Reflecting on arbitrary limits in our classes
Some things in our world are artificially scarce: They are arbitrarily limited, in some way that isn’t inherent to their nature.
When you notice artificial scarcity, it’s worth thinking about it carefully. Why is it artificially scarce? Who benefits from that scarcity? Does it need to be limited? If so, could some sort of flexibility be added in a way that respects the need for limits?
Let’s talk about libraries
Artificial scarcity isn’t inherently bad. Consider due dates on library books. Due dates make the amount of time you can keep a book artificially scarce: There’s nothing inherent about the book that makes it stop working after three weeks. It’s an artificial deadline.
Why do you think this kind of artificial scarcity exists? Who benefits from it? Does that limit need to exist? Is there — or could there be — some sort of flexibility added to it?
I think there are some good reasons for this artificial scarcity. A due date ensures that I can’t monopolize a book. It benefits others who might want to read it. The deadline even encourages me to read books more promptly, and think a bit more carefully about how many books I check out. A little artificial scarcity, not terribly onerous on its own, lets the library function more smoothly and supports its role as a public institution.
A lot of libraries build flexibility into this artificial scarcity. The option to renew a book makes the deadline a little less arbitrary. New books often have shorter due dates or allow fewer renewals to ensure more people can enjoy them, while older books can be checked out for longer. While my local library still has due dates, they’ve entirely done away with overdue fines. Their explicit reason for this is to remove barriers to people accessing the library’s resources.
Wait, isn’t this blog about assessment in higher education?
A big part of my recent development as a teacher has been driven by identifying the sources of artificial scarcity in my own classes, and reflecting on why those things are limited. This has highlighted some places where I’m unnecessarily limiting things in a way that can hurt my students.
I’ve come to believe that if a resource must be artificially limited, then that limitation must be carefully considered, revisited frequently, and that flexibility should be added wherever possible.
As we saw with library due dates, artificial scarcity can serve some useful purposes. But the real danger, I think, is artificial scarcity that we accept without question because it’s “always been that way”. There are many such things in higher education.
Let’s take a look at some sources of artificial scarcity in higher education, with my key questions from above in mind: Why does this artificial scarcity exist? Who benefits from it? Does that limit need to exist, and could there be some sort of flexibility added to it?
In any class that is graded on a “curve”, grades are artificially scarce. Forcing grades to fit under a normal curve arbitrarily limits the number of each grade available in the class.
Why limit grades this way? Some common reasons for curving include increasing rigor, fighting grade inflation, allowing comparisons between sections, and “objectivity”. These reasons are founded on incorrect assumptions, such as the assumption that student learning is normally distributed within small samples, with some meaningful concept of an “average” student.
Curving is a case where I believe the artificial scarcity is wholly a bad thing. We’ve argued often on this blog that grades should directly reflect actual student learning. That requires accepting the premise that grades are not a scarce resource, and that indeed all students in a class could earn an “A” if they all demonstrate the relevant learning (I often tell my students that I would love for exactly this to happen). If we insist on grades being scarce, then that requires us to disconnect them from what students have actually learned, and curves do exactly this by making comparisons between students, rather between a student’s work and clear learning goals.
I say much more about “abundance” vs. “scarcity” when it comes to grades, as well as educational opportunity as a whole, in this post: Abundance and Scarcity, so I’ll stop going on about it for now.
Reassessments can make or break a class. I often recommend that new alternative graders put an artificial limit on the availability of reassessments. Tokens are one way to do this; a careful schedule of new attempts is another.
Why make reassessments artificially scarce? It’s primarily for the instructor’s benefit, and that’s not to be taken lightly: An unlimited reassessment policy can bury an instructor under a pile of grading and make alternative grading unsustainable.
But students can benefit from this artificial scarcity too: I often tell students that I want them to think carefully about a reassessment and not race through it. Having a limited number of reassessments can help students focus on learning from the reassessment process, although there are better ways to approach this (such as including reflective portions of a reassessment).
While a first-time alternative grader should be careful, they should also rethink their limits after gaining some experience. It’s easy to put too strict of a limit on reassessments, making the instructor comfortable but unnecessarily burdening students. If a limit is actively stopping a student from showing their real understanding, then it’s worth finding ways around that.
If a first attempt at a reassessment policy worked out great, perhaps the number of tokens, or frequency of reassessments, can be slightly increased. Or, students can have ways to earn more tokens (perhaps they could even earn tokens for completing an early reassessment or completing grade reflections, thus reinforcing learning-focused actions). Alternately, there can be flexibility within hard limits: I allow my geometry students to revise only one proof per week, but they can submit multiple revisions of that proof within the week, leading to some very useful back-and-forth conversations. If a class allows reassessments on a regular schedule (e.g. new attempts appear on quizzes or exams), perhaps students can have a say in which new attempts are included. One instructor I interviewed for our book uses a “Build-your-own final exam” in which students can request that their own personal final exam includes up to four standards of their choosing.
