Tuesday, February 12, 2013

Not a job talk

I'm giving a talk at University E later this week. Since University E is weak in my particular sub field, I've been asked to talk in the general seminar. This means that I should expect to give a talk to faculty in the audience who would not have had to take the same quals as me had we gone to the same graduate school.

This alone would not be a problem. I have few aspects of my work that have some very pretty pictures associated to them. Usually, the audience coming to a talk with a given advertised abstract consist of people who are at least vaguely interested in the subject matter, so I can assume some amount of common background. After all, it's not really worth several research hours on my end to write the talk, or the combined several research hours of the audience to have me talk if NO ONE is interested in my research. I try to make the talk pretty, and then talk to one or two people in the audience.

But a divergence in interest is not the only complicating factor. University E may have a position opening up soon. I have many many many mixed feelings about this, but there's no point in dwelling on it if there isn't actually a job to be had. It is well known to the department in University E that I would interested in interviewing for this position should it exist. I've been told (as if I needed reminding) that I should treat this seminar presentation as a job talk. I should expect a lot of people in the audience from the department's strong sub-field, who have little interest in my research. I have  no idea how to proceed.

In grad school, I heard a lot of advice about not trying to be something I am not. I should not try to talk about things that I am not an expert on. If I am an expert in the construction of rope bridges over windy ravines, I should not try to sell myself as an expert raft maker when talking to a department of shipwrights, even though many of the tools are similar to those of raft makers.

This should be a point for me to shine in my own subfield. But I'm terrified of giving a talk that falls flat  to most of the audience. I am not talking to one or two people in the audience. I'm trying to impress everyone.

I spent all day yesterday trying to write this talk, and succeeding in making one slide (beyond the title). Any advice would be appreciated.


  1. With the usual disclaimers, my advice would be that you maybe try and think of it as the first 'overview' lecture in a high level course for bright students who are taking the course as an elective. Your goal is to explain why what you do is cool, identify commonalities with other areas (e.g. how building bridges requires rope making, perhaps noting a reference or two to work by rope makers who are raft builders which have informed your work in bridge building at least as a footnote on the slide, or through your choice of illustrative figure), and give a case study or two of how you've taken on and answered a challenging or novel question (built around the question and some cool pictures, and having a slide showing the funding source and the papers from it so far - a sneaky way to 'sell' yourself as someone who can deliver the metrics). You want people to enjoy the talk, and to go away at least thinking that you would be an engaging teaching and are passionate about your work, which is kind of cool even if they wouldn't work on it themselves... NOT, as with a normal sub-field talk, demonstrate your mastery of the subtlties or your clever twist on a standard thing (because that would go over the heads of the audeience and bore them).

    Myself, I usually add a few specialist slides - data, mathematics etc. - after this sort of talk. Then when someone asks a more detailed question, I can quickly pull up the slide (and I usually prepare a few sentences about each that I've practiced or noted, so that I can talk about it efficiently) and show that I DO have the data/maths etc. in my repetoire, I just designed my talk for a specialist audience...

    Since my department includes people trained in all STEM disciplines, as well as social science and humanities backgrounds, I have to give this sort of talk every 2-3 years whenever my turn comes round in the internal seminar series, so I have a bit of experience...

  2. Well, damn. I wish I'd asked you this before submitting an abstract on a very specific recent project. This sounds like very good advice, and something I could do if I hadn't been asked to talk about recent work. Still, the idea of approaching this as the first day of an advanced level course in the subject matter is a useful guide. Thanks.

  3. After JaneB's great advice, I'm not sure I can add anything else! My first thought after reading the post was to think about a colloquium style talk, where you emphasize the importance of your sub-field and your particular work in regards to the rest of the field. The first third should be very basic, the last third you should state your work as well as you can, even if that means that people will get lost. You can always focus on a particular example and hand-wave what the big picture is... Good luck!

  4. My advice is to simplify, and simplify again, at least for the first half of the talk. I used to worry about being patronizing to the audience until I realized that this never actually happens: a more colloquium type audience *always* responds best to talks where they were really guided through in a very clear and simple way. And, as JaneB says, try to connect, and then reconnect, with why this research is interesting/cool/matters even outside your niche. Good luck!

  5. The common thread I've found from successful job talks is that they tell a story. So, start big picture with the larger question/problem that is easy for a general audience to digest and then use your research to explain how you tackle that question/problem. And I agree with anon - speak simplify it whereever you can.