Last spring, I did a short presentation (on Mazur’s peer instruction) for an event at the University of Chicago’s Center for Teaching and Learning, and afterward, I was recruited to become a Teaching Consultant. As a Teaching Consultant (TC), I am one of a handful of graduate students from across the university who have the opportunity to get together and read and talk about pedagogy. We take the knowledge we glean to help other graduate students in their classrooms through workshops and individual consultations. Yesterday, we met to have a discussion about discussions.
Put that way, the topic of conversation is somewhat amusing (especially when you consider that we tried out the technique of leaderless discussion, during which I had to stop the leaderless discussion to discuss how we were discussing – I think we almost drowned in the recursion). Still, it was an interesting meeting, particularly the input from our TCs in the STEM fields. At first blush, you might think that there is no opportunity for students to discuss in a discipline like computer science. The TC described his class, and described the “discussions” as an opportunity for students to respond to questions and also ask for clarification. In other words, the “discussions” were primarily led only at a level that might help the students to remember and understand their readings, but not achieve the higher levels on Bloom’s taxonomy of learning domains (i.e., application, analysis, evaluation, creation).
However, as the conversation continued, the TC also described how students occasionally are given a scenario: they need to build an efficient program to do x, how should they go about it? The TC said that frequently, there is an answer that is considered more or less “correct” or the most efficient, but that the students are given free rein to try to figure out that path themselves. In these discussions, then, the students are most definitely applying their lessons, and certainly evaluating, if they are trying to weigh the best route along which to write the program – and if they actually write the program, you could argue that they are even being creative. Thus, although the discussions at first blush look quite different –it’s not 20 students sitting around a table and discussing Kant or Arendt, for instance—they function in a very similar way as our discussions in the Social Sciences Core at the University of Chicago.
I suppose a devil’s advocate might argue that the instructor in a computer science course does not have to prepare in the same way that we prepare for a long discussion on texts. However, I would still reply that it’s a difference of context, and maybe degree, but not of kind. The instructor in the comp sci course still has to figure out the right questions to ask in order to achieve the desired result (e.g., a better comprehension of the task and issues at hand). And that’s a large part of the art of discussion leading: finding and asking the right questions.
Revision of Bloom’s taxonomy of cognitive domains (Anderson & Krathwohl, 2001).
The type of questions that you ask depends largely on the learning outcome that you seek. If you want retention of factual data, you can ask factual questions. If you want the students to understand the material rather than potentially parrot it back at you, then you have to ask questions that require more comprehension: can you explain x? Can you give alternate examples of x? Moving up Bloom’s taxonomy to the higher learning domains, if you are training the students to analyze and evaluate their material, then you need to step up your questioning: Break down the author’s arguments that led to her conclusion – what do you think the most crucial step is, and why? How internally consistent is the argumentation? If the outcome of x phenomenon were different, what would be the implications for y?
In a class like I teach in the Social Sciences Core, where we focus on teaching the students skills such as analysis and evaluation, such questions as those at the end of the last paragraph are frequently the most fruitful for building those skills. However, learning is a process, and college students – any students, really –can’t be expected to jump from 0 to 60 in x seconds. So, usually, a combination of questions becomes necessary, but I find it particularly helpful to keep the end goals in mind. In my case, that’s usually skills-building.
The students, however, frequently don’t see to that point of the future. Instead, they see the more immediate goals: they need to learn the material to write the papers to get the good grades to pass the course to get the internship to….In other words, they think content is king. What they don’t see is that writing those papers – the very next immediate step—in fact requires them to learn those skill sets, to master those learning domains, on Bloom’s taxonomy.
In preparing the discussion for the teaching consultant meeting, we talked about this myopia and how to get students to understand that their seeming low-stakes project (i.e., having a fun, easy, entertaining discussion in class) has everything to do with the higher-stakes assignments like papers. How do we help the students see the connection between the low-stakes assignment of discussion – an assignment that builds skills, even though the students seem unaware of that construction – with the assignments that they are more transparently graded on?
Thus, there are multiple issues that a leader of a classroom discussion must address: she must learn to ask questions that can be directed to various learning outcomes, and she must help the students see the effects of their discussions, both on themselves and their skills and on their understanding of the material at hand. Determining the best questions in a particular context takes practice, and a little bit of art to make sure that you’re getting through to your particular group of students. Figuring out how to connect the low-stakes assignments and the high-stakes projects like papers is perhaps also an art—you are required to deploy different assignments (minute papers; small group projects; content questionnaires, etc.) to help the students make the connection, and once again, you have to determine which assignments will work best with your specific class.
Why do I classify these two issues with discussion leading as artful, rather than scientific or even procedural? I think I prefer to consider teaching an art, because it means that no matter how many times I teach the same class, there will never be one “correct” formula to use to run the class, craft assignments, and lead discussions. Instead, directed by the outcome I seek, I can turn to any number of possible assignments or a plethora of questions, and put them together in a meaningful, artful way.
Reference cited:
Anderson, L. & Krathwohl, D. A. (2001) Taxonomy for Learning, Teaching and Assessing: A Revision of Bloom’s Taxonomy of Educational Objectives New York: Longman.
Additional reading on discussions (good techniques and tips, although centered around ideals particular to democratic republics):
Brookfield, S. & Preskill, S. (2005) Discussion as a way of teaching: tools and techniques for democratic classrooms. San Francisco: Jossey-Bass Publishers.