Academic Articles 1: Student Teaching

Seneca, in his Moral Letters to Lucilius gives the following advice: “Associate with those who will make a better man of you. Welcome those whom you yourself can improve. The process is mutual; for men learn while they teach.”

This is one of many passages where Seneca displays his stoic sense. And science has backed Seneca up. People really do learn by teaching others. For over a hundred years now researchers have been rigorously studying teaching as a way to learn and the results are compelling.

Looking at some of the recent(ish) research on the educational value of teaching seems like an appropriate task for our first post in the Academic Articles miniseries (a series where we look at academic literature which has an important bearing on debate camp). In this post I will look at three different developments in the literature on student teaching. For each of the three developments I will briefly review the literature (though I encourage you to read the studies and articles yourself) and then mention what I take to be important take-aways for debate camps.

Part One: The Lessons from the Peer Instruction Program

Crouch, Catherine H., and Eric Mazur. “Peer instruction: Ten Years of Experience and Results.” American Journal of Physics 69.9 (2001): 970-977.

Lasry, Nathaniel, Eric Mazur, and Jessica Watkins. “Peer instruction: From Harvard to the Two-Year College.” American Journal of Physics 76.11 (2008): 1066-1069.

Mazur, Eric. “Peer Instruction: Getting Students to Think in Class.” AIP Conference Proceedings. Vol. 399. No. 1. AIP, 1997.

1.1: Brief Literature Review

The Peer Instruction program is a model of teaching introductory physics that was first developed at Harvard University, but which has since been adopted by many other schools. It has a lot of similarities to the TEAL method, later developed at MIT, which has also since been adopted by a lot of schools around the world.

The PI program emphasizes reciprocal peer instruction and discussion rather than passive notetaking. The general model looks like this:

“A class taught with PI is divided into a series of short presentations, each focused on a central point and followed by a related conceptual question, called a ConcepTest, which probes students’ understanding of the ideas just presented. Students are given one or two minutes to formulate individual answers and report their answers to the instructor. Students then discuss their answers with others sitting around them; the instructor urges students to try to convince each other of the correctness of their own answer by explaining the underlying reasoning. . . .  To free up class time for ConcepTests, and to prepare students better to apply the material during class, students are required to complete the reading on the topics to be covered before class.”

The result of implementing the PI program was a huge improvement in student performance. Not only did it improve conceptual mastery and quantitative problem solving, it also decreased student attrition and in some cases as much as halved the number of failing students.

Likely there are other benefits as well. For example, PI probably helps students refine their ability to learn by reading, a critical skill given how much of advanced learning occurs on one’s own.

1.2: Implications for Debate Camp

The remarkable success of the PI program shows something important about how students learn. Often it is more valuable for a student to try and master material on their own, and then refine that understanding by attempting to explain and justify their ideas to others than it is to simply listen to an instructor explain and justify answers by lecture.

This has some clear applications to debate camp instruction. For example, VBI is revamping its seminar program (as an option for students instead of modules) for precisely these sorts of reasons. Seminars allow students to read carefully ahead of time, and then answer questions and explain their ideas to their peers under instructor supervision. This should help develop a far deeper understanding than students will be able to acquire through even extremely well done lectures.

Another way to use these insights is to refine lab instruction. Often labs split up for drills but stay in large groups for discussions. However, as these articles note, that provides less time for each student to articulate their own thoughts. Thus, it is often helpful when labs are having a discussion, say about strategy, to split the lab up into smaller groups of ~4 people and have discussion take place in that smaller context. This gives each student more time to explain their ideas to others increasing individual peer instruction time.

Part Two: Student Tutoring and What it Tells Us

Cohen, Peter A., James A. Kulik, and Chen-Lin C. Kulik. “Educational Outcomes of Tutoring: A Meta-Analysis of Findings.” American educational research journal 19.2 (1982): 237-248.

Roscoe, Rod D., and Michelene TH Chi. “Tutor Learning: The Role of Explaining and Responding to Questions.” Instructional Science 36.4 (2008): 321-350.

2.1: Article Take-Aways

There is not a lot in the Cohen piece that we need to dwell on. I included it, however, because it provides a nice meta-study which helps document the amount of evidence there is that student tutoring does not just improve the performance of those who receive tutoring but also improves the academic performance of those who provide tutoring. The act of teaching other students to learn a subject improves one’s own understanding.

