What it takes to get tenure: ambiguity of the teaching criterion


Getting tenure at a teaching university might be harder than getting tenure at a research institution.

If you don’t like that concept, then try this similar concept on: what you need to do to get tenure at a teaching-centered institution is far more ambiguous than what you need to do at a research university. One could argue that it’s easier to get tenure if you know specifically what you need to do. At most teaching schools, exactly what you need to do to get tenure is vague at best.

In one, you need to convince the faculty and administration of a teaching university that you are excellent at teaching. In the other, you need to convince the faculty and administration of a research university that you are excellent at research.

At research institutions, when you interview for a job, it is typically spelled out exactly what you need to do to get tenure: grants, publications, and train doctoral students. At most places, you’re given a neighborhood of a dollar amount, or a certain set of grant agencies and number of grants you need, and a number of publications in journals with a certain tier, as first and senior author. There may be some subtleties, but when you’re coming up for tenure, it’s clear based on the numbers whether you’re approaching that threshold, and you should be well aware if you are shy of the mark or have well exceeded it. If you’re marginal, then you know that you’re marginal.

The notion that teaching counts in hiring and tenure decisions at research universities is a sham, as recently pointed out by Alex Bond at The Lab and Field. If you’re at a research institution, being a horrible teacher won’t hurt your chances at tenure and being a fantabulous teacher won’t help your bid for tenure. (If you are unliked or extremely popular, however, and your case is marginal, then teaching performance could be inserted as a surrogate variable to help swing the review one way or another.)

At teaching institutions, the story is entirely different.

At your on-site job interview, I wish you luck trying to get a wholly quantitative description about what it takes to get tenure. Typically, you need to be “excellent” at teaching, and “excellent” at either research or service, and mighty good at the third. I think that’s the answer I got at every single one of the 10 or so teaching campuses where I’ve interviewed over the years.

You know how teaching doesn’t matter at research schools? Well, the converse isn’t entirely true. Research does matter at teaching schools, though there may be a lot of flexibility about what counts as research. At lower-ranked institutions, “research” might not necessarily involve publications or external funding, if you really like someone’s teaching. It could just involve keeping students busy in your lab outside of the curriculum and having some of them get into grad school.

Some teaching campuses put specific numbers on publications, which in my experience has ranged between 0-6, with no real specification of impact factor. The expected publication rate before tenure is negatively associated with teaching load, but this relationship has only a moderate correlation. Most places expect you to submit a grant but aren’t horribly put out if you aren’t funded. The research criterion is pretty clear-cut at teaching campuses, and there is also fudge room because it’s not the primary criterion.

Then, what constitutes excellent teaching? Most campuses go with a Supreme Court Justice Potter Stewart kind of definition.

Knowing excellent teaching when you see it isn’t a good way to decide whether someone gets tenure, is it?

How do most people judge excellent teaching, when they are required to make such a judgment? A student of human nature would suggest that it is identified by how much it resembles the practices of the observer. I’ve never met a full professor who didn’t have a moderately high opinion of their own teaching. And we’ve all met plenty of full professors who couldn’t teach their way out of a bag. This does not bode well for effective tenure decisionmaking. (By the way, is the Bush neologism ‘decider’ now a replacement for ‘decision-maker’?)

In practice, there are many factors that are included in the quantitative and qualitative measures of teaching performance at a teaching campus.

The drawback to all of these quantitative and qualitative measures is that they all suck, or at least have poor resolution.

Let’s go over them one at a time. Keep in mind that no school uses all of these measures in concert.

Teaching evaluation forms There is a whole subfield in education research on this topic, and I’m not going to let this site devolve into a bitch session about teaching evaluations. Really horrible instructors get horrible eval scores, and amazingly perfect instructors get high scores. What happens for most of us, the professors in the middle — ranging between not-so-good and run-of-the-mill excellent — is really murky.