Due dates or deadlines for assignments can be another form of artificial scarcity. Due dates are generally arbitrary, since students could certainly complete assignments — and show excellent learning — at different times. (Here I’m thinking about deadlines within the instructor’s control; the end of the semester is an external constraint that’s also a sort of artificial scarcity, but not one most of us can do much about.)
But the scarcity also has purposes, both for the instructor and for students. For instructors, deadlines make grading easier and more consistent, and avoid the cognitive pain of having to constantly switch between different assignments.
Due dates can help students as well, by providing structure to work within. This is a very real benefit: A clear and well-communicated course structure helps students focus on learning. But much like reassessments, this can go too far: Due dates can actively prevent a student from demonstrating their learning. I’d much rather see what a student has learned than enforce an arbitrary deadline.
So, as with reassessments, we can look for ways to be flexible about deadlines. Tokens that allow a deadline extension are one option. Alternately, some instructors accept student work up until they start grading (important: tell students this! — or else this kind of policy punishes students who believe, quite reasonably, that due dates are final).
Deadlines are perhaps the artificial limitation where I’ve seen the most change in my own practice over the past year. I used to use “grace days”, effectively single-purpose tokens that allow a deadline extension with no excuse needed. Last year, inspired by several people I follow on Twitter, I started using a truly unlimited deadline extension policy. Here’s my syllabus language (copied from Spencer Bagley):
A word about due dates
Here is how due dates work in the workplace: They exist, and they’re important. However, there’s a certain amount of flexibility. If you need a little longer on something, you ask whoever set the deadline if you can have some more time. This is usually not a big deal, but if it happens a lot, or if you miss deadlines without letting anyone know, people will start asking you if everything is OK.
That is also how due dates work in this class. If you need to update the due date on anything, just fill out the due date change request form (there’s a link in the left menu on Blackboard). It’s automatic: As soon as you’ve sent the form, you can use the new due date for that one time. You’ll only hear from me if I have questions.
If you ask for lots of changes, we’ll work together to find ways to help you keep up. Note that you may not get timely feedback if you need to change a due date.
The “due date change request form” (an idea taken from Drew Lewis) is a simple Google Form that asks students for the name of the assignment, their proposed new deadline, what they need to do, and what they need from me. I get an email any time a student submits a request. It’s simple and automated.
I’ve used this system in two different classes — sophomore and junior/senior level — and have been very impressed with the results. On the one hand, I see about the same amount of late work as I had when using tokens or even “hard” deadlines (with required excuses for late work). On the other hand, students consistently suggest shorter deadlines that I would have allowed. Some students still forget to fill the form out, or miss their proposed deadline, but that is no different than under any other policies. I follow up with them as promised in the syllabus, and that leads to some useful conversations.
The difference in student attitudes is huge. Removing artificially scarce deadlines is a very tangible expression of my trust in my students. Students understand that I really care about their learning, and I’m not interested in playing compliance games. Could somebody abuse this? Sure, but that’s no different than any other policy. I consider an improved atmosphere of trust and focus on learning to be an excellent result in exchange for roughly the same amount of deadline trouble.
… and so much else
There are many other sources of artificial scarcity in higher education. Some are within the control of individual instructors; others reach much deeper into the institution, such as policies that limit the number of classes that can be taken credit/no-credit, deadlines for dropping classes or getting refunds, or the number of spots available in secondary admissions programs.
These artificial limits exist for some not-entirely-terrible reasons, often having to do with the bureaucratic convenience of massive institutions that have to deal with thousands of students.
Artificial scarcity is not a black-and-white, all-or-nothing issue. My point here isn’t to argue that our classes should have no limits whatsoever. It’s also not to argue that limits are critical and must be enforced at all costs. Rather, my point is that it’s worthwhile to look for artificial scarcity, to examine the things that have “always been that way”, and when we find it, to think carefully about why it exists, who benefits from it, and how flexibility could be added. The results can be more subtle, more carefully tailored to the needs of everyone, and more beneficial too.
Once you start to view the world with these questions in mind, things start to pop out. Where else is there artificial scarcity? How could it be made better?
Thanks for reading Grading for Growth! Subscribe for free to receive new ideas on assessment in higher education, every Monday, right in your inbox.
And they’ve removed late fines without causing major problems people sometimes predict. From my local library’s FAQ on removing due dates: “Studies have shown that library fines do not impact the return time of material.” Keep that in mind when we talk about deadlines on student work.
Also, not everything that’s limited is artificially scarce. Some examples: Seats in a classroom are inherently limited by space. Hours in the day are fundamentally limited, and that limits what instructors can do. The end of the semester may be an artificial deadline, but it’s not a limit that instructors can generally escape from. But there are plenty of things that are limited, but not inherently so. That’s what I’m focusing on here.
If you haven’t encountered “curving” as a current and very real phenomenon, check out this article, by two big names in Chemistry education, published just last year: “Grading on a Curve as as Systemic Issue of Equity in Chemistry Education.”
Robert wrote about the issue of “real world” deadlines, and whether deadlines in higher education prepare students for the “real world”, in some detail earlier this year on his personal blog.