Wednesday, March 25, 2015

Why giving a chalk talk is such a dreaded academic past time

The academic job search can be a challenging time in one's career, but it is also a great time to really reflect on where you see your research in the next 5-10 years. One of the biggest excitements comes when you actually land your first interview. With the interview ahead of you, you begin seriously toying with the possibility of joining this institution. In fact, your odds of actually getting the job just went up tremendously, so pat yourself on the back! But, of course then comes actually prepping for the interview. This shift from cloud in the sky to a real possibility is a pretty big shift in the job application process. Most of the parts of this process are pretty well covered by others here, here and here and include advice you may be more familiar with if you've been in academics for some time. You've likely seen job seminars and met with scientists in one-on-one fashion before. For the most part, you've even given a dissertation seminar so you know what to expect from most of the individual pieces.

Though, the one piece of the interview process that is most likely to be completely foreign to you is the chalk talk. There are several reasons for this, but for the most part the reason is that typically only faculty are invited to attend these talks, and also not all departments require a chalk talk in their searches. I too found myself in a pretty scary place when, in addition to going on my first job interview - a collectively foreign process - I was faced with giving a chalk talk - a pretty foreign concept. Even my PhD advisor's experience with the chalk talk was pretty limited and so I found myself consulting the web and asking friends recently off the market. Unfortunately, I found this level of preparation completely insufficient for the first time I had to give a chalk talk. In fact, my first chalk talk was the worst talk I have ever given and was a complete embarrassment to myself as a scientist. I don't say this for pity, but to emphasize how little you can ever prepare for something like this. Not surprisingly, I didn't get a job offer after my first interview and went on the market a second year. Knowing how terrified I was to ever do this again, I was determined to do better the next time. So, soon after my interviews were over, I asked the faculty to attend a chalk talk in my department of a job candidate. As it turned out, the chalk talk I attended was given by an already established assistant professor, but it was a pretty good model for many of the insufficiencies in my own chalk talk. So, I put it out of my mind until the next job cycle came back around and I was invited for another interview. This time, I was determined to identify all the holes in what I did wrong and I am happy to say that the chalk talk I gave that ultimately landed me an offer went amazingly well. But it was a process and it certainly did not start out that way!

Here are a few reasons I think I didn't do so well and the things I learned in my preparation that made a world of difference in my delivery.

1. Don't draw parallels to other experiences - chalk talks are in their own category!

While for the most part you haven't been told much about chalk talks, you will undoubtedly start researching and talk to others, and start to form an idea of what it's like. You may even try to relate it to a similar experience you've had in your academic career. I was told that it is a test of "How well you think on your feet" and that it should mostly be off the cuff rather than a prepared talk. So, I related it to my preliminary exams in graduate school. You may also relate it to a committee meeting. For me, drawing parallels was a horrible mistake for several reasons, most importantly due to the differences in the audience and expectations. A chalk talk audience can range from the entire department to only the search committee, but often you don't know ahead of time who will be there, so you just need to be prepared for anything. The expectations for a chalk talk are also very different than other talks. You are expected to give a broad overview of your research program and each individual project you will discuss. This means that there will be at least 10-15 minutes in the beginning where you should be doing the bulk of the talking. Of course, you should assume that no one has ever heard of you, read your CV or seen your earlier job talk. Despite what you may have been told, you can practice and even rehearse this part. For the most part, you can also rehearse your explanation of the big picture and major motivation behind each project you present. Do not under any circumstances try to 'wing' these parts of a chalk talk. Remember, this exercise is also an evaluation of how well you will teach, so imagine you are explaining your work to a student in your office. The 'think on your feet' parts come after you are done with the overview, and the motivation/big picture for your first project. But even then, you don't just open it up for Q/A like at the end of a seminar. In fact, in the off chance you have a pretty quiet audience, you should be prepared to talk the entire time if needed. Then let them interrupt you in a more natural way. Of course, you can use the transition between topics as a chance to open it up for questions just in case you aren't getting too many.