At my university the forms are called PTEs: Perceived Teaching Effectiveness evaluations. The key word here is “perceived.” Are students good at knowing whether their instructor is effective? Often, yes. However, there are a huge number of systematic biases that go along with these forms, suggesting that we need to avoid using the numbers in a comparative fashion. Upper division courses have higher scores than lower division courses, which have higher scores than non-majors courses. This might be independent of teaching effectiveness. There are age and gender biases that affect student perceptions of effectiveness, and associations between the grades received by the students and the perceived effectiveness of the instructor are not necessarily causative. How you dress in the first weeks of class can really matter, too. From discipline to discipline, mean evaluation scores are quite variable. If you want to measure improvement in the same course, with the same professor, with the same student demographic (including time of day the course is taught), then this might be a good measure, at a coarse resolution.

If your tenure case is being evaluated at the level of the college or the university, and your scores are being compared against colleagues in other disciplines, or who teach different kinds of courses, that isn’t fair. I don’t know of a campus that specifies a specific threshold score for evaluations (at least officially), and that is a good thing. However, unofficially some campuses or committees are expecting scores to be above a certain level. If that’s the case, then faculty need to learn the little tricks to make sure they don’t do things that cause students to lower their scores. (That’s a whole other set of posts.)

Written remarks by students The voluntary responses by students on evaluation forms are potentially telling. Students can offer specific and useful praise, and also tell damning stories that very clearly can explain instructor performance. Recurring similar comments by multiple students are particularly valuable. However, most student responses are idiosyncratic and it’s very difficult to distinguish between a student with a legitimate grievance and one who is bitter about their own performance.

Classroom observations Faculty members in the department may be requested or required to sit in on a certain number of hours or lessons before offering a recommendation. These observations are effective so long as the observer is capable of identifying effective instruction. This is heavily subjected to the biases of the observer, especially as scientists typically have no training in teaching methods. For example, when I was a junior faculty member, I made sure to implement the methods of active learning in science instruction that I learned as a graduate student in the College Teaching Certification program and as a Preparing Future Faculty fellow. So what happened when I was observed by my senior departmental colleagues sizing me up for tenure? I’ll always remember this, word for word: “You need to be less Socratic and lecture more. You should be using powerpoint and use more detailed information.” Never mind the fact that all of the current research on science education told me to do the opposite of what they said. After all, these professors were the ones evaluating my tenure file. So, when they were in my classroom, I had to lecture, even though I knew this was an ineffective approach.

How could classroom observations be effective? The people doing the observing could know what they hell they are doing and could be well trained in evaluating effective teaching. This happens in public schools. In the state of California, to be come a fully credentialed K-12 teacher you need to go through an evaluative induction process, the Beginning Teacher Support and Assessment (BTSA, pronounced “bitsa.”) To be a BTSA evaluator, you need to be trained to observe and score the performance of teachers, and this training process involves a calibration of standards and a long list of specific criteria. One BTSA evaluator observing one set of instruction comes up with a score very similar to any other BTSA evaluator; that’s the way the system is built.

What about teaching-centered universities, how do senior faculty do their observations? They show up, if they care to spare the time, and they then may fill out a cursory form if one exists, and then include whatever observations they choose to include or not in their letters. I can’t think of an evaluation that is more subjective nor disconnected from whatever objective measure that could exist. (I’m not saying that I’m any less guilty than anybody else, mind you. Of course, most faculty would be peeved at the notion that they need to be trained to recognize good teaching.) Regardless, in some teaching schools, classroom observations aren’t a required or even optional component of the tenure portfolio. Oftentimes, the only thing that tenure committees know specifically about what happens in the classroom is by hearsay from students.

I was impressed that once, my all-time favorite dean chose to sit in on my classroom for half an hour, and when he wrote the letter for my file he referenced specifics from what he saw in my classroom. He didn’t do this for lack of being busy, and I appreciate the time he spent in directly evaluating me.

