One thing that faculty really struggle with is whether or not, and how much, to give students credit for their clicker question answers. You want to give students some incentive to participate, but grading opens a whole can of worms. One of my faculty workshop participants explained the dilemma very astutely:
“If I do not assign “grades” or “scores” to clicker activities, many students will not participate in the clicker activities. Only the motivated students participate and the “feedback” I get from the responses are skewed towards high-performing students. Consequently, my goal of reaching poorly performing students becomes thwarted.
If I assign “grades” or “scores” to clicker activities then I worry about students operating multiple clickers so their friends will get good marks without having to attend class. I think this would be frustrating to students who “play by the rules” and are still not doing well in the class.
If I back off a bit from “grades” and assign “points for participation” to clicker activities, in addition to the multiple clicker concern many students will click random responses just to “prove” to me they are participating in the discussions. In this case the real-time feedback is useless to me. Occasionally, it turns into a game and a few students intentionally utilize their clickers in a way that brings about the biggest reaction (laughter) from the class. I think this would be frustrating to students who really want to learn.”
That comment outlines three different ways to think about grading: No credit, credit for correctness, and credit for participation.
Points affect student conversations
So, how do we think about the best way to grade clicker questions? First, I want to share the results from a very pertinent study from 2006, “The effect of grading incentive on student discourse in Peer Instruction,” (James, Am. J. Phys., 74(8), 2006). This researcher did something that is really challenging to do, both in terms of time and logistics – he actually listened in to a wide variety of clicker discussions in introductory astronomy courses at his institution. He categorized student discussion as to what was happening, such as whether they were stating their answer, posing a question or idea, disagreeing with their partners’ idea, etc.
In one class, the instructor’s grading scheme emphasized the right answer; the clicker score was 12.5% of the student grade, and incorrect responses got only 1/3 credit. Note that this scheme is similar to what I see a lot of instructors use. We’ll call this the “correct-answer grading” scheme.
In the other instructor’s class, the clicker score counted for 20% of the course grade, and incorrect and correct responses counted for the same amount. We’ll call this the “participation-credit” scheme. (Note that James calls these the high-stakes and low-stakes classes, but I disagree; they both count pretty heavily towards the student score, but one emphasizes the right answer).
Dr. James found out that in the “correct-answer grading” class, student conversations were much more likely to be dominated by one member, usually the one who was more knowledgeable, and most of the conversation focused on that student’s answer choice. In the participation-credit classroom, however, the conversations were more balanced.
Additionally, in the correct-answer grading class, students were more likely give the same vote as their clicker partner. That suggests that the instructor in that class is getting results that are misleading as to how well the class actually understands the question; students are more likely to vote with what they think might get them points, as opposed to giving the answer that reflects their actual thinking.
I’ll note that this is a small study, and the comparison is between two different instructors and two different sized classes, but the results have been demonstrated in different types of courses in a second paper by the same author. In that paper, additionally, they found that high-stakes classrooms led to more passive conversational styles.
So, if we give a lot of points for correctness, it’s likely to shut down student conversation, and mislead us as to the level of student understanding.
Points affect student motivation
Another thing to consider is that extrinsic rewards (i.e., those that arise from something outside of yourself) are less likely to lead to intrinsic motivation. I’ve been reading some educational psychology, and apparently they warn strongly against giving points for students to engage in something, as they’re less likely to personally value the activity. However, they do recommend giving points or other rewards if the teacher thinks that students are not likely to engage in the activity otherwise. For some of our students, that may be the case, but I’d argue it’s hard to know for sure without trying it.
So, how do we encourage students to participate? I personally feel that a combination of intrinsic and extrinsic motivation might work best. Extrinsic motivation is the sense that you are doing something for some external reward (e.g., points). Intrinsic motivation is the internal sense that this is something that is important for your own learning. See my previous post on Getting students on-board with clickers for some ideas on creating buy-in. I think this can address some of the concerns about students playing a game with their clickers, and not really engaging with the questions. If the questions are interesting, and they understand that the discussions matter for their learning, and there is a culture of participation in the class, then most students should engage.
A possible compromise
What we do at Colorado is to offer, for example, 2 points for participation, and 1 point for correctness, so that there is a little sense of accomplishment for getting the right answer. But then the clicker questions only count as extra credit, offsetting a poor homework or exam score. Some instructors choose to make it more like 10% of their course grade, but I’m a bit nervous about this approach for the reasons discussed above.
In the end, you probably want to choose a grading scheme that works well for your particular pedagogical approach. For example, in our physics courses, it’s very important to us that students deeply engage in conversation about these difficult physics concepts, so we have this mixed correctness/participation scoring system which works well for that. I have another physics colleague, however, who strongly believes that students should only be intrinsically motivated, and he has developed a pedagogical approach that repeatedly asks students to examine their own learning. He’s a master of this, and I think most of us would aspire to that approach, but it might not work for all of us, or not right at first. Really, this is all about student motivation; I highly recommend this short white paper summarizing the research on student motivation.
But one thing that is clear to me is that it would likely be quite detrimental to only provide points for correctness.
Please weigh in on the comments; what types of grading schemes have you seen work?
(This is a repost from an article I wrote for the iClicker blog).