I’m a tenured faculty member in a physical science discipline at a primarily undergraduate institution (PUI). These means that my institution has very few graduate students, my department has no graduate students, and my primary duties are teaching undergraduate classes and conducting research that involves undergraduates. That doesn’t mean we don’t do good research–I’ve published in prestigious journals and I have colleagues who have received very large and prestigious grants–but it does mean that we have to sometimes be a little more flexible and clever in how we approach certain things.
My department is wrapping up a tenure-track faculty search. We had an amazingly talented and accomplished pool of people, and it was hard to select among them. Most of the advice that you hear is correct: In order to get a job at a PUI, you need to have a good record of publishing, at some point (either as a grad student or postdoc) you should mentor an undergraduate (preferably more than one) on a project, and you should teach a lot and teach well. “A lot” is discipline-dependent. If you’re in a discipline where everyone spends a lot of time as a TA, then you should also get (at some point) some experience solo-teaching a class as the instructor of record. If you’re in a discipline where most people barely even TA, then a track record of doing more than the minimum of TA work, and getting some sort of accomplishment under your belt (e.g. TA of the Year award, or helping to rewrite part of the lab manual) might be enough.
Here’s one strange thing, though, that I haven’t seen anyone give as advice, but somehow many candidates have been doing it: People are giving research talks that are primarily about their field in general rather than their specific contribution to it. I don’t know why. This has happened for a few searches in a row now. We’ve searched in different subfields, we’ve had different search committee chairs (i.e. the candidates have interacted with different people when planning their interview and getting instructions for their job talk), and we’ve seen this problem in male and female candidates, US-born and foreign-born candidates, pale candidates and candidates of color.
In these problem talks, the vast majority of the time will be spent explaining the background to their project. I’ll pick an example that isn’t related to any of our recent searches, so as to keep this anonymous. Imagine that the candidate were talking about quantum dots, and they spent the entire time telling the audience what a quantum dot is, why quantum dots are important, and what the important questions are, but spent little or no time on what they themselves do in quantum dot research, and what contributions they have made and the techniques that they use. That may be a great overview talk for a department colloquium if you are an established person and you’re just there to enlighten the audience. Go do that…ten years from now. Don’t do it now. When you’re interviewing, you need to tell us what YOU do. I wish I knew who was putting it out there that you need to give that talk now when you’re trying to get hired. DON’T DO THAT!!!
That doesn’t mean you shouldn’t give a lot of general background, especially for an audience dominated by undergraduates. However, if the department give you, say, 50 minutes for a talk and 10 minutes for questions (typical numbers, adjust accordingly if local custom is different), then you should spend maybe 20-25 minutes (give or take a few) explaining the basic concepts underlying quantum dots and demystifying some key abstract concepts that some of the undergrads are not familiar with. Along the way, try to illustrate things with examples from your work. “And here’s a picture of the Quantum Dot-Making Machine Thingie that I have used in my work”, not “Here’s a picture of somebody else’s Quantum Dot-Making Machine Thingie.” After that, you should be telling us about specific things you’ve done and showing us your results. You might have to simplify some of it for the students, and sometimes you might need to go over their heads a bit to reach the faculty, and that’s OK if done in moderation. What’s important is that in the second half of the talk you hit the level of the students when you can but never at the cost of telling the faculty (i.e. the people deciding whether to hire you) what it is that you actually do. We aren’t sitting in this talk to be persuaded that quantum dots are important. We know that quantum dots are important; that’s why we decided to interview a quantum dot expert. We want to know why your work on quantum dots is particularly important.
The other thing we want to know is if your research is feasible with undergrads. That often means it should be (comparatively) inexpensive, but not always. (There are private undergraduate colleges with endowments that exceed the GDP of small island nations.) What’s more important is that you need to make it clear that a lot of the work on your projects can be done by people who only need a short time to get up to speed on some key techniques. Make that very explicit in your written research plan. “Undergraduates could contribute to this project because the Quantum Dot Analysis Machine Thingie is easy to run if you’ve just had a year of chemistry and a physics class on optics.” Or whatever. The more explicit you make that the more likely it is that we’ll interview you.