Is it time to kill the traditional student presentation?

The traditional student presentation format - ten minutes of powerpoint, two minutes of questions - rarely works well. After three or four presentations, a good chunk of the audience zones out. Getting a good discussion going is hard - and there isn't time for an extended conversation anyways. Then there are the logistical challenges - loading presentations, allocating the unpopular time slots, dealing with late students and no-shows.

This term, my class did poster presentations instead - and we are never going back to powerpoint.

What is a poster presentation? The poster itself can be anything from a deck of 8 or 10 slides printed up and glued to a bristol board to an artistic rendition of the main research findings. "Presentation" is a misnomer. There is no time or need for the presenter to launch into a five minute monologue about their research. Instead, there is a dialogue between the poster presenter and the poster viewer "What is your research about?" "What did you find?" "What does this show?"

Why are posters so much better than the traditional student presentation format?

1. Posters get students talking. At any one time, one third of the class (say) will be presenting, and the other two thirds will be walking around, looking at the posters, and asking questions. Instead of spending a three hour class listening/checking facebook/watching the clock, students are actively engaging with each other and talking about research.

2. Each individual presenter gets much more attention, and much more feedback on their research. The key difference between posters-in-classrooms and posters-in-conferences is that, in a classroom, there's a captive audience. If each audience member checks out each presenter's poster for just three minutes, the presenter is getting far more opportunities to talk about, explain, and discuss their research than they would with a traditional stand up/speak up/shut up powerpoint presentation.

3. Poster presentations build skills students can actually use. Many of my students will never get up and present their research at a conference or for a job talk. However a large percentage of them will end up in job interview where they will be asked "Tell me about your paper." "What was it on?" "What did you find out?" "What problems did you encounter, and how did you solve them?" It's really useful for students to practice answering questions, and explaining their research clearly and concisely.

4. Posters harness students' good natures. Students - at least the ones I teach - like to have a chance to find out what other students are doing, and have helpful and collegial conversations. Poster presentations provide a venue where those interactions can take place. The traditional seminar question-and-answer format doesn't work as well, because students are afraid of either looking stupid themselves, or making someone else look stupid.

5. Posters are great for presenting econometric results. As anyone who has sat through a lot of applied micro presentations can testify, presenting empirical results well is something even experienced economists find challenging. There's room on a poster to present a couple of good sized tables of econometric results - and people can get close enough to actually read them! 

6. It is a far more efficient use of class time. For example, fitting 12 posters into a three hour class is a breeze. Fitting 12 powerpoint presentations in is brutal. 

Hopefully by this point I've persuaded one or two people to think about using poster presentations instead of the traditional powerpoint format. So what are the challenges?

1. Time management. Although the student feedback on posters was overwhelmingly positive, one of the complaints I received was "I didn't have time to talk to everybody." This was an issue because I was evaluating students in part on the basis of the comments that they made on other people's presentations, so a student who didn't comment on everyone's presentations was penalized. 

2. Grading. Poster presentations tend to be a bit chaotic and unstructured - this is what's so wonderful about them - but that makes grading challenging. One solution is to evaluate the poster and not the presentation. This year I asked the students to prepare a timed "three minute thesis" explanation of their research and evaluated them on that. Hardly anybody managed to stick to the time limit, but it wasn't terrible. Next year I'll experiment with allowing the students to assign the presentation marks to the other students themselves.  I also asked my TA to give each presenter a five minute question/answer session on their poster - one way of reducing the chaos would be to have those five minute question/answer sessions held during TA office hours. Honestly, the issue with grading is as much about perceptions as reality - anyone who has graded in-class presentations knows that it's extraordinarily difficult to evaluate these well, and the grading decision is usually made in about 3 minutes. But it's important for students to feel that grading is fair.

3. Venue. The switch to posters was possible in part because Carleton has a new multi-purpose lab with white boards and comfy chairs that is an ideal venue for poster presentations. It's possible to use a traditional classroom and tape posters onto the wall with masking tape, or balance them on desks, but it's not ideal.

