Before March 2020, I was not a very flexible teacher in terms of course policy. I was quite rigid on deadlines unless there were emergencies and exams were “one and done.” After frantically creating and teaching an online course for spring term 2020 where I tried to be very flexible to accommodate the wide range of home situations faced by my students, I was exhausted from managing the numerous emails and messages on Slack asking for extensions. I needed to find a better balance between increasing flexibility for students and the amount of time I spent negotiating logistical details. That summer, I read *Small Teaching Online*, which is where I first heard about the idea of using “Oops tokens.” I have now used tokens for three terms, and plan to continue using them for all of my courses.

**How I use tokens**

The idea of a token is to create a system where students understand how much flexibility you have already incorporated into the course, while also knowing the limits to that flexibility. Initially, I only gave students four tokens at the start of each course. I rather arbitrarily chose that number, thinking that there would be nine assignments and roughly 21 exam questions, so it somehow “felt about right.” I soon realized that four tokens was not quite enough for the majority of students, because they would ration their tokens rather than using them as intended. I now allow students to earn up to three additional tokens, and have seen far fewer students feel like they need to ration them.

Students can spend a token in my classes in two ways:

**Option #1: **They can spend a token to receive a 48-hour no-questions-asked extension on an assignment. **Option #2: **They can spend a token to rewrite an exam question (where the score on that new question replaced the score based on the original question). The form of this exam “rewrite” could either entail making corrections to the original question or attempting a new question on the same topic. For example, here is the “original” exam question and the “revision” question on maximum likelihood estimation from my statistical inference course:

Original: An experimenter has reason to believe that the PDF describing the variability in a certain type of measurement is given by:

for 0 < y < ∞, and 0 otherwise, for θ > 0. Assume that you have a random sample, Y_{1}, …, Y_{n}, from this PDF. Find the maximum likelihood estimator of θ and argue that it does indeed maximize the likelihood.

Revision: Below is the likelihood function for θ based on a random sample, X_{1}, …, X_{n}, from a PDF where 0 < x < 2 and θ > 0. Find the maximum likelihood estimator of θ and argue that it does indeed maximize the likelihood.

To manage the bookkeeping of token-spending requests, I use a Google Form that students fill out in order to “spend” a token. In addition, I track the number of tokens each student has remaining as an ungraded column in my learning management system (i.e., Moodle) gradebook. Instead of managing email chains or Slack messages, I update the token counts each Friday. (I warn students in my syllabus to keep track of their totals in real time, since I will only update the counts once a week.) Batch processing the tokens spent via the Google Form is a time save on my end, and is easily tracked by exporting the results of your Google Form to a spreadsheet.

I highlight rows that had been processed, making it easy to track what requests had come in over the past week (though I’m sure there are other ways to do this!).

To earn the three additional tokens in my course, students can complete tasks that I view as beneficial to their learning, for example:

- coming to student hours (also commonly known as office hours) and introducing themselves,
- asking a substantive/reflective question (i.e., a question about a topic, not simply “I don’t know how to do this”) more than 24-hours before a homework deadline, or
- reflecting on major misunderstandings on homework or exam questions and filling out a “mistake report” Google Form (adapted from a form Robert Talbert uses). This form helps students reflect on their learning process by having them identify a misunderstanding/point of confusion from an assignment or exam and reflecting on how they addressed this issue, including how my (or my grader’s) feedback was used.

I add an additional column to my learning management system gradebook for tokens earned.

**Benefits**

Overall, students appreciate the token system, though many look at me like I’m crazy when I introduce the idea. More specifically, students appreciate (and comment on) the ability to receive a few no-questions-asked extensions for those particularly stressful times of the term.

From a teacher’s perspective, I enjoy giving students a limited number of revisions on exam questions, avoiding common pitfalls with one-and-done exams. I also appreciate the fact that they can’t revise every question, which still provides an incentive for students to keep up with the material. Additionally, I receive fewer emails and Slack messages asking for extensions. There are still emergencies to deal with, but this seemed like an efficient system.

In the “real world”, students will experience a limited amount of flexibility (outside the academic bubble), so this system reflects that. You can probably receive an extension on a deadline or two, but eventually you have to do your job or there will be negative consequences.

**Challenges/pitfalls**

Students face two primary challenges as they adjust to this system. Some students feel the need to “hoard” their tokens in case something goes horribly wrong; other students spend all of their tokens revising the first exam. These challenges haven’t deterred me from using tokens, but have made me think about how to discuss them with students early in the course. Additionally, some students don’t track their token totals as carefully as they should, and forget that the token column isn’t necessarily their true “balance.” There were a few times where I had to tell a student they could not submit all of their revisions for credit since they had run out of tokens. This was a rare occurrence, and the time savings of batch processing outweigh the distraction of constantly updating the gradebook.

One challenge that I faced when using tokens was remembering to update the token counts. That doesn’t sound challenging at first, but I was slow to do this a few times before I added it to my calendar. I also found that choosing the right number of tokens is tricky, and requires some trial and error. I’m still not sure that earning up to seven tokens is the magic number for my ten-week term, but it’s closer than four. If I were only granting extensions with tokens, then two or three might be about right, but since I also allowed them to be spent on revisions of exam questions, the number needed to be higher.

**Acknowledgements and additional ideas**

I want to reiterate that the idea of using tokens to add flexibility to my courses is not something I devised, and that I borrowed ideas from a variety of sources. In addition, Linda Nilson talks about the token system as being a fundamental component of specifications grading, helping students adjust to the rigor of the new system by adding flexibility. (See Eric Reyes’ post for his take and his syllabi.)

In addition, I have seen other people use tokens that students exchange for different types of flexibility in a course, such as

- An oral exam question revision rather than a written revision.
- Drop an unsatisfactory homework grade
- Increased flexibility in attendance policies

Other resources that helped me implement this system include:

*Small teaching online*- The care and feeding of tokens by Robert Talbert (this is where I got the token spending form template)

Here’s a token to get you started in your classroom.

## Leave a Reply