A reading seminar where every student reads, writes and contributes to the discussion in class

I remember reading seminars as follows: one student spends the entire week preparing for a powerpoint presentation, which often turns out to be stressful for the student and somewhat boring and uninformative for the audience. The other students only glanced over the paper and so any discussion quickly falls flat. I therefore decided to have multiple short presentations without powerpoint (less preparation, more fun to listen to, plus repetition is good for learning a skill). I also decided to use short writing assignments as homework to make sure that all students were prepared to contribute to the discussion in class. At the same time, I wanted to keep things manageable for everyone.

1. Learning to present: every student does multiple short presentations without powerpoint.

No powerpoint: I didn’t want students to spend too much time preparing a presentation. I believe that often, when students spend a lot of time preparing presentations, they focus too much on making powerpoint slides and not enough on informing the audience and telling a story.

Short presentations: Doing an engaging 45 minute presentation is extremely difficult, and a skill that most postdoc don’t have, so why do we use 45 minute presentations in our graduate seminars? I decided in stead to let each student do three 10 minute presentations.

Feedback: After each presentation the presenters got feedback (from the other students and myself), so that they could improve their presentation skills during the semester.

Easy listening: An added benefit of 10 minute presentations is that it is much easier for the audience. Each week started with three student presentations, one on the background and main question of the paper, one on the data and the results of the paper, and one on the conclusion and implications of the paper.

2. Practice writing: every student does a different writing assignment every week.

Graded homework each week: A paper discussion can only work if people have read the paper. If students don’t read, they may spend most of their energy to try to hide that they didn’t read (I know I was in that situation!). So even though I understand that life and research get in the way of reading, I really wanted to make sure that the students were prepared for the seminar. To do that, I made every student do a written assignment every week that would count towards their grade (unless they were presenting that week).

A different assignment for each student: I had a long list of assignments so that each week, many different assignments were done AND so that over the course of the semester each student did many different assignments. This guaranteed that the students read the paper, but each with a different question in mind.

There were several types of written assignments. Descriptive: 1. Describe the background and main question of the paper, 2. describe the data and the results, 3. describe the conclusions, 4. describe which virus the paper is about. Critical: 5. What is your opinion of the paper? 6. What do you think the authors should have done differently? 7. Play the devil’s advocate: why should the paper not have been published? Summaries: 8. Summarize the paper in your own words, as if writing to a friend, 9. summarize the paper using only the most common 1000 words of the English language, 10. summarize the paper in a graphical abstract, 11. summarize the paper in a tweet. Meta: 12. Who are the authors of the paper? 13. How often is the paper cited, do you think it is influential?

Short! Each written assignment could not be more than 150 words, to keep the workload manageable for me and for the students.

Surprisingly hard: Some of the assignments were harder than the others. Summarizing the paper using only the 1000 most common words from the English language turned out to be very hard, but some of the students did a great job (see here and here). The graphical abstract was also hard for some students, but others liked it just because it was so different from their usual work (see here and here). The ”devil’s advocate” writing assignment was always very interesting to read.

Easy: Grading the written assignments was quite easy. I simply gave a plus or minus for 5 categories (answered the question, scientific accuracy, clarity, grammar and word count).

Revisions allowed: After a request from a student, I decided that the students could redo any assignment where they had gotten less than 100% because I believe that feedback is most useful when it can be applied to a revision.

3. Promoting equity: thanks to the written assignments, every student could contribute to every class.

Everyone contributes: One of the nice things about the homework schedule with different assignments for everyone is that in class, I could ask each student about their homework. This way, each student contributed to the class, promoting equity, and the brief discussions of the homework assignments always let to questions from other students. Even if I didn’t ask, some students would volunteer to share information they found while they researched for their homework. For example, I remember someone remarking at the end of a presentation: “In your presentation, you said this result may be very important, but I found that the paper hardly has any citations even though it was published ten years ago, so I think it may not have been picked up by anyone.”

Sharing homework: I also encouraged the students to share their written assignments on the online forum we had for the class, so that the other students (and not just me) could read them. Sometimes they led to interesting forum threads. I also published some of the written assignments on my blog, after asking the students for permission. This way even more people could enjoy them.

A better way to get feedback on a talk

A few weeks ago I had an important talk to prepare, and I wanted to get some feedback to make the talk as convincing as possible. So, as is usual in the lab where I work, I scheduled a practice talk a few days before the actual talk, but then I decided to do things slightly differently.

The usual way to get feedback:

1. Give a practice talk for the lab.

2. Ask people to give feedback directly after the talk. 

While this method is at least a hundred times better than not giving a practice talk at all, I thought it could be improved.

Here are some of the drawbacks of the current method:

1. The people in the lab are not necessarily in the best position to give feedback. For example, they may understand parts of my work that are difficult for another audience.

2. It takes too much time (for me and the others) if everybody needs to tell me their feedback.

3. I will probably not remember everything that the audience told me (except the general message that there is always a lot to improve).

4. Not everybody will actually state their opinion. Some people are too shy to speak up and a consensus may grow during the discussion, whereas I would like to know the variety of opinions.

So this is what I decided to do:

1. I scheduled a practice talk, as usual, but in stead of letting people tell me their opinion, I asked them to write it down. I explicitly asked them to write down what they liked about the talk as well as ideas on how to improve it.

2. I also screencasted a part of my talk with Quicktime (this records slides and my voice) and sent it to a collaborator in another city, so that she could make sure my presentation was correct on her part of the project. In principle, the screencast could be sent to a friend in another field to get the opinion of an outsider. 

How it worked

First of all, I discovered that after my practice talk, several people were eager to tell me their feedback directly. I asked them to write first, but then I also got oral feedback and this led to some interesting discussion about the content of my talk.

The written feedback from the practice talk was extremely useful. I read everything immediately after my talk. It hurt a bit. Just like honest feedback always does. I noticed that there were a few opinions in the written notes that I hadn’t heard in the seminar room. Two people wrote that they liked the first part of my talk, whereas the consensus in the room seemed to have been that this part wasn’t very good yet.

I put the notes away and worked on my talk for two more days (yes, it was an important talk!). Then, when I thought I was ready, I looked at the feedback again. Surprisingly, this time the reading was not painful at all! Now it was like reading a to-do list with only items that I had already done, interspersed with compliments.

I was happy that I had asked for the positive feedback, because it was motivating to read as I was sitting in my foreign hotel room, it got me excited about actually giving the talk. Rereading also helped to make sure that I hadn’t missed anything important (the axes were unlabeled on slide 15).

The video I sent to my collaborator in another city was also useful. It was easy to do (see link with explanation), it felt good to have my collaborator involved in the process, and she sent me an email with detailed feedback.

All in all, my conclusion is that it is very useful to get feedback in writing. It saved time, I got more opinions and I was able to reread it a few days later. In fact, I used the written feedback method for my students in Munich in the four years that I taught scientific presentations. I should have known that what was good for my students is good for me!