Sunday, 19 August 2018

Creating a market for in-class extra credit

I began offering extra credit in my ECON110 (now ECONS102) class in 2012. There were two reasons for this. First, as with many first-year university classes, my class had suffered from declining attendance over the previous several years, and offering extra credit for completing certain activities in lectures seemed to be one way to arrest the decline in attendance (and it worked!). Similarly, we had introduced marks for tutorial attendance in our economics papers in 2007, and that demonstrated the potential impact on attendance. We introduced extra credit for random spot quizzes in ECON100 (now ECONS101) in 2016, and it had a similar effect on lecture attendance (at least, until the current semester where attendance has been surprisingly low). Second, I already used some class time to conduct extra activities such as small experiments or surveys that I would use to illustrate various economic concepts, and it made sense to reward the students who provided the data for those illustrations.

As you can see, extra credit has been part of my practice for several years, and is now being used by several of my colleagues as well. So, I was really interested to read this 2016 paper by James Staveley-O'Carroll (Babson College), published in the Journal of Economic Education (ungated earlier version here). In the paper, Staveley-O'Carroll describes the creation of a market for extra credit, where students can buy and sell extra credit, which he has been using in his classes:
The EC [extra credit] generated by answering clicker questions, as described in this article, offers an inexpensive alternative reward to create a realistic market with properly aligned incentives. By answering questions in class, students create EC. The EC, however, is not transferred directly to the student who created it. Instead, a market system is implemented to price EC and allocate it to the students who desire it the most. This market can be manipulated by the instructor to give students hands-on experience with many economic topics such as inflation expectations and game theory. Moreover, extensions of the experiment allow the instructor to add a financial intermediary, bonds, and stocks to the basic market framework, expanding the range of topics to risk aversion, hedging, and peer-to-peer lending.
This system combines the incentives associated with extra credit with a real-world market mechanism that the teacher can use to illustrate some of the concepts being developed in class. There are many pit market experiments that teachers already use, so this provides an interesting extension of those. This sounded pretty exciting to me. However, as I delved deeper into the paper, I realised that it was very practical for the class sizes that Staveley-O'Carroll was dealing with (between 7 and 30 students), but would soon become very unwieldy without some serious back-end automation in my classes (with up to 320 students).

On top of that, the mechanism by which the extra credit currency (Gronks) converts to extra credit marks relies on balancing the supply of extra credit marks and demand for extra credit marks for each piece of assessment, to determine the price (in Gronks) for each point of extra credit. I could see that would require a lot of explanation for some students to get their heads around. So, while it sounds like a pretty cool idea, it might be something to park for classes at a slightly higher level than first-year.

Finally, the article doesn't provide any evidence that the extra credit market improves student understanding of economics. Nevertheless, Staveley-O'Carroll does find that:
...student feedback has been almost unanimously positive, with many students noting that the in-class currency is the best part of the course.
So, at least the students enjoy it!


  1. The days of being rewarded for attending are gone, this is an academic course and the fact that simply attending increases your grade (reflecting your competence) is a joke.

    1. People are rewarded for attending work. It seems to me that NOT being rewarded for attendance is the outlier.

      Attendance turns out to be highly correlated with grades, so the overall grade including extra credit reflects competence to the same extent as the grade without extra credit does.

  2. Simply attending a job and not doing anything will not earn people any reward.

    Also, correlation doesn't equal causation, and if attendance and grades equally reflect competence, why bother trying to challenge the system? Let the grades speak for themselves.