Promoting student growth with engagement credits

Possibly a better way to handle attendance, participation, and other forms of engagement.

Photo by Mathew Schwartz on Unsplash

Perhaps the biggest problem facing instructors today is how to get students engaged in their class work. What does “engagement” mean, and how might we elicit it? 

The term “engagement” is often used as a combination certain behaviors; of feelings, beliefs, and emotions; and things we do to elicit all these. It’s a bit amorphous, which makes it hard to say or do anything meaningful about "engagement". It’s helpful to look at the etymology of the word: “Engage” derives from an older word meaning "pledge", as when a couple gets engaged to be married. Other words that derive from the same root are mortgage and wage, pointing toward toward "a state of committed involvement, in an activity into which we have entered willingly and with the intention to complete"¹.

What many have been seeing lately with increasing frequency, is disengagement, which manifests as absenteeism, being intellectually "checked out" even if physically present, opting out of required readings, and more. You can sometimes see disengagement on the faces of students, and you know there's more going on than meets the eye.

We can't make students be engaged; we can only create an environment where engagement is a more compelling choice than disengagement. Teachers have been trying to do this for over a century using a predictable model: Attaching points to things like attendance and participation, and make it part of the course grade. This approach is problematic for various reasons which I'll discuss below. In this post, I want to describe a simple alternative I've been using for the last few years that seems to work well, but avoids the various traps of points-based approaches. It uses a notion I call engagement credits.

Why the traditional approach is problematic

The traditional approach to eliciting any sort of behavior from students in a class is to place a numerical value on it and make it a variable in the course grade formula. Engagement is no exception: Typical policies intended to promote engagement boil down to tallying attendances and making the number a percentage of the course grade; or taking a purely punitive approach by stating that students with a certain number of absences will have their grades lowered, or possibly even automatically fail the course².

This approach is problematic for at least three reasons:

• It introduces potential inequities into the course. As David wrote here, channeling Joe Feldman, participation and attendance can be considered "non-academic factors" and their assessment is susceptible to instructor bias. They often reflect a student's freedom from personal constraints such as family, health, and financial issues which would make attendance and participation difficult, and some students -- through no fault of their own -- have more such constraints than others.

• It conflates attendance and class participation with "engagement". Attending and participating in class are two forms of engagement, but not the only two. A student who fills out a survey, interacts on a discussion board (above and beyond, perhaps without, a requirement to do so), or works through an optional online practice problem set is also engaged with the class and the learning process. And yet, attendance and participation often have pride of place in course syllabi. I suspect this is because they are easy to quantify and therefore instructors will tend to optimize the notion of "engagement" around them.

• Goodhart's Law. This states that when a measure becomes the target, it ceases to be a good measure. Goodhart's Law haunts traditional grading generally speaking, but when applied to "engagement" – when we attach points to it and make it part of a grade –  we actually get disengagement. Students will go through the motions and appear to be engaged but actually they're just grinding (in the video game sense) and are not really present with their work. 

One response to the problems of traditionally graded engagement is to completely exclude it: To have attendance, participation, or other related forms of engagement have no place in the course grade and focus only on evidence of learning. I used to take this approach. I would record attendance but not grade it or include it in the course grade. Instead I used the data it as a means of keeping tabs on students and checking in with those who had numerous absences. 

But recently, I ran a simple correlation between my students' attendance rates and their final exam scores and found a very strong positive correlation (the p-value was on the order of 10^-5). Correlation is not causation, but this result confirmed my belief that engagement, including attendance and participation, are not “non-academic” factors at all but in fact play an important role in the evidence of learning students produce. So it has a place in the grade students get in my classes. The main question is how to handle engagement, in a way that avoids traditional traps.

Enter engagement credits

Back in the day, I played Dungeons and Dragons³. In that game, and most other role playing games, you have the concept of experience points: Points that you earn by doing relevant and useful stuff like killing dragons, casting spells, and completing quests. As you accumulate points, you eventually earn enough to level up and become a more powerful character.

One day not long ago, I thought: Students should earn experience points for doing successful "quests" in my classes. I even drafted an entire grading system based on nothing but accumulation of "experience points" which could be done in a wide variety of ways, including the usual stuff like taking exams and coming to class but also alternative activities like oral quizzes in office hours, creating a poster presentation, etc. These all carried experience point values, and the results all went into a single pool of points that you could use to level up through different grades.

I never fully fleshed that system out; others (example, example) have done so, and probably better than I would have. But a vestige of the idea made it into my courses, namely the idea of earning credit for doing any relevant, useful instance of "engagement" as defined earlier. I called them engagement credits.

It's almost embarrassingly simple how this works:

• In the course design process, I picked out some standard "acts of engagement" such as attending a class meeting, doing satisfactory work on a pre-class assignment, and working through an optional online homework set. In the syllabus I attach a point value to each thing. The standard is 1 point for class attendance and 2 points for completing a class prep⁴. Other items were more valuable. For example, completing the "startup assignment" (described here) was worth a whopping 30 engagement credits. Overall, I scaled the point values so that earning 100 engagement credits would be easy if you just came to class and did the pre-class work.