Assessment data Perhaps we could look at student performance using assessment data, looking at student knowledge before and after individuals pass through your course. These kind of assessment data aren’t common, and anyway, most science faculty are in full rebellion against regional accreditation agencies that are requiring assessment in curriculum design, and using assessment data like this could actually annoy some faculty members who might think that you’ve gone over to the dark side of assessment. I suppose you could use these numbers but just not call it assessment and maybe get away with it.

Student letters I think few campuses do this, but it happens in my undergraduate institution. I was asked by my college to write letters of evaluation for faculty members in whose courses I was enrolled. The college requests letters from some students who are listed by the faculty member, and also randomly (or perhaps haphazardly) selects other students from rosters of recent courses. I imagine that these letters would be a lot more informative than whatever would be in student evaluations. They do this for both tenure and promotion to full professor.

Hearsay It is stunning how students are willing to discuss my colleagues in front of me, as long as I’m not involved in the conversation. Just the other day, I was in my lab sorting ants, and some of my research students were going on and on in great detail about an instructor in our department, who is a close colleague of mine. There was a mix of criticism and praise. They were talking like movie reviewers or restaurant critics. I wasn’t involved in the conversation, but I was sitting easily within ear’s reach where they were saying all kinds of things about my colleague that they would never say directly to this person. This kind of overheard conversation happens all the time, especially if you’re teaching lab sections. It’s unprofessional of the students to do this in front of other professors, but I guess they’re not professional. I arguably have more indirect information about my colleagues’ teaching from this route than any other. If I believe most of what I hear, by the way, then most people in my department are incredibly awesome. Regardless, this isn’t a valid source of information for evaluating teaching performance, though I imagine that in some environments this is probably the source of information with the greatest sway.

External evaluations Research universities require external letters from experts in the subfield of the tenure candidate to evaluate their tenure file. So, teaching universities must get outside experts to evaluate the teaching of candidates in their subfields of expertise, right? Ha! That’s a good one!

What it takes to be “excellent” at teaching is being able to convince the other faculty in department that you fit that label. Faculty use a variety of information sources, including not not limited to the information above. Ultimately, the assessment is a holistic gestalt-based system. Kind of like how honey bee colonies use guard bees size up the pheromonal composition of bees landing at the nest to decide if they belong, academic departments work the same way. If you don’t fit in, then the guard bee will pounce on you.

The biggest way to not fit in is to not teach well.

However, another way is to teach well, but teach differently.

It’s often said that tenure is about “fit.” Some people say that’s vague: how do you define fit? It’s nothing that needs any special definition. Either you fit in or you don’t. Either you have the same values and the same approaches with respect to education, or you don’t.

This is why it’s sometimes hard to get tenure in a contentious department (read: snakepit) in a teaching institution. Even if you’re careful to not take sides in any weird departmental politics, everyone involved in the tenure process will be called upon to assess your teaching. This is going to involve a meeting where your teaching is discussed. If the department has divisions about teaching philosophy or approaches, this will emerge in the criteria used for evaluations. If one side really likes what you do and explains why, then the other side might end up in disagreement. This is not good. You can ameliorate this by how you sell your teaching approach in your tenure file. You don’t want to make the mistake of arguing that you have worked hard to find the most effective approaches to teaching and that your assessments show that students learn effectively. What you want to do is communicate that your teaching fits in with your department, and that you have worked collaboratively with your colleagues so that you have learned how to teach well from them. You don’t want to say anything that is overtly contrasting existing practice, because, ultimately, the people in charge of deciding whether your teaching is excellent will compare your work with the template of their own work. Just like guard honeybees that use their own smell to decide whether to reject outsiders.

Even if you have a history of demonstrating teaching excellence at a teaching institution, a fresh pair of eyes with a different perspective, or a different agenda, could look at the same record and come up with a credible argument that the record fails to demonstrate excellence. Without anything changing, the environment can shift so that what is perceived as “excellent” in one year might not be acceptable the next year.

