Bring on the Data Blitz!

Aug 19 2009 Published by under Academia

At some of the smaller conferences that Sci has been to (which are, by the way, often the best ones) there is a new and improved thing called "the data blitz". Usually, this is where everyone gets one slide to present their poster/talk and to get people interested in it. Usual time is 1 min. 1 minute. Get to the bottom line and get there fast.
Sci will admit that she is not the least verbose person she knows, and that, when this data blitz opportunity first presented itself, she was a little freaked out. But with some guidance and thought, it's actually pretty easy, unless the content of your poster is really THAT complicated, in which case, why would you put all of it on a poster? And so Sci thinks she's pulled off her most recent data blitzes rather well, though improvement is always necessary.
However, it seems that in many people, a data blitz instills more than the usual modicum of fear, resulting in an enormous number of things that just shouldn't happen in a data blitz. And so, in the tradition of things you should never do in a powerpoint presentation, I present: the things you shouldn't do during a data blitz:

1) All usual presentation rules still apply. Face the audience, speak clearly and slowly, don't ACT like you're rushed for time. But you're pressed for time. Focus on the main point of your data, what were your findings, and what do you think they mean?
2) Remember, you usually have only one minute. How much of your data can you explain fully and clearly in a minute? Hint: if it's got more than two graphs on it, you're going to go over.
3) If you have animation on your data blitz slide: TAKE IT OUT NOW. I'm not kidding. Most of the time, you don't have control over your slide, and you don't want to pause in your story to say "oh, could you click the...and again...ok, one more time...". Also, NEVER cheat when you have one slide by using a butt-ton of animation to post fifty million graphs on it. It's annoying, and is going to take you WAY longer than you have.
4) If you have more than one sentence of text on your slide, you have too much.
5) DON'T go over time and keep going, pretending like you didn't hear the timer. We all heard it, we know you did, too. Wrap it up.
6) OMG. You have ONE slide. ONE minute. There is NO excuse here for font smaller than 18. Hell, there is no excuse here for font small than 44!
7) If seeing someone read their slides aloud in a talk is bad, it sends you straight to hell in a data blitz.
8) And there is nothing more annoying than seeing someone literally put their ENTIRE poster up on one slide, and try to race through it in one minute. If Sci catches you, she will hunt you down.
So...if you shouldn't do all of these things, what SHOULD you do?
A) Some people like to be extemporaneous in talks (Sci does, though she has her transitions down pat). But for a data blitz, it's a good idea to plan out what you want to say, and the best and shortest way to say it. That way, you know you'll be as clear as possible while still getting out your main point.
B) If you need to put on more than two graphs you have TOO MANY. If you cannot tell your story in two graphs, tell the first bit, or the last bit, or the weird bit, or the most awesome bit. Leaving some information out is a good thing and will make people want to come to your poster and hear the rest. No need to fill them in and have them skip.
C) If you really don't have one graph that tells your story well, post a model (as long as it's not too complex and crazy), or a picture. If you need to, just post your title. But don't ever overwhelm people with graphs. A pretty, simple model seems like just the thing.
D) Be enthusiastic and happy about your data, even if it's unexplainable. Who knows, you might get people to come by and give you insight. Don't be apologetic or self-effacing, we WANT to see your stuff, or we wouldn't bother with this conference.
Sci will be honest, though. The best data blitz she ever saw was a motivational poster, photoshopped to be part of the presenter's title. Pretty awesome, that. Unfortunately, it didn't mean people were going to come to their poster.
Now go forth! Blitz Sci with DATA! And include any further suggestions below.

6 responses so far

  • Janne says:

    There was such a presentation at a meeting I was at last month. One minute per person, split up over two days (half the posters presented first day before lunch, half did so on the second day).
    I have to say I don't care for it. The problem is, you have fifty people lining up to give their one minute spiel. Even if everybody keeps to one minute, each changeover is another ten-twenty seconds. If you're part of the waiting group you're focusing on your own short speech and completely miss what the others are saying.
    Worse, when you're just siting down and listening you realize that an hour of short snippets - many of which you have little direct interest in - soon blur into one unfocused fuzzy blob of boredom. You sit back, do your email or daydream of lunch and inevitably miss the posters that _would_ perk your interest.
    I think that, at least when the number of posters are more than two dozen or so, it may be better to group the posters and have somebody present them all as a group (there's this long list of people in the program committee; set them to work): "Next, the group about neural adaptation. Here we have Dr. Swanson with her study about NMDA-mediated adaptation in the presence of nicotine *flip*; Mr. Adams and Dr. Tweed shows us how song patterns change in the common weed warbler *flip*; .."

  • David says:

    I went to a conference recently that had a similar thing, 5 minute speed papers. I really enjoyed it, and many presentations (the ones that didn't attempt to squish 20 minutes into 5) were excellent, and piqued my interest enough to ask the speaker to explain in more detail afterwards.
    There were plenty of presenters, even senior ones, who had little grasp of how much material was appropriate, and didn't get even half way through their slides. They were forced off -- no standing there awkwardly while the timer ticks over.

  • They were forced off -- no standing there awkwardly while the timer ticks over.
    Did they use the hooked end of a cane around the speakers' necks? Like in a vaudeville show? Because that would be conference job EVAH!

  • gradstdentsteve says:

    1. the smaller conferences are the better ones: I've just gotten started doing conference in IR/poli sci, and I've found the same thing.
    2. data blitz: I love this idea. LOVE IT. I haven't seen it, but it sounds great. but...perhaps, an idea after reading Janne's comments... putting an hour's worth of data blitzes together is definitely going to be worse than an hour long panel w/ the typical 3-5 papers, since it'll be harder to dig into anything. and the greater motion of people and materials will be distracting. though, it would be easier to drop in and out while you think about other things. but, perhaps it would be organized better if the blitzes were matched to the panels. that is, have the poster presenters do their blitzes at the beginning or end of panels that relate to the material on the poster. kinda like targeted advertising

  • David says:

    Did they use the hooked end of a cane around the speakers' necks? Like in a vaudeville show?

    That would be awesome. Unfortunately, it just involved the chair walking up to the podium to introduce the next speaker. A few cheeky presenters used those precious last seconds to squeeze the rest of their talk in one exasperated single breath sentence.

  • There was a Datablitz like this at the APA Convention last summer. I think they had 2 minutes each to talk. It was really entertaining!

Leave a Reply