Flipping the Classroom

Houtan Emad, below, is a student in a flipped version of Math 200, taught by Louis-Philippe Saumier, above. (Photo by Hugo Wong)

For generations, students have spent their days furiously copying lecture notes from a chalkboard and their nights churning out assignments based on those notes. This familiar model is the dominant one for most university classes, but a few UVic instructors are turning this model on its head. Instead of letting students practice concepts at home after absorbing material in a lecture, a new format, dubbed “flipping the classroom,” does just the opposite.

“Flipping the classroom, to me, is just simply a movement to push the initial delivery of information and content out of the classroom and online,” explains Jamie Kemp, a PhD student and sessional history instructor at UVic. “The beauty of that is that you’re able to spend your class time doing more interactive things, so really applying knowledge and engaging with the course content in a more dynamic way.”

The idea, already used on a smaller scale in the humanities, first gained traction in the sciences in the early 1990s, when Eric Mazur, a professor of physics at Harvard, noticed that while his students did well on complex problems, they were still struggling with fundamental concepts. In an example from a presentation he calls “Confessions of a Converted Lecturer,” Mazur tested his students by pairing simple, conventional physics problems with application questions. He found that students in general did worse on the application questions and sometimes had serious misconceptions about basic concepts, but many still managed to get the right answer in the end. This meant two things: that students could do well in the course through rote memorization alone, and that teachers could be misled into thinking that they were teaching effectively. Mazur also noted that students would be frustrated when their “recipes” (ones that did not consider the underlying concepts), wouldn’t work in subsequent problems.

Mazur is not the only one who noticed a lecture’s shortcomings. Jason Siefken, a sessional instructor and the first math instructor at UVic to flip the classroom (for Math 100 last summer), sees other problems, including the lack of student participation. He felt that there was a better way to use valuable time with the instructor.

“In essence, the professor is doing nothing but telling you things,” says Siefken of traditional lectures. “You’re not getting any benefit of interaction . . . Why would you not record that, make it available to anybody on their own time and then do things that actually rely on interaction during the class time?”

Siefken wanted to make a change, so he attended a workshop led by a senior UVic math instructor, Alfonso Gracia-Saz, who is cross-appointed with the Learning and Teaching Centre, which serves to enhance teaching effectiveness at UVic. Gracia-Saz’s workshop introduced math instructors like Siefken to elements of Mazur’s teaching method. Though the attendees had taught math conventionally in the past, with lectures and homework, they agreed to give it a try.

Louis-Philippe Saumier, another math instructor who attended the workshop, is currently using this method in his summer Math 200 course. Some of his students, who registered for the course not knowing what Saumier had in mind, were skeptical at first, but many are coming to like it.

“Some people are buying into it more and others are attached to the lecture,” says Saumier, “but I ran a survey earlier in the class and 90 per cent of the students are pleased or really pleased by this format . . . especially the ones taking the course the second time. These are the ones that failed the traditional method. They come to see me, and they say, ‘It’s 9 000 times better than what I had; all math courses should be taught this way!’ ”

The night before class, Saumier’s students view instructional YouTube clips or read online articles. The video clips are a mix of pre-existing videos and ones that Saumier creates himself. For many topics, he makes two sets of videos: one set slower-paced, the other faster-paced. This allows students who have already seen the material to quickly review concepts without testing their patience, while students who are seeing the concepts for the first time can watch the slower-paced videos and still have time to take notes and get more thorough explanations. In class, Saumier asks students to solve problems based on the online course material. Students indicate their answers through iClickers — wireless remotes that allow students to respond to multiple-choice and short-answer questions. A bar graph showing the class’s answers then appears on the screen, which helps Saumier gauge whether the class understands the concept. If a majority of students responds correctly, Saumier can introduce new topics; if there is a mix of answers, he knows to explain further.

This kind of polling data is valuable to instructors like Saumier, and it is more effective than simply asking students to raise their hands or scanning the room for confused faces. As his colleague Siefken notes, “University students are experts at nodding and looking like they’re getting things . . . Say you lecture and you look around for confused faces, and it looks like everybody’s got it, and you go on. This is absolutely not true. Most people haven’t gotten most of what you’ve said, so this is where the clicker component of the courses comes in . . . They first think about their problems individually, and then we [the instructors] have them vote with their clickers, and then there’s peer discussion and revote. And when they vote with their clickers, it’s not for them; that’s for the instructor.”

Whatever the discipline, lectures provide students with largely passive learning opportunities. Flipped-classroom proponents seek to turn that tide, but the size of the class often determines the methods used to increase engagement. In large classes like Saumier’s, technology like iClickers can help. However, in smaller humanities classes, there are different, lower-tech ways to improve engagement.

As a sessional instructor of Medieval Studies 451, a course on medieval manuscripts, Kemp flipped her classroom earlier this year by relying on readings instead of technology, assigning around 200 pages of reading a week (but with plenty of pictures as one student concedes). Students then wrote 300-word responses to that material and discussed their findings in class. Unlike math and science courses, she points out that elements of the flipped classroom have been “put into practice in humanities departments forever.” Her small class size allowed her to not just increase student participation by asking for their opinions, but base entire classes on discussion.

“The reason that I started with the flipped classroom is that the first time I taught Medieval Studies 451, I taught it in a relatively traditional, seminar format. I did some lecturing and had a lot of very stilted discussion, and it just wasn’t a very interesting class. So when I had the opportunity to teach it again and again, I’ve tried to focus on turning it into something that students couldn’t do at home.”

