Interview season is about to begin in earnest for science academic job seekers, both at the faculty and postdoctoral level. One of the most important parts of an interview, is a seminar that the candidate has to deliver to the department or the research group in which (s)he will be working, if (s)he is selected for the position advertised.
By the time the invitation arrives for an on-campus interview, an aspiring probationary faculty has typically already delivered a number of talks at conferences, or in academic science departments or national laboratories. For a postdoctoral applicant, on the other hand, this may well be the first important research seminar given in a non-familiar setting.
Physics graduate students, for example, may have given talks to their own PhD advisor’s research group, or (more seldom) to their own department, maybe presented some short (10 minutes) talks at some APS meeting, as well as, of course, their PhD thesis defense . All of these talks are either fairly short, or before a “friendly” audience.
Many posts have been written on how to give talks. Rather than providing my own contribution to eternal debates, e.g., whether equations should be featured on the slides , on how many slides, whether to use movies or not and so on, I prefer to make a general observation drawn from my personal experience. I regard it as possibly relevant primarily to applicants for postdoctoral positions, who will typically be asked to give a talk about their PhD research work to the group of their potential postdoctoral advisor. As usual, being myself in physics my opinion reflects in part some of the quirks of my field, but I think some of the remarks below might have broader relevance.
It’s All About the Big Picture
The most common mistake that I see made by junior scientists, when delivering a presentation on their own research, is that of failing to place their work in a broad context. They focus their talk on very narrow, technical, experimental or computational details, and no sense is given to the audience as to
a) why that was deemed a worthwhile research subject to begin with
b) what outstanding questions were addressed
c) what new results were obtained.
d) how the work has advanced knowledge of the subject
e) how the work goes beyond, or complements the effort of others.
e) are there any future possible developments ?
This is understandable in part. After all, most of the time spent working on one’s doctoral research work has to do with finding solutions to very specific, technical issues, over which a graduate student may be stuck for months. This type of obstacle can literally become the nightmare, the obsession of a PhD candidate; overcoming it is typically cause for celebration, especially when that paves the way to the completion of the project and of the doctoral dissertation.
However, one ought not lose sight of the fact that that part is of little or no interest to the outside society. We all appreciate that the completion of an original piece of research work entails solving a number of highly nontrivial technical problems, hence our general respect and appreciation for anyone capable of doing it. Still, it is the greater significance of the project and its impact on the field that most of us in the audience would like to hear about .
It is the responsibility of a PhD advisor to ensure that his/her advisees be capable of presenting effectively their research work to the outside world. Knowledge of the literature, of alternative approaches or techniques employed by others, of the limitations and advantages of the technique utilized vis-a-vis other ones, is a must for a freshly minted PhD, who should be regarded as a world expert on the subject of his/her dissertation .
A (nearly) PhD graduate unable to articulate reasonable answers to questions such as the ones listed above (which will inevitably be asked by an audience of non-specialists), will come across as someone who has simply “followed orders”, without even developing a real understanding of the context in which his/her research should be placed.
What is the problem with that, one may ask… After all, Isn’t a postdoc mostly expected to help in the lab, or in any case to bring technical expertise needed to expedite the completion of the various research projects ? Well, that is only true in part.
Sure, a postdoc must be technically competent, but (s)he will also (and arguably primarily) be expected to come up with ideas for projects (not only for him/herself, but also to assign to graduate students) and carry them out independently, suggest new lines of research, contribute to the mentoring of graduates students (beyond helping them with “manual” work) and to the writing of articles and proposals, as well as act as an effective spokesperson for the group, at conferences or in the interaction with other scientists.
A good interview talk is not the one that has the right number of slides, equations, figures and so on, but rather the one which will give the audience the impression that the candidate is ready to take on such a task. I know you candidate really want to tell everyone about the brilliance of that sampling trick that you included in your code, or about that new ingenious way of growing thin films that you have thought of — and make no mistake, those things are important, but your talk should be mostly about vision.
 It never ceases to amaze me that so many soon-to-be Physics PhDs find themselves interviewing for a postdoctoral position having never before given a one hour talk. Their doctoral advisors, as well as their departmental graduate chairs, are seriously at fault.
 I find that debate quite silly. Of course equations should be used, when they are required. Making up slides consisting exclusively of equations is probably a bad idea, but from that to not using equations at all, for fear of scaring off an audience of fellow scientists, there seems to be a leap.
 I say “most” because sitting in the audience there may be a fellow graduate student stuck with the same technical problem — no worries, (s)he knows that the person presenting the work must have faced it and will be sure to ask at the end of the talk.
 Essentially knowledge of all the background information that allows one to write a research article once the data have been produced.