What I learned from teaching a seminar class

I have been teaching a seminar course this semester, and through coaching students I think I learned more about what makes a good presentation than by reading “5 useful tips” type of articles or attending any number of  “Improve Your Speaking Skills” classes. By coaching and watching students give research seminars I became acutely aware of common mistakes, some of which I am also guilty of, most of which are easily fixed.

For the record, I do not consider myself to be a remarkable speaker, and I can use improvement. However, I do not think I am a poor one either. These tips are not primarily for you, they are for me. They are basically a collection of dos and don’ts which bubbled up over the past few weeks. Most are obvious and have been talked about by many others. Some are less obvious. As this tips collection is for myself, they are very eclectic, just like your presentation should not be.

1. Know your audience. Who are you going to give a talk to? Then tailor every bit of your presentation to that audience. If you are speaking to a crowd of bioinformaticians, you probably do not need to introduce MrBayes, or what HMM stands for and what it is good for. If you are talking to experimental biologists, then yes, you do.On the other hand, do not dwell on common molecular biology assays. Your goal here is to be clear and engaging. We all know that visiting seminar speaker who is coming from a tangentially-related department (i.e. a biophysics speaker in a molecular biology department), and who fails to engage his audience because the first two equation-filled slides (perfectly acceptable in her department) triggered the audience’s mathophobia.

2. Familiarity breeds comfort. Start with something you are sure most of your audience already knows. Many speakers fear doing that, because, if the audience already knows this stuff, why bother telling them again? Won’t they be bored? Well, if your entire seminar covers what they already know, then the answer would be yes. But if it’s only the first five or 10 minutes, then your audience will love you for placing them on familiar grounds as a staging area for their foray into uncharted (for them) territories.

3. First motivation, then outline. Many speakers start with an outline. Probably not a good idea. The reasons for why you created such an outline would be lost on your audience. Start with a motivation: what is the scientific question you are asking? Why is it important? If you have a brief story that bolsters that motivation then tell it. This could be anything from an interesting clinical case for a health-related seminar to a counter-intuitive or surprising bit of logic for a computational seminar. An initial compelling story with an unresolved problem will create the audience engagement you want. Like any good  campfire storyteller, your goal here is to get your audience to want to hear more. Now that you have hooked them with a motivation and an unresolved story, present your outline.


A laudable goal. You could be more specific though.


4. <cliche>Say what you are going to say, say it, then say what you have just said</cliche>. If you have attended any kind of “improve your public speaking” seminar, this point must have been stated 1,000 times. So for the 1,001st time: yes, do it. Your audience is not familiar with this material as you are. You need to state things several times, without seeming repetitive. This rule can (and should) be used in a nested fashion: apply it to your whole talk, and then apply it to every important sub-section of your talk.

For your entire talk. Say what you are going to say: “Today, I am going to talk about how to survive the zombie apocalypse”.

Say it: (your talk: acquiring food, water, guns, securing the perimeter, etc. etc.)

Then say what you’ve just said: “Therefore, you have to be very determined, have a good aim, travel in groups and don’t make noise at night”.

5. Use new jargon/abbreviations/acronyms/initialisms sparingly. Generally, only use abbreviations if they are (1) common jargon for you and your audience (2) they really are essential to save time. Other than that, avoid them. When introducing new abbreviation/jargon, make sure that you repeat the components and the acronym several times, over several slides. In that way, even your most inattentive audience member should have got it. That being said, do use the jargon common to you and your audience. This will make everyone more comfortable.

6. Explain the graph. When you present a figure with a graph, explain exactly what the X-axis represents, and what the Y-axis represents. Remember: you spent a long time generating this graph and poring over its meaning, but your audience may be seeing it for one minute tops To them it is completely new. Give them time to take it in, and walk them gently through it, holding heir hand. If the graph shows the results of an experiment, make sure to say something like: “if the flies were all expressing the stinky gene, then we should expect the red line to look like this, and the blue dots like that. If  none of the flies were expressing the stinky gene, then the red line would be here, and the blue dots there. BUT BECAUSE (raise voice) the red line is here, and the blue dots are scattered, it means that only this batch of flies expressed Stinky”. (Replace flies with Raccoons where appropriate.)

This may take some time to explain

6. When outlining methods, cartoons work best. When explaining the methods, a flowchart with graphics beats bullet-point sentences every time. ‘Nuff said.

Gene W Yeo, Nicole G Coufal, Tiffany Y Liang, Grace E Peng, Xiang-Dong Fu & Fred H Gage Nature Structural & Molecular Biology 16, 130 – 137 (2009)


7. Spell out the conclusions. Don’t expect your audience to infer the conclusions themselves. It took you weeks to understand your mess of a lab notebook, so how can you expect them to do that in fifty minutes?

8. Be aggressively hesitant. All research talks have that “sketchy bit” in them, where you are not really sure if these results mean anything, or mean what you think they do. Many speakers try to sweep that part under the carpet, or downplay the problem. Don’t do that. On the contrary,  highlight the problem. Everybody has this type of problem in their lab, and you will create and instant bonding by saying: “hey, I’m not sure what is going on here either”. Another consideration is that, if you attempt to hide or downplay your “sketchy bit”, someone from the audience will call you out. That may unravel the good part of your presentation together with the part you tried to hide. Not good.

9. You should have listened to your mom. Poise may be less important in academia than in a business setting. Still, wear a clean shirt, don’t fidget, stand straight, don’t mumble, enunciate and maintain eye contact.  Also, take your finger off the laser pointer trigger unless you are aiming it at the screen. While on the subject, use the pointer sparingly. It’s not a security blanket.

10. Less is more: Don’t read your seminar from slides. Your slides are not  the place to put the stuff you are supposed to say, that’s what your gray matter is for, or if you have to, index cards. Words on your slides should be kept to a minimum. Slides are for explanatory figures, graphs, and very brief statements.

11. Keep to the time. Shoot for five minutes less time than you are allotted. When rehearsing if you are going overtime, there is only one solution. No, speaking faster is not it. No, combining slides is not it either. Killing some of your material is. Yes, that material is precious, and important, and you worked very hard to get these data. It all amounts to a hill of beans if your audience is fidgeting in their chairs 15 minutes after the seminar was supposed to have ended. Newsflash: at this point you are not the brilliant scientist anymore (if you ever were), you are that tiresome nag that is keeping them from lunch, or uncollegially cutting into the next speaker.

12. Rehearse, rehearse rehearse. Boring, yeah. Do it anyway. A lot. Rehearse in front of your cat, your wall, your spouse, your lover, your drinking buddies, anyone you can get a hold of. You will be bored, they will be bored, and you will improve.





Share and Enjoy:
  • Fark
  • Digg
  • Technorati
  • del.icio.us
  • StumbleUpon
  • Facebook
  • Reddit
  • Twitter
  • FriendFeed
  • PDF
  • email
  • Print
  • Google Bookmarks

Comments are closed.