• But importantly: Other opportunities to earn engagement credits constantly came available. For example, completing a Five Question Summary survey earned 5 points. Last semester I found this article on deliberate practice, and I really wanted students to read it, so I made a Google Form for responses and awarded 10 engagement credits for good faith responses. Students were welcome -- encouraged, even -- to create their own engagement credit opportunities. So, if you had issues making it to class or completing pre-class work, or if you just didn’t want to do a particular task, you could just engage in some other way. 

• In my syllabus, a total of 90 engagement credits was needed for a grade of A; 80 for a B; 70 for a C; and 50 for a D. For perspective, the class had about 30 regular meetings; sp doing the class prep work and then attending the class would therefore get you to 90 credits automatically. But we also had the rule that if you met all the requirements for a course grade except engagement credits, you earned that grade but with a "minus" penalty; having major issues with engagement wouldn't hurt the course grade badly as long as the "academic factors" of the grade were going well.

My students have found engagement credits useful: They provide a light pressure to complete “quests” that get and keep them engaged, without turning it into a stressor or a mindless game of point-grubbing. The fact that there are many ways to earn the credits, including activities of their own design if they want, makes the engagement more real than if we limited it to just attendance and “participation”. 

In day-to-day practice, I have to do a fair amount of daily accounting work since engagement credits tend to happen through small but frequent activities. Much like tracking your checking account transactions, if you don't take time every day to record engagement credits, it's easy to fall behind and hard to catch up. This semester I set a goal of consistently recording attendance each day and used a habit tracker to help. It took some discipline, but in practice this took no more than 10 minutes a day and produced a lot of valuable data (see below).

The course LMS, for a change, can be really useful here. In Blackboard and probably in other platforms, gradebook items can be given a tag or category. I created one for engagement credit-bearing items and then a separate column that kept a running total of all gradebook items with that tag. This way, all I had to do was stay consistent with entering the individual items, and students could check their totals at any time.

How it works overall

I have no formal data on how well the engagement credit approach has worked, but it passes the eye test in my view. Compared to classes where engagement wasn't a part of the course grade at all, the classes that use engagement credits seem to have levels of participation and attendance that are, if not perfect, then consistently reasonably OK. (I don't have the experience many of my colleagues have, of week after week of rampant absenteeism for example; my attendance rates this semester hovered between 75% and 85%. Again not outstanding, but consistently reasonably OK.)

Using engagement credits has forced me into better habits as an instructor, especially the habit of tracking attendance. Attendance data are extremely valuable, especially if you as an instructor are concerned about absenteeism. They can alert you to students who have missed several days in a row, for example. And I was able to use attendance data after the course was over to do useful statistical analyses, for example the aforementioned correlation between attendance and final exam scores⁵.

Overall, I find engagement credits to be a good compromise between heavy-handed points-based approaches to engagement, and not including engagement in the course grade at all. They are easy to implement and cost nothing. They could even be implemented partially in a course currently underway or in a course that uses traditional points-based grading. Most importantly, they encourage the engagement we seek and open the door for conversations about why this is so important for learning.

1. This quote, and a more detailed look at what “engagement” means and how we might study it, are from a paper I co-wrote in 2019 with Anat Mor-Avi about active learning spaces. It’s freely available at this link.

2. Here, one might counter by saying that engagement is assessed in many ways in a class, not just attendance and participation, for example by requiring out-of-class readings or homework. One could argue that the only thing in a course that we assess is engagement. I’ll leave that discussion for another time; here I am focusing only on those forms of engagement with a class, that are not directly engagement with class concepts (i.e. learning the material).

3. Incessantly. And, this was the early 1980s, in the South, during the good old days of the Satanic Panic. My Southern Baptist parents were absolutely sure that I was conjuring demons and sacrificing virgins, or something, and headed straight to you-know-where.

4. This sends the signal that I find preparation for class twice as valuable as attending the class for which you prepared. Which I think is accurate, and also equitable since completing a class prep does not require a student to be physically present at a particular location at a particular time of the day. However, I am considering making them equal in the future since attendance is really important in my view.

5. I am really surprised by the number of faculty I know who seem very upset by recent trends in attendance and absenteeism, and yet they don’t collect any data on attendance in their classes. Doing so is as simple as passing around a sign-in sheet or running a clicker question each time your classes meet. If attendance is that important, why not collect data on it? And how important is attendance to you, really, if you don’t? The data can help us understand and perhaps do something about the so-called “disengagement crisis” we face. For example, it’s often claimed that students with low high school GPAs are the ones skipping class. But has anybody tried collecting data on attendance and running a correlation on absenteeism rates and GPA?