2. With mixed expectations from different faculty, you're very unlikely to please everyone!

It can really be hard to go into a talk assuming that several in your audience are almost certain to be disappointed in your performance, but this is the case with the chalk talk, especially for interdisciplinary searches. Most of the folks I interact with in my postdoc come from more traditional math/physics/statistics backgrounds. They thrive in a chalk talk because they are used to being at the board for general conversations, but if you are a biologist this may not be the case. So faculty with different backgrounds will have different expectations. Then there are the old school folks who may have a very different opinion of what it takes than a newer faculty who may not even have had to do one in the past. In fact, many search committees now-a-days encourage candidates to use powerpoint, but most advice suggests NEVER to do this no matter what the search chair tells you. I think a big part of this is to temper the differences in expectations between young and more established faculty members. I am sure this is the result of seeing many more of these types of talks over the years which alter how high your expectations might be. It's the same for seminars. When you were an undergrad attending conference talks, they probably all seemed amazing, novel and insightful. Now that you are a seasoned scientist, your expectations are very different and you might come away from a conference thinking that everyone is still presenting the same work they talked about 5 years ago. Perspective makes a huge difference.

I think the most important thing to realize is that you are reaching a stage in your career where you need to convey that you are an independent scientist. You are not seeking the approval of your graduate committee or your advisor. You are interviewing to be their colleague and as such you need to show that you have opinions and ideas that are going to make you an important part of their department. Of course, just because pleasing everyone isn't necessarily your goal, offending anyone is a huge mistake to make! You want to avoid anything that might suggest that you would be a hostile colleague or someone who is going to work in a bubble.

3. It is easier to criticize ideas than results

Part of the reason that you are not likely to please everyone is that it is much easier to criticize an unfunded idea than a published paper. The ideas you are presenting typically have not been validated. You likely haven't even submitted any of them for funding or gotten much feedback before now. This is where giving a practice talk can be very helpful (see below). You want to vet your idea to an audience prior to being in a high stress job interview situation. It may also help to go ahead and write up a 1-4 page sample grant proposal. This is a useful exercise and the writing will absolutely come in handy when you go to write the grant anyway. Perhaps this can even be turned into a young investigator award application. Chances are you have a starting point in your research statement, so go from there and develop your ideas into at the very least a specific aims page.

If you are short on preparation time, consider using notecards. The first time, I didn't have the notecards and found it very hard to use a 3 page grant proposal during the middle of a talk. Instead, a combined approach worked best for me. I had a manila folder laid out in front of me with notecards on each project and my overview. I also had a notecard on funding sources and on answers to expected questions from many of the links above. I also had the full write up for each of three projects in the folder as more of a crutch. I didn't refer to it, but it helped to know that if I got any hard questions, I could refer to it.

4. Practice makes perfect - conquer your fears in a safe environment!

If you can, you should absolutely practice your chalk talk before doing the real thing. The practice talk should be done in person, with at least 5-10 people, and at least half of them should be faculty. While a fellow postdoc or grad student can provide great feedback on projects and ideas, they may give you a false sense of security based on their minimal expectations of a chalk talk. Alternatively, a faculty member in the same talk may have a very different interpretation of your performance and likelihood to get the job. Of course, since different profs will have different expectations, you should have a least a few there (see #2)!

If it was your first time ever giving a talk at a meeting would you ever consider doing it without giving a practice talk? Then, you should consider the same is true for a chalk talk. A practice will also help change your expectations and identify places where you are less knowledgeable on your project. It may even help you develop your ideas further as you talk through it with others. If time is a factor, I had my practice chalk talk the Thursday before my Tuesday talk (though I left for my interview Sunday AM). I sent the email Monday PM to nine people in my department (6 faculty, 2 research scientists and 1 postdoc) and 5 people showed up (3 faculty and 2 research scientists). So, in relatively little time I was able to assemble a small group of very helpful people to make me feel 1000% more confident when I had to do another chalk talk!

In addition to giving a practice talk, if you can I would strongly encourage attending a chalk talk in your department before you ever have to give one. This may be a year before you want to go on the market. While these may only be open to faculty, they almost always make exceptions for postdocs on the market. That also means that there may be a postdoc or two at your chalk talk there looking for pointers.


  1. I gave a chalk talk at the Broad Institute on genomics of butterfly mimicry. I decided this was impossible, so resorted to pasting pictures of the butterflies up rather than trying to draw them. It went down just fine, I think. Remember, visuals are important, even in chalk-talks!

  2. Good point - I was encouraged to draw the fish I talk about in my chalk talk. Though, sometimes that may not be easy for intricate graphics such as range maps, etc. The advice I've gotten regarding this is to print out a LARGE image to show, even if it spans multiple sheets and you tape them together (think old school dot matrix printers!). I also like your idea of pasting it to the board itself! I have been encouraged to avoid handouts at all costs, mainly because it pulls the audience's attention away from you and may lead to more questions from folks paying less attention as they look over a handout.