Kemp first learned of Mazur’s method, called “Peer Instruction,” after hearing about it through her prior work at an education technology company, and adapted it to fit her needs. For her, the peer element is an important part of the experience. Those 300-word responses motivated her students and encouraged them to challenge the material they encountered, unlike typical courses where the assigned material is often the only view presented.

Kirby Delaney, a recent history graduate and a student in Kemp’s class, spoke fondly of the discussions.

“We would all sit down together and discuss the article, and we’d get into quite heated debates, with some people really believing what the author had to say and other people just thinking it was a waste of time. It was really interesting, because you do a lot of reading for history classes, but you very rarely get a chance to discuss them in that way. You usually just accept that whoever is doing the writing is correct, and you never get a chance to actually insert your own opinion.”

“Even though it may be reading-intensive,” says Kemp, “you don’t necessarily have to ‘get it all’ on the first try, because your peers in the class will help you and you will post in forums and you will work out the meaning of that text as a community. It’s so easy to forget that learning is a fundamentally social undertaking.”

In addition to extensive classroom discussion, Kemp also found time to further extend the experiential learning by incorporating enrichment activities, like allowing the students to handle real manuscripts in the university archives.

“It was so hands-on, which is interesting for a medieval studies class,” says Delaney. “We got to go down on a couple of occasions to see what UVic has for manuscripts, and they’ve got some really beautiful ones hidden down there . . . They became not just pages with writing on them but almost creatures with little histories of their own.”

In traditional classrooms, instructors often assume that many students skip assigned readings, so class time may be squandered reviewing that material. In flipped classrooms, students are forced to do the learning work in advance, and that can sometimes be perceived as more work. Students and teachers, regardless of faculty, agree that simply absorbing the material once at home is not enough, and that the thorough preparation and self-discipline demanded by a flipped-classroom model can be challenging.

When asked if the flipped classroom was better than the regular one, students Houtan Emad and Spencer Perks could not say for sure. Emad and Perks, each in their second year of engineering studies, are currently enrolled in Saumier’s flipped Math 200 course. They like the added participation and anonymity that the iClicker system provides, and agree that they’re getting a more in-depth understanding of the material. However, they also say that while this method of teaching might be appropriate in the summer term, when their course load is lighter, they may not have enough time to watch the videos for a flipped course while balancing five other courses and the associated labs during the fall.

“It would be fatiguing to have more than one flipped class,” says Perks, “because it takes a lot of initiative on your part to actually go home, watch the videos and pay attention — especially since it’s on the Internet. You can easily open a tab and go to another YouTube video. It’s right there.”

Emad, like practically every science student, has been used to the lecture format since high school and has optimized his study habits to accommodate the standard classroom’s strengths and weaknesses. Because the flipped model is so new to him, he hasn’t fully adapted to it yet, but as a first-time student of Math 200, his ambivalence is not unusual.

The added workload is noticed not just by students, but by the instructors as well. As Saumier is teaching 200-level university math, existing online resources are more scarce, so he must make many of the videos himself. Kemp is charged with planning enrichment activities and monitoring online forums in addition to answering student queries in the off hours. “I would say [the flipped method is] high-touch and very personalized in terms of the learning experience, which takes a lot of time,” says Kemp.

For Saumier, Kemp, Delaney and others who like the method, the added workload does not detract from the benefits they see in this teaching method. In Delaney’s case, Kemp’s flipped class made her seriously consider pursuing a master’s program. She didn’t feel ready for a higher-level degree until she took the class, which she says made her feel “valued and intelligent as a human being and not just the person sitting in the back and taking notes.”

The effectiveness of peer discussion is cited extensively in Mazur’s publications. In a 2001 article published in the American Journal of Physics, Mazur compiled 10 years of data after his initial foray into Peer Instruction (PI) in 1991 (one incarnation of the flipped classroom). He administered the same diagnostic test on basic Newtonian mechanics at the beginning and end of the term, wondering how much students would improve after taking his course. The beginning-of-term results stayed fairly consistent (hovering between 67 and 71 per cent), but after changing his teaching style, the gains at the end of the term increased (from an 8 per cent gain in a traditional lecture to a 14 per cent gain in a flipped classroom). As Mazur refined his methods, students improved further, like in 1997, when his students scored 25 per cent higher by the end of the term.

At UVic, the results are still preliminary. The flipped classroom has only been used in the math department three times to date, so no concrete conclusions can be drawn yet. However, the results are promising. In an email, Siefken stated that the failure rate for students who participated by using their iClickers in class was four times lower than that of those who did not. The failure rate for flipped math classes as a whole is also lower than the average for equivalent lecture-based math classes, according to Siefken’s data.

Even the most die-hard flipped-classroom advocates acknowledge the importance of a great lecturer, and they do not see the lecture disappearing anytime soon. Still, instructors like Saumier would prefer to see students have more options — not only in what material they study, but in how that material is delivered. Universities often offer multiples of high-demand courses, and Saumier says if some were lecture-style and others were flipped, students could choose the method that works best for them.

For those instructors looking to take the flipped classroom for a spin, Kemp suggests experimenting by flipping a single unit of the curriculum and evaluating the results, or adopting parts of the method. Every idea behind the model is designed to improve student engagement and deepen students’ conceptual knowledge, whether it’s with iClickers or just old-fashioned reading. This alternative classroom model shows much promise and might make traditional lecturers take notes of their own.

Leave a Reply