This is different than research institutions, I think. It’s harder to argue against grant dollars and a list of publications on a CV. You could argue that the journals aren’t of a high enough impact or that the grants are from the wrong agency, but the research bar at a research institution is far, far more tangible than the teaching bar at a teaching institution.

I would guess that if you are unambiguously above the bar that’s been set for you for research productivity and funding, and you haven’t entirely botched something else, you should be golden. Even if there are academic disagreements about your work, if you’ve got the grants and published in the right journals, then that is likely to be fine. This is particularly the case if you’re at a unionized institution, in which the tenure process is more transparent than at an institution with an opaque process with secret information, because the faculty lack the power to make sure that the process is fair.

Of course, at nearly all universities, tenure rates are quite high, except for a few Ivies that have a de facto policy of hiring Assistant Professor positions as glorified 7-year postdocs. When people don’t get tenure, it might be because performance is not up to snuff, but it can also happen because the department is toxic or incompetent. Other crazy stuff can happen, too. Regardless, the lack of specific quantitative criteria in the teaching criterion create an element of hap into the process that makes it less predictable, which makes it a source for anxiety if not a source of difficulty.

In short, the amount of work it takes to be an excellent teacher doesn’t necessarily correspond to the amount of work you have to do to get tenure at a teaching institution. To do that, you (most likely) have to be an excellent teacher and you also have to do the work to convince your colleagues that you are. In some places, this is harder to do than others. In some places, you don’t even have to be an excellent teacher, as long as you are able to create that perception. There’s the rub.

5 thoughts on “What it takes to get tenure: ambiguity of the teaching criterion

  1. To me it seems judging a good researcher vs a good teacher are not as different as you suggest. Good research in many cases is just as subjective as good teaching (really bad research and absolutely amazing research are often easy to identify, but differentiating among those in meaty part of the bell curve is hard). The difference in practice between research and teaching is that we devote a huge amount of time and effort to assessing research quality. Every time you submit a paper or grant you undergo a subjective review. Thus research quality is easier to gauge at the tenure level because your cv is an aggregate of many independent (but still subjective) evaluations. Everyone has at least a handful of anecdotes about how they were unjustly rejected by a biased editor or referee. If we were judged only by a couple paper or grant submissions it would seem just as unfair, subjective, and stochastic. However in research, compared to teaching, assessments are more likely to balance out to provide an accurate picture because you are assessed so many different times.
    I think tenure at a teaching school could be made just as “objective” as at a research school, but it would require putting in the same amount of work as we put into the peer-review process of publications and grants. For evample by having 2-3 independent (not colleagues or close collaborators) reviewers for each class you teach, each semester or year. Over the course of 6 years at a teaching university you would have over 100 independent assessments of your teaching, which would probably provide just as accurate of an impression or your teaching abilities as your publication and grant list would of your research abilities.
    The $65,000 question: is all of this time and energy put into getting a more accurate picture of teaching quality worth it to teaching universities?

  2. These are really good points. I want to think about this a lot more.

    I don’t think research institutions, in practice, seek to tenure “excellent’ researchers. They want to tenure productive researchers. There are venn diagrams that could be drawn (and I’m tempted to do so), but clearly productivity ≠ excellence.

    These universities outsource this evaluation to the research community that decides who gets to publish in which journals and who gets funded.

    Whereas. teaching institutions almost uniformly claim to want to tenure excellent teachers. They don’t have the option of outsourcing this quantitative evaluation to external entities. And nobody, I don’t think, make a reasonable attempt to actually measure this on their own campus, other than the evaluations by students at the end of the semester, and occasional evaluations. There’s a push-and-pull with “freedom to teach how you want” and “learning actually how to teach well” and I think any effort to make sure that pre-tenure faculty get substantive teaching evaluations would be accused as Orwellian.

    I’d be interested in hearing from anybody who comes from a university that really does carefully measure teaching, and actually holds mentors accountable for doing genuine mentorship, including observations. I do have a feeling that nobody’s going to contact me about that, though.

Comments are closed.