4. Ideas spread - bad ones as well as good ones. Suppose one or two students in the class make the same mistake - entering "highest degree" as a continuous variable, rather than as a series of categorical variables, for example, or including an explanatory variable which is clearly endogenous but adds considerably to the regression's r-squared. With a traditional powerpoint, the rest of the class would be too zoned out - or the mistake would flash by so quickly - that no one would notice it. But with posters, because each student is actually looking at every other student's work and trying to understand it - bad ideas can spread throughout the class.

5. Letting students know what is expected of them. Students are all familiar with the traditional student presentation format, and know precisely what to expect. Poster presentations are strange and unfamiliar. So it's really important to take time to explain what's required, how the sessions will work, what the poster should look like, and so on.

But all of these challenges pale in comparison to that one moment when you stand in the middle of the classroom, and look around, and see that every single student is focused, engaged, thinking, talking, and learning... 

Comments

[Christine]

Warning: if you get your students to do a poster, and they are in a competition with students in the Sciences they will lose. The poster technology (software and printing, I imagine) that they seem to use for their conference poster sessions is way way cooler than printing up a few slides and posting them on bristol board.

[Patrick]

They're just using Powerpoint. Create the presentation poster size, and get it printed. If you want to know how, Google will tell you. There are lots of online resources for creating poster presentations.

[Stephen Gordon]

Yes! Yes! A thousand times yes!

I first saw posters at the Bayesian meetings, and I got so much out of them that I insisted on introducing them when I organised the CESGs all those years ago. Add beer and snacks, and it's the best session of any conference you can name.

[Linda]

A few of my colleagues have introduced poster presentations - in classes of various sizes and levels. Their comments were similar to yours, Frances. For one large undergrad class, the students did group posters - so class size doesn't seem to be a barrier. But I was told the first time through was a lot of work for the instructor, especially those who had never produced a poster before! I had intended to try it this fall...and forgot. But this may spur me to try it in the spring session.

[Frances Woolley]

Christine - and the business school students! Though it's important to avoid being text heavy - for posters just as much as for powerpoint - and some science posters I've seen are more like a paper printed out on a 4'x6' board than a presentation.

Patrick - any suggestions? Or links?

Stephen - I think the typical attitude of economists towards posters is 90% (or more) about social status and 10% (or less) about the actual realities of posters. The sad reality is that, in economics conferences, posters sessions are for losers - people who weren't good enough to make it onto the program - so people don't attend. Unless there are sufficient quantities of beer involved! The best way to break that is to have high profile sexy people in the poster sessions. I wonder if the fact that you saw them working at the Bayesian meetings isn't a coincidence - I can see them working really for econometric theory.

Linda - interesting. Would be very interested to hear how your colleagues graded them. Hope you do try them - I can imagine you really enjoying the one-on-one discussions with students about their topics - and the students getting a lot out of your comments!
'

[Chris J]

Posters are easily made in powerpoint or open office. They avoid the title/bullets nightmare that is bad powerpoint. Google ghettysburg address powerpoint for the best bad powerpoi t ever.

[Stephen Gordon]

Frances - Yes, that is an issue. I think the way to address it is for the organizers to fin one or two heavy hitters who are willing to be a good sport and do a poster. In my case, I was able to get Don Andrews to do a poster, and used that as leverage with people who thought they were being relegated to the second tier.

[Avon Barksdale]

I think the success that you are seeing in the poster session stems from physical restrictions that people can violate in a talk. You said that the poster session is confined to 8 to 10 slides and that there is no time to go on a 5 minute monologue about the research. That's why it's better!

