next »

[abedelia]

I'm almost 3 years into a postdoc in public health and going on the job market this summer/fall.
I have a strong publication record and I'm a co-investigator on a major grant.

My PhD thesis work represents a very coherent story about a related set of research questions on a single disease system. My postdoc on the other hand, has been a mishmash of small, unrelated projects that all draw in one way or another on my expertise and interests but to an outsider would seem very fragmented. I have no good excuse for this other than my postdoc has been a terrible experience (overbearing, micromanaging supervisor who has me jumping from project to project at his whim coupled with my own naivete and lack of backbone).

Despite this, my CV looks good (I've managed to get some papers out of my postdoc, although I don't have a lot of enthusiasm for them) and I've landed an interview at a great school. I am totally lost when it comes to putting together my job talk. If I talk about my PhD thesis work, it will be, I think, a well-received talk, but I can't pretend like I haven't been doing anything for the last 3 years. If I talk about my postdoc research, it will be fragmented and thin. I could rework the format of my talk and take it from a careful narrative about two major studies (what my PhD thesis talk was about) to a higher-level overview of all the research I've done in my PhD and postdoc. That might sound more impressive but it would be even more fragmented and I am not sure I could find a strong common thread to tie it together with.

I'm seriously considering a second postdoc to try to get back on track and onto a more meaningful research trajectory before I go for a faculty position but for now I need to give this interview all I can. How can I sort this out?

Thanks for any advice.

[cranefly]

What do you plan to work on as a faculty member--your PhD-related work or post-doc related work? That is what you should talk about. Bring in the big questions that you are looking at, and present some of the ways you have approached those questions, and add in some future directions.

[anon99]

As cranefly said find a single unifying theme for your previous work that relates to what you want to do.  You may have to go really big picture.  If you did your PhD on reeds and your postdoc on weaving, basketweaving might be a unifying theme.

[lab_gal]

The advice I've been given (that worked well for me) is to tell the strongest (most convincing) story you can, even if that's as simple as "Baskets are cool". A job talk is not about convincing the committee that you're qualified to do research (your CV and letters demonstrate that)... it's about convincing the committee that they want you as a colleague: that you are well spoken, persuasive, insightful, can think of your feet, answer questions, and (most importantly) likely to gain independent funding.

I'm a young investigator, but I've given a lot of talks and just gave a few job talks. FWIW, here's my advice...

You don't need to integrate everything. You just want to be convincing about what you do choose to talk about. I think it would look strange to not include some things from your postdoc, but don't worry about including it all as a part of the hypothesis. You can incorporate bits and pieces of your prior work without needing to include every single part.

I suffered from a similar problem, though in a slightly different order. My PhD was very fragmented with three nearly independent projects. My postdoc dealt with some impressive preliminary research but still needed a bit more time to get major results. I thought I was going to bomb the talk. I wrote the first draft as if I was writing a grant, trying to convince someone that my planned approach was a good one. Only AFTER that, I thought "now where can I throw in stuff I've already done". I structured it so that the first 20 slides told an introductory story all revolving around seemingly unrelated methods in my field of interest.

Let's say I do basketweaving. I introduced myself and gave everyday examples of why baskets are useful. Then I talked about how to construct them (PhD paper #1), what they look like from a distance (#2), and different weaving techniques (#3). Even though 1, 2, and 3 dealt with baskets that were completely unrelated to what I was proposing, everything was a way of saying "isn't basketweaving cool". I did NOT go into details. I gave figures from my published papers and spoke in general terms (literally, "I used XYZ technique to understand ABC phenomenon"). Then I gave the motivation for the research problem, some preliminary postdoc data, and finally a (thoroughly hypothesis driven) explanation for what I would do in the position. I was told by two separate people in two separate job talks that it was "The best talk I've ever seen. Ever."

My point being, if you can point out the relevance and interest in techniques that are fragmentally or peripherally related to what you want to do from the perspective of the big picture, people are going to be impressed. You don't need (or want) to include everything you've ever done. If you give the same level of detail you would give a normal research talk (to your lab group), they're going to be bored. It will be more impressive to give a cohesive talk that demonstrates why your approach (i.e., your scientific style from a broad perspective) is exciting than to give one that is a resume of qualifications.

Be certain to reference your own papers. Mention verbally that you are a "Co-investigator on a grant that addresses exactly this issue". Etc. But I would suggest you keep the talk as cohesive as you can, even if that means backing up and looking at it from a less detailed perspective than you are used to.

Hope that helps. Again, this is just the approach I took - there are many ways to structure a job talk, and it tends to be field-specific (I'm in STEM).

[merinoblue]

I like the advice here. Based on what I've seen, I agree with lab_gal.  I've seen two job talks where the candidates floundered because they presented their current research, which they didn't know well. When it came to answering questions, they fell apart. In the same competition, the strongest job talks I saw were those where the candidates presented their dissertation research. They knew it well and it showed. I don't think it mattered to us how recent or "old" it was.

[onthefringe]

What kind of place are you interviewing at? Where I am (STEM, R1), we want to hear a well-crafted, interesting, exciting story that leads directly into the research you plan to do here. I would be very surprised to see a talk that consisted only or mostly of research from someone's PhD rather than their postdoc, unless it was clear from their talk, letters, etc, that they would be able to pull together a fundable project based on their PhD work. In my field of STEM (Biomedical research) we are very "what have you done for me lately". We recently had a job talk where someone referenced a paper he coauthored as an undergrad, and despite the fact that it was a Cell paper and relevant to his current work, many people scoffed a bit.

In contrast, my husband works in a similar field but at a SLAC. For various reasons, he ended up going back to his thesis field to develop more undergraduate-friendly research projects. His job talk related entirely to his dissertation, with an explanation of why he intended to go back to that research area.

Can you pull together one of your dissertation related projects and tie it through to one of your published postdoc projects to at least make it look like you have a trajectory?

[87735501111]

In my social science field, it is common to just present one strong unpublished paper, and add a slide or two at the end that indicate your future direction and goals. Some of the introduction could talk about your broader research, but even for a 1.5 hour talk, it would probably be too much to detail everything you are working on.

However, the "present the big picture" type talk also sounds like a good format, and I suspect people would like it. The advantage of presenting one paper would be that you can go into detail with the technique, but I could see the big picture style working well if you were coming in with say, a different set of methods from most of the faculty or with a different subfield area.

[lab_gal]

In my social science field, it is common to just present one strong unpublished paper, and add a slide or two at the end that indicate your future direction and goals. Some of the introduction could talk about your broader research, but even for a 1.5 hour talk, it would probably be too much to detail everything you are working on.

However, the "present the big picture" type talk also sounds like a good format, and I suspect people would like it. The advantage of presenting one paper would be that you can go into detail with the technique, but I could see the big picture style working well if you were coming in with say, a different set of methods from most of the faculty or with a different subfield area.

That's a very good point. I suppose the best approach in any case is to think about your target audience. (I rarely present to people who care about the details, since I work in a scientifically obscure field). Take a step back then and think about what would impress your likely audience - if it is an audience of your immediate colleagues, perhaps more detail is needed.

By the way, you can always ask who your audience will be... people from the department, elsewhere, etc

Regardless, I think "don't try to incorporate everything" is usually appropriate. Even when going into great detail, think carefully about how to pick the important details and leave out the boring stuff (or stuff they can ask you about after your talk).