research poster presentations

It varies by field, but poster presentations are fairly common in computer science.  Many conferences typically split things into full papers with traditional presentations and short papers with poster presentations (although I recall ACL was debating making the presentation decision independent of the paper length).  The peer review process varies by conference; sometimes short papers go through the same process and other times they have fewer reviewers.  Typically they have less competition though.

I recall my two undergrad posters (2001-2002); I didn’t know how to print something on poster paper, so I printed individual 8.5×11 pages and glued them to a poster board.  Yeah, I was that guy.  Admittedly, that was a long time ago and the other students did similar things.  I was clueless about presenting too.

My first graduate poster presentation was at IUI 2006 and I had learned to make a reasonable poster, but I didn’t know how to present it.  I’m still embarrassed about that.  Granted, it was my first publication as primary author and maybe my first conference, and also my first conference alone in a foreign land (though Sydney is really nice and has a great waterside park/arboretum).

The purpose of the poster itself is two-fold:  Firstly, it needs to augment your dialogue.  People will come up and you’ll talk to them, using the poster as a visual support device.  Secondly, the poster is an advertisement.  You’re like a vendor at an auction trying to attract customers’ attention and time.  In this sense, posters are a sort of competition for the attention of the people walking around.

Let’s be clear:  The poster isn’t the main attraction.  You and your research are the main attractions.

As for poster design, I’ll give a bulleted list of guidelines I follow:

  • it should look professional (i.e., print it on a poster printer, make it look clean and organized)
  • clutter is bad.  Everything has cost and benefit – adding a bullet has some informational benefit at the cost of distracting from everything else on the poster.
  • design the poster for older folks
    • font size should be the largest you can get away with.  If you need to use a small font, it probably means you have too much text.
    • contrast is important.  Don’t hurt people’s eyes with too much contrast, but if they can’t make out your delightfully subtle grey subtitles, it’s as if they’re omitted for some people.
  • break down the research into a hierarchy, which is visually apparent
    • corollary:  It shouldn’t look like a grid of powerpoint slides; mix up the shapes a little.
  • content:  Focus on the most important goals.
    • essential background knowledge (the bare minimum they need to understand your sub-field)
    • essential methodology – remember to focus on your novelty, especially compared to your other publications
    • essential results – people need to know “Is this good?”.  Even if you can’t objectively answer that question, you can give examples.
  • pictures/screenshots are good, diagrams are good, graphs are good.  Equations can be good (depending).  Text is a sometimes-necessary evil.  Large data tables are also generally bad.
    • On the other hand, a poor diagram can be fatal.  You need to spend time on them; they should be both visually interesting as well as reasonably informative.
  • use color; you’re trying to interest people.  The poster viewer is like the reader of a novel – they want to be drawn in and excited.  I’m not great at picking color schemes, so I like to use Color Scheme Designer to come up with ideas for the primary/secondary colors.  Once you have colors you like, you just need to convert the hex color codes to whatever your application uses (or guess and check).

I feel like this list is overly long, so I’ll just describe some methods of creating posters then stop for now.

Most people use PowerPoint or Keynote to make their posters.  I’ve tried making posters with OpenOffice, Keynote, and Inkscape before.  Keynote was the best of the three.  OpenOffice is slow on Mac and Inkscape isn’t really suited for text (I had trouble with varying colors of underlines, for instance).  Regardless of what you use, the first thing you should decide is the size.  Consider the conference specifications and also the specifications of the printer you’re using (there’s a for-pay poster printer here at UD, Kinko’s/etc will have a different one).  I usually pick the biggest size I can get away with.

Once you have a size, resize your slide to some scaling of it.  Usually I need to make it in half scale because the programs don’t support a full-scale size.  Then I design the poster, save to PDF, and bring it to the print shop.  Be sure to jot down the size you want (and make sure it matches the aspect ratio of the file).  Usually they need to manually scale it up in Acrobat or something for me.

One last note is to beware of colors – your monitor and the printer almost certainly will look different.  Some places will let you do a scaled-down proof to check colors (although it might not be free).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s