Podcast Episode 64: Collaborative Exams with Teddy Svoronos

- - posted in assessment, economics, episode, podcast, policy

Teddy Svoronos is a lecturer in public policy at the Kennedy School of Government at Harvard. He’s most well known for his creative use of technology in the classroom, but he’s actually someone who thinks about pedagogy first and lets that dictate all of his tech choices. In this episode he tells us how he gives exams where students work independently first, and then teach each other during a collaborative second stage. He also shares some analysis he’s done of the results that have encouraged him to increase the amount of collaboration in his classes.

You can subscribe to the Teach Better Podcast through your favorite podcast app or simply subscribe through iTunes if you don’t have one yet.

Show Notes

0:00 Intro

0:57 Welcoming Teddy Svoronos.

2:08 What is a two-stage exam? Trying to make learning happen in the exam, too. Why Teddy was interested: only 40% of students looked at the solutions for the midterm exam problems.

8:11 Getting timely and non-threatening feedback from peers. Learning collaboration and communication skills.

11:03 Measuring how well people collaborate. The collaborative score is one standard deviation higher than the average of the group. Comparing the collaborative score with the total knowledge represented by the best individual answers. Some groups do better than the sum of the knowledge of the individuals in the group. Problem difficulty as interactions amongst the elements.

The data set now consists of 11 exams and 900 students over three years of midterms and finals.

15:54 The origins of two-stage exams and some details. The Eric Mazur connection. Time for individual exams vs. group exam. In the second stage, students work on a subset of questions they’ve already answered–the more complex ones. About the rubric.

19:57 Doug’s summer workshop at Stanford with Carl Wieman. Using two-stage exams, including at the beginning of the class on a pre-requisite. Doug tries a two-stage pre-requisite exam following Carl Wieman’s suggestion.

22:11 Benefits on the student side: discovering what you don’t know. Students also learning by teaching each other. Learning transfer is improved by giving an explanation after you got the right answer.

And students hear different ways of thinking about a problem. Polling as another way to support student collaboration: how to and how not to.

29:04 Class size is about 70 students. How Teddy decides whom to call on and tracks participation. Trying to avoid implicit biases in who’s called on. Identifying students who would benefit from answering a specific question. Brief pre-class exercises (quizzes) ask for an explanation for each answer, and Teddy uses those to decide whom to call on in class. Teddy talked about how he makes short online video modules on the Teaching in Higher Ed podcast.

32:50 Teaching and learning technology innovation as a question of scale: making a 70- or 200-person course more like a 30-person course. The challenge is individualizing learning in a larger course in a data-driven way. Teddy’s aiming for a system that will figure out whom to call on for the greatest benefit. Learning technology and teaching strategies as ways to reduce social distance in a social network. Social media as an analogy for learning technology. The importance of a comfort with technology. The future of teaching is trying to figure out how to adapt the best elements of face-to-face learning to a range of environments.

37:20 Social engagement as a necessary pre-requisite. What Teddy does as Think GROUP Share. Block randomizing groups in two-stage exams with some distribution of previous performance. Randomizing groups once for the whole semester. But the two-stage exams use different groups than the rest of the class meetings. The two-stage exam then, as an unintended consequence, starts to measure collaboration as a skill, rather than student comfort with their existing group.

41:55 Teddy’s student group projects. Currently students don’t incorporate the final feedback into a report. Everyone comes ready to present, but only a few groups present.

46:51 Two contrasting examples. Edward’s course design is to remove the instructor as much as possible from the network. Doug’s end-of-semester projects get a poster session. Posters are projected instead of printed. Students circulate and defend their posters in rounds. Students vote for the best poster. A written version is due later.

Metaphors. The professor as party host. Unlocking motivations and relationships by reconceptualizing the teacher’s role.

50:53 A memorable mistake you made in the classroom. Letting the students correct you. A comfort with making mistakes in the classroom–because that’s how we learn. Tabby Worinos.

54:51 Signing off.

Comments