The Roscoe article is the far more interesting read. The article opens with an excellent survey of the literature discussing why people learn by tutoring.  Several mechanisms are worth mentioning. First, tutoring requires you to explain ideas and, inevitably, explanations that are clear to you are opaque to others. Thus, tutors must to reach for alternative explanations, be it new analogies, new framings, new examples or what not. Generating these alternative explanations encourages conceptual mapping which deepens understanding. Second, tutoring helps students identify and fill gaps in their own understanding. “Tutors may need to utilize metacognitive processes such as comprehension-monitoring and metamemory, which involve evaluation of the quality of one’s own knowledge and understanding.” This encourages students to further develop and refine their own understanding and helps students get over the problem of ‘not knowing what they don’t know.’

Following this discussion, the article presents its own study on what sort of tutoring had the largest effect on the tutors understanding (an unfortunately neglected subject). There are a lot of results worth considering, and I highly recommend reading through the article. But one of the main lessons is that tutors learn more effectively when their role goes beyond ‘summarizing’ or ‘reexplaining.’ The more that the tutor’s role involves their own synthesis the more the teaching provides opportunities for metacognitive reflection. The danger is that students may prefer to “deliver rather than develop their knowledge.” Effective tutoring which developed “[r]eflective knowledge-building activities involved self-monitoring, knowledge integration, and generation of inferences.”

2.2: Implications for Debate Camp

It is studies of these sorts which lead me to think that giving students the opportunity to develop and teach their own lessons is incredibly valuable. The most effective thing student teaching can provide is the chance to develop reflective metacognitive awareness of one’s own understanding. The chance to develop and teach a lesson, under supervision, is basically unparalleled in encouraging just such reflective metacognition.

A second application is to find ways to encourage tutoring between students of different levels of experience. When there is a difference in experience it forces students to come up with different types of explanations. One thing I’ve tried before, though never at VBI, is to team up with a ‘sibling’ lab. Each student in my lab would be paired up with a student in the other lab and the older student would be encouraged to be a ‘tutor’ to the younger student (meeting every day, or every other day, and teaching a concept either that the older student had just learnt or that the younger student wanted to learn). This can be easily arranged by lab leaders and often resulted in some extremely positive learning opportunities (note, the program was not wholly successful, I think there were several pairings who just never met up for tutoring).

Beyond advising student tutoring, however, these studies also tell us something important about ‘explanation’ as a learning tool. I have, since I started coaching, insisted that there is one drill more valuable than any other for getting good at framework debate. I’m also suspicious that no debater has ever listened to me. The drill is to find someone who knows nothing about either philosophy or debate (I would normally pick my youngest sister or sometimes a parent) and sit down and explain your framework to them until they can explain it back (answering any questions they may have). This was a long and laborious process; however, it was extraordinarily helpful in making sure I understood my own positions. Having to explain ideas without using any of the ‘shortcuts’ that exist in debate really pushes one to identify the gaps in one’s own understanding.

Part Three: How we Learn from Expectation Alone

Nestojko, John F., et al. “Expecting to Teach Enhances Learning and Organization of Knowledge in Free Recall of Text Passages.” Memory & cognition 42.7 (2014): 1038-1048.

3.1: Article Take-Aways

There are a couple of really interesting aspects of this article that I want to mention. First, the authors did not assess the educational impact of teaching, but instead assessed the impact of an expectation that one would teach. As the authors put it, they wanted to know if “having an expectancy to teach, without actually teaching, enhance learning, as compared with having an expectancy to be tested?”

To answer that question, they had students read through a 1,541-word passage. Some of the students were told they would have to teach the material afterwards and some were told they would have to take a test on the material. Then all the students were given a test. Interestingly the students who were reading the material to prepare to teach did better on the test than those preparing to be tested.

Just expecting that they would have to teach material resulted in students better able to recall and organize the main points of the passage.

The second interesting thing about this article is that the authors performed a second study to try and understand why students who were expecting to teach learn more effectively. What they found is that an expectation to teach is at least partially explained by improving the student’s organization of free recall information. This provides compelling evidence for the explanation that “participants expecting to teach put themselves into the mindset of a teacher, leading them to adopt certain effective strategies used by teachers when preparing to teach—such as organizing and weighing the importance of different concepts in the to-be-taught material, focusing on main points, and thinking about how information fits together.”

3.2: Implications for Debate Camp

What this shows is that some, though by no means all, of the benefits of student teaching can be gained just by instilling in students an expectation that they may have to teach the material (even if they end up not teaching it). This is very practical information. One possible application, which was used by some of the PF labs last year, is to tell students that one of them will be selected during each lab session to briefly teach the main ideas of the modules they attended. If you can guarantee that each student will have to teach numerous times throughout camp (and if they never know which day they will need to teach) it helps instill an expectation in students which will significantly improve student learning during modules.