spring 2011

CS376: Research Topics in Human-Computer Interaction

Discussion Guidelines

Each student will (co-)lead one discussion during the quarter. Meet with Scott and Chinmay at the end of the previous class to discuss your plan. On your discussion day, submit your preparatory materials (notes etc.) via the online course submission system instead of your critique.

Most importantly, begin your preparation by defining a very clear learning goal for the class. What will students understand at the end of class that they may not have at the beginning? Decide on 3-4 key ideas from the readings that you would like the students to understand deeply, and structure your discussion strategy around that. As part of that, your discussion should accomplish the following:

• Briefly summarize the papers for those who may have not read them as closely as others. (This is most important for papers where a critique is not required.) This should take no more than 10 minutes. You can summarize the papers together or separately, whichever you prefer.
• Lead a conversation about the papers, covering topics much like those that would be covered in a critique. You don't need to provide all the content, but you need to be willing to step in at any point when no one else is providing the content. Don't plan on giving a 45-minute monologue. Do come in with a clear set of goals for what you would like to convey.
• Get people talking! One great way to do this is to break people into pairs or small groups for a minute or two to think about a question. Pairs and small groups give more students a chance to participate, and they help students get ideas and words flowing. And by a minute or two, we literally mean 60 or 120 seconds. Inside of class, pair/small group discussion longer than a minute or two tends to lose momentum and fall off track.
• Cover both high- and low-level parts of the readings. What does this research mean? High level concepts are important, they help us anchor on the topic and give us some motivation for a research topic. How was this research accomplished? What technical concepts and methodological strategies were employed? Really dive into a key detail of how a study was performed, or whether the right research question was asked. If it's a technical paper, the discussion should help students deeply understand the key technical ideas.

We're very open to students trying something innovative or different during the discussion they lead. If you are going to do something "unusual", make sure to talk with us about it several days beforehand so that we can help you determine if it is appropriate and achievable.

In general, avoid the projector. There are two reasons for this. First, the projector is loud, making it more difficult for some students to be heard. Second, somehow when there's a glowing picture at the front of the room, students tune out and stare blankly at it rather than engaging with the discussion. If you have a simple diagram, come to class a few minutes early and draw it on the whiteboard. That said, short videos and demos can be great. One strategy that can work well is to begin class with a short video or demo, then turn the projector off.

Before class on the day of your discussion, read through all other students' critiques. (They're viewable through the course submission system after 7am). Weave ideas from your peers commentaries into the discussion and credit their author's. It's a nice way to encourage the excellent, thoughtful work that students do in writing their commentaries, and also bring those ideas and their authors into the classtime discussion.

Arrive early to class on your discussion day so you aren't flustered. Open up the blinds in the back to let some natural light in. (And, if you like, the windows.) Encourage people to sit towards the front -- it gives the discussion more energy. An important part of teaching well is leading class with energy and enthusiasm.

After class, grade the student commentaries. We recommend grading literally right after class while everything is fresh. Don't spend a huge amount of time on this -- essentially, the goal is a "check", "check-plus", "check-minus" grading system, based on the depth of student's intellectual engagement with the paper's core ideas. In general, the majority of commentaries get a 'check', with exemplary commentaries getting a 'check-plus' and weaker ones getting a 'check-minus'. That said, this class is not graded on a curve, so any particular day may or may not match that general trend. Enter grades through the admin interface. You will have admin access for two days following your discussion day.

Finally, see the grading page for details on how the discussion you lead will be assessed.