Most people – and that includes professional economists – have no idea how to give a talk. An hour long talk should not have any more than 10 slides and probably less. There is no need for the wind-up, no need for the monologue about why you did what you did. Start with the result. One table. One graph. Explain the result in your own voice. Lead with why people want to hear about your work in the first place. A talk (or a poster session) is not the place to prove every detailed theorem or show array after array of numbers. No one – not even the best – has instant memory access to every detailed result in her head and she can't follow dozens of numbers for computation.

Talks are a great way for students to share their work if the professor imposes the structure that every student wants to violate. Students want to show that they “did it right” so they want to include heaps of stuff. Stop them. Get them to talk about the result, the idea. Prevent them from going over 10 slides. Make sure that if they display a table that they talk about every singly column, why it's important, and what patterns emerge. In short, get them to honestly teach their colleagues. If professor imposes this structure, the talks will be as good as poster sessions.

[Linda]

I found this site very useful when I was preparing my first poster. http://colinpurrington.com/tips/poster-design

[Frances Woolley]

Avon - it is nine years - almost to the day - since Nick made his very first appearance on WCI, with this classic post:

http://worthwhile.typepad.com/worthwhile_canadian_initi/2006/11/how_to_present_.html .

Saying much the same thing.

Linda, thanks, that really gives me some good ideas for how to advise students next year.

[Christine]

Patrick: cool! Thanks for info. Will have to figure out printing process (can likely handle software elements pretty easily, now I'm told it can be done).

[Christine]

Have been playing a bit with videos as an optional assignment in one class. It has led to some really quite good as well as entertaining results.

[Tel]

4. Ideas spread - bad ones as well as good ones. Suppose one or two students in the class make the same mistake - entering "highest degree" as a continuous variable, rather than as a series of categorical variables, for example, or including an explanatory variable which is clearly endogenous but adds considerably to the regression's r-squared. With a traditional powerpoint, the rest of the class would be too zoned out - or the mistake would flash by so quickly - that no one would notice it. But with posters, because each student is actually looking at every other student's work and trying to understand it - bad ideas can spread throughout the class.

So you are bothered by a communications medium on the basis that it can effectively communicate?

[Frances Woolley]

Tel: "So you are bothered by a communications medium on the basis that it can effectively communicate?"

It's a challenge. Not something I'm bothered by, just something to deal with.

Christine - interesting! Will any make it into the "how to think like an economist" series?

[Patrick]

Frances: How about this :)

Here too: http://guides.nyu.edu/posters

Poster presentations build useful skills for entry into the private sector. The new hires are often stuck with working the trade show booth ('cause everybody hates doing it). It helps if you've had experience repeating the same pitch and answering the same questions ad nauseam.

[W. Peden]

Hmm, this looks like a great idea. I've had students doing presentations this year in my history of science classes (5 minutes, seated, no powerpoint) and while they've almost all done very well, I do think that engaging their non-verbal communication skills would spice things up a bit, and encourage them to prepare more in advance.

[W. Peden]

(5 minutes in most cases. I've had some extremely good 15 minute presentations as well.)

[Frances Woolley]

Patrick - thanks for the links. I hadn't really thought about the preparation for the private sector angle until my students started saying to each other things like, "Have you ever been to a job fair? It's going to be like that." Now I'm totally convinced that this is a skill our students could use and don't get enough of.

W. Peden - I like the idea of the no powerpoint presentation. Don't know if it would achieve quite the same thing as a poster, but it's an interesting alternative to the traditional powerpoint in class presentation. Worth thinking about.

[W. Peden]

Frances Woolley,

One of the good things for the students, I think, is that they get eye-contact from their classmates, which (a) makes them know they're being listened to, and (b) means they can gauge things like whether or not they're going too fast.

If they had done a statistical study and had some graphs, I suppose I'd allow slides of graphs, but that doesn't apply in this particular class, or any other classes that a humanities person like me is likely to teach. In some more advanced philosophy classes, having them do stuff on a whiteboard would be good. In formal logic classes, I like to get everyone to prove at least something on the whiteboard at some stage of the term.