Preventing Clicker Cheaters

Another question that came up at today’s TLT Group seminar on teaching with clickers was the question of how to prevent and/or deal with students who cheat using clickers. By cheating, I mean that in classes where clicker questions are counted as part of students’ grades (perhaps as quiz grades based on the accuracy of student responses or simply participation grades based on effort), Student A can skip class by giving his clicker to Student B, who responds to clicker questions during class using her own clicker as well as Student A’s clicker. This gives Student A credit for clicker questions even though he has skipped class. What to do about this situation?

I mentioned that short of biometric scanning, there isn’t a good technological solution to this problem, that it’s more of a classroom management issue. In small classes, instructors can usually spot a student “double clicking,” and in larger classes, TAs might be enlisted to do the same. Also, instructors can make clicker questions fairly low stakes (maybe 5-10% of the course grade) so that (a) there’s less incentive to cheat and (b) if a few cheaters get away with it, they don’t gain too much benefit from it.

I spend a couple of pages in my book talking about this issue, since I’m asked this question just about every time I speak about clickers. However, the group from the University of Nevada-Reno at the seminar today had what I think is a great solution, and it’s a technological one! They use the “pick a student” feature that their classroom response system provides to randomly select students who respond to clicker questions during class. (These students are presumably asked the share reasons for their responses with the entire class. The Reno group didn’t specify.) If a student is selected but isn’t in the room, he or she gets in “big trouble.”

This sounds like a useful approach to the cheating issue. I’m glad the Reno group shared it today! I’ve talked with a few instructors who regularly use the “pick a student at random” feature and find it extremely useful for increasing student participation in class without making students feel that their instructors are picking on them. So that’s two good reasons to try out the “pick a student at random” feature if your system makes it available to you.

How have you approached the cheating issue? How about the “pick a student” feature? Do you find it useful?

Jim says:

April 24, 2009 at 10:56 am

We had one prof here at SDSU use the pick-a-student feature regularly, initially simply as a participation tool, but one day the “picked” student turned out not to be in the room – that was an aha moment for both prof and students! (Our system only “picks” a student if his/her clicker has joined the session.) The prof shared this with other clicker-using faculty and it has become one of our standard suggestions when asked about the cheating-with-clickers issue.

One caveat: don’t leap to conclusions about the worst. Other profs have noted that sometimes roommates/friends end up inadvertently switching clickers and may not notice until something like this occurs.

One other suggestion: if some clicker questions reappear on tests, it:
– helps students better understand the types of test questions you are likely to offer,
– and, assuming you write solid conceptual-level MC questions for tests, helps ensure stronger clicker questions,
– and, ups the ante a bit in that students will realize that if they actually come to class and participate in the clicker questions, they are more likely to know what to do with similar (or even the same) questions on the test – this adds motivation to be present in class beyond simply getting clicker participation points.

By the way, UNR has a great clicker support group and their adoption trajectory parallels ours at SDSU in many ways.

derekbruff says:

April 24, 2009 at 12:19 pm

Thanks for these thoughts, Jim. I really appreciate hearing from someone who’s worked with faculty using this approach to prevent cheating. You make a really good point about not assuming the worst.

I also appreciate the positive side of this approach you highlight. I find that students really appreciate it when coming to and participating in class helps them prepare for graded assignments (like exams). The stronger the alignment is between what happens in class and how students are assessed, the better from the students’ view. That’s based on the student feedback I’ve seen from many courses.

This shouldn’t be too surprising, however! Alignment between instruction and assessment makes sense for a lot of reasons. But I think you make a great point that clicker questions can make this alignment more visible to both students and instructors. I started adding multiple-choice, conceptual questions to my exams last year when I realized that if I’m spending so much class time on clicker questions of that sort, these questions must reflect learning goals I find important. So they should be on my exams! And since I had gotten fairly good at writing these kinds of clicker questions, coming up with test questions of this sort was not hard.

Clickers in (Very Large) Economics Courses says:

August 10, 2009 at 12:02 pm

[…] I’ve mentioned before, this feature can help prevent cheating with clickers (one student responding with an absent […]

Upcoming TLT Group Online Workshop in November says:

October 22, 2009 at 1:07 pm

[…] faculty members, instructional developers, and educational technologists. (I blogged about a few good ideas that emerged from the workshop.) The feedback on the workshop was great, and the TLT Group […]

Leave a Reply Cancel reply