Tuesday, January 13, 2009

Lots To Talk About

Recently, various people have asked me -- in person or by email -- about what to include or not include in an interview talk (or talks) in the specific case in which you have many possible things to talk about.

Even if your only talk-worthy research experience to date has been your graduate work, your research may have been sufficiently complex and multi-faceted that you are faced with a major decision about what to talk about. Things get even more interesting when you have had various research experiences (e.g., grad school + postdoc(s)) and have an even bigger decision about which research projects to talk about in an interview presentation. Perhaps it is obvious that one research project is the #1 best topic for a particular job, but what if it's not obvious? How do you decide what to talk about?

Although having various projects might make for a more complicated decision, having these options can result in a very interesting talk. I worked on very different things for my postdoc and graduate research, and I found that the interview talk that worked best for me (and was most fun to give) involved bits of both projects.

I was discussing this recently with a former student who has an upcoming interview at Awesome University and who had to make a similar decision about what to include in the interview talks. She will be giving two talks as part of the interview, and was considering talking about Project A in one talk and Project B in another talk.

My advice, which may or may not have been helpful or even good, is not to divide the research so neatly. I think it would be much more interesting to do a bit of integration of the commonalities of the two projects. Perhaps I am biased because I found that this worked well for me, but I think it will also be effective in this case.

In the case of one talk and two projects, you can do a bit of integration, then focus specifically on the project that is most relevant and/or cool.

If you have two talks and at least two projects, Talk 1 can still mostly be Project A and Talk 2 can mostly be Project B -- it is important to give a coherent talk, after all -- but if you can successfully integrate some elements of different projects in a Big Picture kind of way, then you show your audience (and the hiring committee) that you are driven by first-order questions and can see broad connections among topics and methods.

If you have two talks and one project.. that's tricky, but perhaps one of your talks will be a general talk and one will be more specific, so you can talk about your one project in different levels of detail.

Being able to integrate components of different projects is a useful skill in general, and is an approach you can take throughout your career when giving invited talks about topics that may touch on various projects with which you are involved. It can be difficult to do, but that's just part of the fun.


Anonymous said...

Regardless, it is very important not to try to impress during your job talk with the sheer quantity of work you have performed. In a 45-60 minute job seminar, it is only possible to make two or three main points that the average audience member is going to take home. If you overwhelm the audience with material they will (1) take home nothing, (2) hate your fucking guts.

Anonymous said...

I also have some interviews where I have two talks coming up. One is a departmental colloquium and the other to a smaller research group. My thought was to give a very broad talk to the full departments, as they are quite diverse, and only briefly touch on my work and even then to focus on placing it in context.

At PDF-U I have attended many job talks of both flavors. I can see the pressure to present a big result to the whole department but it seems more important, in that setting, to demonstrate that I can communicate to everyone and that I can be a good department member. No one ever seemed happy when the colloquia are too technical.

Am I crazy?

Anonymous said...

You also want to consider your audience. I had one version of my job talk that I gave to broader audiences (e.g. a department with few people in my field and/or a talk attended by many undergrads or people from other departments) and one that I gave to more specialized audiences (e.g. a department where I knew there were many people doing similar work to mine). Of course there were many similarities between the talks, but the emphasis was somewhat different and this seemed to work well for a range of talk scenarios.

Anonymous said...

P.S. Amen to PhysioProf! A candidate who thinks his/her work is so important that he/she can't abridge or summarize even the most minor bits of it is a MAJOR turnoff. It's okay, people, we realize you have done many wonderful things in your illustrious postdoc careers. Show us some skill at selectivity. If necessary, you can always have a slide titled "The Myriad Other Fascinating Things I Have Worked On" and invite people to ask you about them later.

Josie said...

Is it acceptable practice to ask the department that has invited you what topic(s) they would like you to talk about - particularly when you have many possible projects/side projects to choose between?

Anonymous said...

It's acceptable, sure, at least in my own social science department. Let it be either the department chair or the search committee chair whose advice you take. But after you read the job ad, read the university's mission statement, the goals of the university's current president, etc., you may not need to ask.

Anonymous said...

In my job talk I started by saying "the kind of questions I study are ..." and gave a few examples (each with a little illustration on the slide show and a reference to my paper about that question).
Then, having shown that I have done lots of stuff, I said that in the talk today I will focus on question XXX as an example of the approach I take and the results it can yield.
35 minutes later, in the conclusions and future directions, I went back to the big picture and other questions that I want to pursue.

It worked for me :-)

Global Girl said...

I just gave a job talk. I decided in a completely different way. (Significant here is that it was for a technical consulting position rather than for academia.) I wanted to convey as my main point that I could select, summarize and explain technicalities without jargon rather than any of the research results. (They can see my greatness from my resume already. What they can't see there is my communication skills.) In fact, I'm pretty sure any interest in my research was fleeting and nearly purely a function of how entertaining I was to listen to. Therefore, I simply looked for a nugget of research that has fairly clear boundaries (unless one is in my field) and that doesn't require tons of background explanations. That was all. No one looked asleep nor bored, and the questions showed that they had listened and thought while listening. I have no idea if this would have the same effect in academia, probably not. I wasn't trying to convince anyone that I'm supercreative in research.