for IQ. These would be distinguished by looking at which type of teacher is better at teaching very-high-IQ students.

Steve, who do you think would be better at teaching basic classical mechanics to future high energy theorist, assuming the prevailing attitude of researchers?

Anthony_A said...

Did they control for age of instructor? In general, my best instructors at college were the younger ones. Is there a differential age or experience structure between tenured and adjunct faculty?

David Coughlin said...

When you are teaching standard courses, courses that might transfer to any college or university anywhere without deep consideration, using standard textbooks, then it should be enough to have teachers who are capable of teaching to minimums. These are the 1n1, and probably the 2n1 courses in physics. The peril isn't so much, I think, that the management deans will try to recover the semester's worth of cash from the department, rather than allowing the extra course release. It is that they will try to commoditize the whole curriculum.

I took the second semester of quantum mechanics twice. It is required for graduation. The first time was taught by a hardcore theorist. I was screwing around at the start of the semester and skipped the first two weeks, and I never caught up. The class started with 28 people and ended with 7. I got a D. The second time I took it, it was taught by a collider physics professor. I pretty much only went to class to turn in homework and take tests, I got an A. I would not trade the experience of that first class for anything because it was so intellectually stimulating, but that is no way to approach intro level pedagogy [at least at a large, public university].

steve hsu said...

Teaching future theorists, or even teaching students that are likely to go on to obtain PhDs, is quite different from teaching premeds, or engineers, or physics 101 to humanities and social science students. Of course I, personally, would rather have a brainy professor with a really deep knowledge of the material. But on the other hand you run the risk of some really bad, unprofessional teaching from an eccentric genius. I've experienced the whole range of possibilities ...

I agree with you that the variance is highest among the researchers. The brainpower thing plays a role, and the potential for aspie inability to connect with students or understand what they are confused about. I think the main issue is that researchers have not (primarily) been selected for their teaching ability, whereas adjuncts or permanent lecturers could and should be.

It is interesting that the weaker students in the Northwestern study got the most benefit from non-tenure stream instructors. This is what I would have predicted.

I like David's story below about his QM class. It certainly rings true to me. No offense to experimenters, but it is an usually a big mistake to have a serious QM course taught by a non-theorist.

oregonlocal said...

"Teaching future theorists, or even teaching students that are likely to go on to obtain PhDs, is quite different from teaching engineers"

Some of the smartest people in the world are engineers. As an example I give you Archimedes.

MUltan said...

Or how about this: universities could actually start giving tenure to good teachers rather than trying to shortchange them. Teaching is, after all, what the students pay you obscene amounts of money for. Everything else that schools do is just overhead as far as the undergrads are concerned. If this fraud and parasitism of administrators and research professors on the students isn't corrected, a more efficient method of measuring knowledge than the diploma will soon come around and eat the universities' lunch.

Blog Archive