Discussing PI, teaching, and awards with Zingaro

In commendation of the stellar teaching that is exhibited by new professors at U of T, an honor dubbed the Early Teaching Career Award is distributed each year to select professors that are nominated and supported by the students they taught. Every year, up to four recipients are chosen for this award, with the stipulation that the recipients be within the first six years of their teaching career.

One such professor, Dr. Daniel Zingaro, assistant professor in the mathematical and computational sciences department, won the award last month in February. With a primary focus on computer science and mathematics, he is soon to reach his sixth anniversary as a professor at U of T.

“I have won some research awards […] this teaching award means more to me than those research awards do, because my students helped me win this award; they determined that I should win it. They wrote reference letters, and that’s a big factor in who wins this award—what is the actual impact on students?” Zingaro says.

Zingaro goes on to explain the importance of student input in winning the award over other factors, such as his application and the input of fellow colleagues. “By virtue of winning the award, I know that [students] were very complementary, so that was really nice; to have an award that, in some sense, was determined by the students.”

When prompted on how he tends to structure his lectures, Zingaro had much to say. “My claim to fame is peer instruction, or PI, as it’s called. It’s designed to replace lecture by these multiple-choice questions that test conceptual understanding.” Having done much research on education and its applications, some interesting studies resulted in his adoption of this teaching style. Notably, Zingaro points out that “If you do a lecture for an hour or two, students learn very little from it. I think this is an extreme case, but there are examples in physics where the prof has tested students from the beginning of a lecture on some upcoming material, and then they have a lecture. Then they get tested after the lecture, and they do worse than before!” He remarks that while leaving students confused after lectures isn’t always counterproductive to learning, as it may motivate them to learn more, the ultimate verdict is that students are digesting much less material from traditional lectures than expected.

As an overview of PI, the technique operates as follows: students expose themselves to course material before lectures, whether that is through readings, videos, or other mediums. Students then come to lecture with an incomplete, but workable understanding of what will be taught. They’re given a multiple-choice question that is specifically designed to facilitate conceptual understanding; each wrong answer points to a common misconception that students may have about the question posed. Generally, students will work silently and independently on the question first—a marking scheme based on participation and not correctness ensures that cheating for correct answers doesn’t happen, while providing an environment free of stress. After the first submission of individual answers, students then work in groups and re-do the question. “What we see in the research is that the correctness on the first vote, the individual vote, is usually something like 40-50 per cent or so. And then it goes up dramatically for the group vote; it goes up to 70 or 80 per cent. And you might think, well, maybe they’re just copying from people who they perceive to have more understanding. But we did more research using isomorphic questions, and suffice to say, we have a lot of evidence showing that there is a little bit of copying, obviously, but for the most part, it does demonstrate conceptual understanding of what they’re learning.”

Zingaro has done a plethora of research based on PI, and regardless of what is examined—student opinions, exam grades, conceptual understanding—classes structured around PI simply perform better than lecture classes. In fact, his main area of focus when it comes to research does not relate to computer science at all—he obtained a Master of Education through OISE, the Ontario Institute for Studies in Education, and then went on to obtain his Ph.D. in Education at UofT.

To provide some insight into the work that professors do, when asked after his opinion on the obligation of professors to teach and whether all professors enjoy it, Zingaro gave the question some thought. “At U of T, we have two kinds of professors; we have a teaching stream, and we have a research stream. We have breakdowns for how we’re expected to spend our time. So, a teaching stream is 80 per cent teaching, and 20 per cent service. The research stream is, I think, 40 per cent teaching, 40 per cent research, and 20 per cent service.” Although “service” was not clearly defined here, it seems to refer to responsibilities such as service to committees, organization of conferences and events—in general, activities of leadership.

Zingaro is ambivalent on the distinction between these two streams. “The teaching stream professors… a lot of us do research, and a lot of us are experts in pedagogical research. So, I find it somewhat concerning that we don’t have the word ‘research’ in our titles, even though we do a lot of it. Correspondingly, the research professors do a lot of teaching, and they don’t have ‘teaching’ in their title.” Regarding the question posed, “Are all professors passionate about teaching? That’s a clear no. I think some professors don’t enjoy teaching very much at all […] they are influencing a lot of students that take their classes, and I’m sure it comes across to students’ which professors care and which don’t. But that said, the demands of research are immense. It’s very challenging to be an expert in both teaching and research. It’s doable, but I think most professors choose one.”

Zingaro may be found teaching various CS courses within UTM, he also aspires to dabble in teaching some math courses in the future—the disciplines are, after all, intimately connected.