A graduate student whose (excellent) second-year paper was accepted at a few conferences came to my office last week to ask me how she should prepare her conference presentations. Because I have never given much thought to how I actually do prepare for conference and seminar presentations, I told her I would write a blog post on the topic after thinking about it. So here is my list of tips on how to prepare conference and seminar presentations, in no particular order. I’m sure I’m forgetting many things; please feel free to include your own best tips in the comments section.
(Note: If you are doing theory or your presentation contains some theory, I also suggest reading William Thomson’s A Guide for the Young Economist, which offers good if dated advice for budding economic theorists.)
- Whatever you do, make sure you know exactly how long you will have to present, and prepare accordingly. There is nothing worse than showing up for a talk expecting to have the usual 75 minutes only to learn that the norm at that institution is to have a 45-minute talk followed by a 15-minute Q&A. In this case, I think the old rule of thumb of one slide per minute applies. Though this does not mean that a 75-minute seminar needs 75 slides (most of my seminar presentations have fewer than 60 slides), it really does mean that a 15-minute conference presentation should have no more than 15 slides.
- What your presentation should include is really a function of time. For example, when presenting my work with Tara Steinmetz and Lindsey Novak on female genital mutilation (FGM) in a seminar, I go into how there are four types of FGM, present a diagram that shows the area excised under each type and what each type looks after healing, I discuss the physiological and psychological consequences of FGM in depth, etc. But when I presented at the CSAE conference last month, where presenters only have 15 minutes, earlier this month, my motivations occupied three slides, and were essentially “FGM affects 100 million women worldwide and has really bad consequences on their health; trust me on that, okay?” (though I still had five intro slides…) In other words, the less time you have, the more your motivations should be highly concentrated and the quicker they should answer the “Why should we care?” question.
- Should you practice your talk? Absolutely. Practice over and over, and time yourself. The more you advance in your career, the less you’ll have to practice your talks, but as a beginner, you have every interest in practicing under a time constraint. As I mentioned when giving advice on the job market, I never practice my talks and I have done reasonably well following this foolhardy strategy, but this does not mean you should be equally foolhardy.
- One thing graduate students consistently get wrong in presentations is the level of technique. Sure, you just spent the last few years doing almost nothing but learning highly mathematical concepts and methods, and you want to show off, Don’t. At an economics conference, most of your audience will have been there and done that, and your display of technical ability impresses no one, really. You can never go wrong assuming you are presenting to an audience of smart college graduates with no experience in your field. This means you should emphasize the motivations and intuition, and define technical concepts in plain English.
- Inevitably, you’ll have to get technical and lose some people when you present your theory or empirical framework or identification strategy. That’s okay, as long as you bring them back at the end in your conclusions, and as long as you try to explain your theory, empirical framework, or identification in plain English as you go through your more technical slides.
- Always have an outline slide, unless you use a Beamer theme that shows the outline on top and highlights which section you’re in, as in this case. It comforts your audience in that they know where you are taking them with this presentation.
- On a related note, always provide a preview of your results. This isn’t a murder mystery: it’s only when people know where you’re taking them that they can enjoy the scenery along the way.
- I am a big fan of using LaTex and Beamer for presentations. Almost every computer in the world can read .pdf documents and has working “Ctrl” and “L” keys. PowerPoint, however, will sometimes crash on you, or it will not display on a PC the equations that looked so pretty on your Mac, and among economists, I suspect Prezi is interpreted as a sufficient statistic for one’s lack of content. Plus, LaTex does math beautifully. PowerPoint
does it horriblynot so much.
- For your introduction, use Keith Head’s introduction formula: Hook (titillate your audience with a strong start or broad motivation), Research Question, Antecedents (the four or five studies closest to yours), Value Added (what you are bringing to the table relative to those previous studies), and Roadmap (which is really your outline slide).
- Never, ever have a literature review in your slides. If literature reviews are boring to read in papers, they are insanely boring to listen to during presentations.
- After your introduction, present your theoretical framework, empirical framework, data, results, limitations, and conclusion. Again, depending on how much time you have, you might want to maintain some of those steps to a minimum. One trick that few people seem to know about when presenting is the Magic Appendix Trick: You can have 15 slides for your conference presentation, followed by 30 appendix (i.e., not part of the main attraction) slides which you can resort to if people ask to see them. This is a good place to put descriptive statistics, robustness checks, proofs of propositions, additional graphs, etc.
- If you have a theoretical model in the context of an empirical paper, unless your theory is your main contribution, it might be sufficient to just present your assumptions and testable predictions, and have your full-blown model in your appendix.
- As above, so below, and so your presentation should follow the order in which you discuss things in your paper. It’s also perfectly fine to self-plagiarize here and cut and paste whole sentences from your paper. Writing is rewriting, and hopefully by now your paper is beautifully written. There is no use reinventing the wheel at this stage.
- If you can tell your story with a graph or picture, do so. My two papers which were the most successful in seminars are my aforementioned paper on FGM with Steinmetz and Novak and my forthcoming paper on food prices and food riots. In both cases, the paper contains a graph that essentially tells you the whole story in one simple, self-explanatory picture. Even when presenting to the smartest people in the world, a picture is really worth a thousand words.
- Tables of empirical results should focus on your coefficient(s) of interest. This means that you should have a line in there that says “Control Variables? Yes” for those cases where you do include controls, “Village Fixed Effects? Yes” for those cases where you do include village fixed effects, and so on. Again, stick the full-blown results in the appendix, and present only the results that are the most relevant for your talk as part of the main attraction. See slides 22 to 24 of my CSAE presentation.
- Do not read your slides. Do not learn them by heart. Keep the tone of your presentation conversational. Abstain from making jokes: as a grad student, you want to signal that you have competently investigated an important question and provided a technically sound answer to it. Keep your jokes for when you are a senior scholar in your field.
- Development students: Though you are undoubtedly proud of the fact that you’ve done fieldwork, but unless a picture you took while in the field is absolutely necessary for your audience to understand a point you’re making, avoid fieldwork pictures in your presentations (doubly so for pictures with smiling developing-country children, which are incredibly cliché…)
- Likewise, maps have become pervasive in econ talks these past few years. If you are exploiting some spatial source of variation, go ahead and include a map. But if you’re just including a map because you think your audience won’t know where The Gambia is, put it in the appendix.
- Similarly, I always thought it was a bit odd when people added a last, one-word slide that either said “Questions?” or “Thank You!” Audiences are generally not shy about asking questions, and when they are, they’ll find you after you’re done. As for thanking people, I find that the best thing you can do is thank people for their time and attention either at the beginning or at the end of your presentation.
- I have never had to present a poster at a conference, but here is a list of tips that strike me as sensible. Perhaps the smartest idea for posters is to print your poster on location: since most conferences are held in college towns, you’ll easily find a copy shop where you can print your poster. That way, you will avoid having to travel with your poster, and risking your poster case getting crushed by that inevitable guy on your flight who tries to jam-pack his enormous-size “carry-on” luggage in the overhead compartment by pushing as hard as he can on everything else inside.
- When questions arise, answer them to the best of your ability. If you don’t know, say that you don’t know. If your answer is tentative, explain that your answer is off the top of your head. If a question is obviously of little interest to most people, or takes you too far afield, politely offer to discuss it with the person who asked after the talk.
- Above all, have fun. Giving talks is the most effective way to communicate your excitement about your research. If you are not having fun, chances are people in the audience aren’t either, and if they aren’t having fun, they cannot get excited about your research, which means that your impact will be much more limited.