Playroom and lab meetings

We’ve been trying to figure out how to set up a weekly lab meeting–something where people take turns giving updates on their research, along with a “Hill Street Blues” sort of summary of progress on ongoing projects. My impression is that lots of science labs do this effectively, but I’ve struggled to set this up with my own research groups.

My current thought is that what’s most important is that each person knows his or her role and also knows what to expect in the meeting. Also, the “lab meeting” is not the only way I meet with collaborators and students. We also have various weekly seminars, open Playroom time when people can come in and share ideas, and various meetings of two, three, or four people on specific projects.

Anyway, I received the following suggestions after a recent attempt at a lab meeting:

It was really productive if the spirit of the meeting is very rational and direct. If someone doesn’t understand something, say so. If someone doesn’t like something, say so. If something smells fishy, say so. A group should be cohesive enough not to require diplomacy.

0. Random conversation a bit prior to the beginning of the meeting.
1. Meeting starts strictly on time. Those that come late are frowned upon. The 0 step is intended to make the time waiting social and pleasant, and encourages people to come 5-10 minutes early.
2. Informal seminar (20 mins + 10-20 mins of discussion): One person presents half-baked work for everyone to comment on. During the presentation, people eat the lunch they brought.
3. Progress report: each group member describes in roughly a minute what they’ve been working on the past week so that everyone understands it. If they’ve been to a symposium or seminar or whatever, they give a quick report. Immediate problems they have are discussed, but not more than 2-3 minutes. Anything more than that is moved to one-on-one meetings. This encourages people to have something to report on at the end of each week, and to know what other people are working on.
4. Group decision-making: issues (updating a web page, organizing xyz, arranging for abc, etc) are presented, people are assigned to tasks that need to be done.
5. Team meetings and random conversation. If one person is involved in several meetings, this needs to be scheduled.

Avoid one-on-one interactions. Many people come to these meetings to get consulting from you, and while your insights may be interesting for others, it’s not a good use of time for other group members.

I think that people can work efficiently in the playroom with each other, i.e., there shouldn’t be the sense that I am the center of the meeting and that everyone has to wait for me. Ideally people can help each other (for example, Yu-Sung could help David with R, and David could help Juli with regression modeling, etc.). I’ll have to figure out how best to work things out.

Having a schedule–even a rough schedule–of what will happen is a good idea, I think. It’s much better at a meeting like this if people know what to expect. For example, one-on-one meetings are fine if they are done in the corner of the Playroom and people know when to do them. I’ll to set up something like this. It’s pretty bad that I’ve been a professor for almost 20 years and am still so disorganized…

2 thoughts on “Playroom and lab meetings

  1. I'm curious – how large is your group? Part 3 might not work if the group is too large, and if the group is too small, having to prepare too many presentations can become unpopular.

    I don't know the answers either, and I think you do need a critical mass, so that there are enough people prepared to say something. Well, in Finland you do.

    Bob

  2. One thing that helps the group engage with step 2 is to take binary yes-no predictions from everyone on some aspect of the study (group 1 will outperform group 2 or whatever). Then pull the predictions out when the study is done and see how people did. It is fun to see which lab member is most accurate at the end of the year and people seem to think more carefully about other people's work when they're forced to make a decision about it. Presenters also have to put their work in terms that are easy for everyone to understand to allow for this sort of basic yes-no prediction.

Comments are closed.