education, mastery-based grading Zachary Diamond education, mastery-based grading Zachary Diamond

Upgrade your Gradebook

“My gradebook is more of a progress tracker than a record of each student's performance, and as such it serves more roles than just record-keeping”


"Simplify, simplify"

- Henry David Thoreau


In keeping with the grading theme from my last post, I'd like to delve a little further into the structures I use to keep my assessment as objective as possible and eliminate, the personal, emotional connection that students often perceive between their grades and their identity. I've spent many, many hours discussing this topic with my coaches and supervisors and trying to discover new and better ways to grade, and I think that's because, quite simply, grading is hard. As is usually the case, I've found that the best solution to the challenging problem of grading isn't to work harder, but rather to adopt an entirely different approach - to change the underlying structure of my gradebook and my grading scheme. Before diving into the solution (which is to use binary, mastered/not mastered grades on single tasks arranged in a sequence; a process strongly influenced by the Modern Classrooms Project, of which I am a fellow), let's explore what makes grading such a difficult process.

When you grade something, your task is to examine a piece of student work and, from it, determine objectively how much the student has learned of the topic at hand. There's a lot to unpack in this seemingly simple task: we must find a way to quantify learning in a way that applies to all our students, but each individual student brings so much to bear on each task that it's nearly impossible to come up with an objective heuristic, rubric, or measuring stick of knowledge and learning that can be calibrated to produce a true and fair grade in every individual case. Furthermore, if teachers haven't thoroughly considered and designed a system for objective grading, they may not be able to assess work without unconsciously surfacing other considerations. In my first few years, when I was presented with a piece of student work, I would consider the student's personality and how they may react to the grade I gave; I would consider the student's previous performance and whether this submission was an improvement or a decline; I brought to bear unexamined biases and even prejudices (that, frighteningly, may have been borne out in the patterns I saw in my gradebook demographically). While none of these things are on the rubric, they would pass through my mind as I graded, and I felt that to be "objective," I needed to learn to just ignore them, which would create a sort of cognitive dissonance that made grading a difficult and stressful series of decisions I had to make while factoring in some considerations and discarding others.

Now, there are tools and techniques for dealing with this by objectively describing whether and how the piece of work conforms to certain standards (rubrics are one of the best); but even so, there is always ambiguity in assessing the quality of a product if there are multiple levels that we have to choose between (especially if some aspects of the work conform to one level, but other aspects conform to another level, lower or higher). Grading this way is mentally exhausting and time consuming - I teach between 160-200 students whom I see every two days, so daily grading means putting myself through this intellectual and emotional wringer 90 times a day, which was not something I could conceive of as a young teacher (it was one of those "ok, but how on Earth do you actually do that" questions). And I was right - without careful planning and an intentional approach, daily (or even weekly) grading is not sustainable; neither the world's greatest rubric, nor any amount of hard work, planning, or time management could have made this possible without burning me out. There is a better way, and, unsurprisingly, the solution is not to grind harder, nor is it to learn to ignore the conflicting thoughts and feelings that swirl around as we seek to land on a perfectly objective grade; rather, we need lean our ladder against an entirely different wall to approach formative assessment in an unambiguous way. The solution is progress tracking.

Through the self-pacing structures of the Modern Classrooms Project, I've learned that the best way to address to this problem (and the key to in-class grading) is to limit my formatives to a simple task that requires mastery of one single skill or piece of knowledge. Importantly, the task must be so simple as to negate the need for a grade at all. Rather, a simple "mastered" or "not mastered" must suffice to determine whether the student can move on to the next lesson or assignment, and there is no ambiguity as to the quality of the product - either it's done or it's not, and no rubric is involved because the task can only be successfully completed if the student has mastered the content.

My upgraded gradebook, therefore, is an ordered sequence of binary switches, of 0's and 1's, and when a student submits an assignment, it takes only a few unambiguous and stress-free moments to glance at it and see which they get (i.e., whether they mastered the lesson and can move on or should go back and revise). Each assignment is a single step toward completing the project; for example, in my current unit students are making a remix and the initial sequence is: 1) Choose a song (which is very easy for me to tell if the student did or didn't), 2) Set the tempo and key in our DAW (again, easy to tell at a glance), 3) Import the song into the DAW (easy to tell - you get the idea), 4) Line up the song with the DAW grid, and 5) Add new loops and sounds from the DAW loop library. These steps may seem inane and overly simple, but that's actually the point - chunking the lessons into incredibly small steps makes them both achievable for students and very easy for me to "grade." The ease with which I can evaluate these tasks also allows me to transform my role from grader of quality into gatekeeper of content - if you haven't mastered this task, you can't move on to the next one until you do. This means students who are moving along have demonstrably learned something, and the progress data quantify that learning rather than me having to try to figure out how "smart" a student is from what they've produced (in other words, achievement is measured objectively by progress, rather than subjectively, by quality, and a student who is struggling is behind, not dumb which is a huge distinction in terms of developing a growth mindset).

My upgraded gradebook - a progress tracker!

My upgraded gradebook - a progress tracker!

But the benefits don't stop there. My gradebook is more of a progress tracker than a record of each student's performance, and as such it serves more roles than just record-keeping (although it serves a record-keeping function as well - a student who has completed 4 of the 5 steps has learned 4/5ths of the content in an objective way). A student who is behind stands out on my tracker; without having to delve into their work and try to determine how much they know through interpolation and divination, I can very easily narrow my focus on those students who need help to catch up. Extrapolated across the entire class, the progress tracker provides actionable data on how the entire group is performing (traditional grades don't provide this data, because if there is any more wiggle room in assessing a piece of work than yes/no, the aggregate data of how every individual kid does on that particular piece is ambiguous; every kid in the class might get a 4/5 on the same assignment, but the missing point may be different for each of them, indicating a different misunderstanding in each student! I would separate those 5 points into 5 separate, tiny assignments). Furthermore, having this aggregate data on where every student in the class is at in the unit allows me to strategically group them together and find effective student helpers (since I can see who's ahead).

It's important to note that a gradebook set up as a progress tracker has implications for the very structure of your curriculum, and requires careful planning to establish a sequence of tasks, each with a clearly demonstrable and binary (mastered/not mastered) response, that leads students to the final outcome of possessing the knowledge and skills required to pass a unit. In my music class, every lesson represents a single step in the process of creating a song, and their learning is represented quantitatively by how far into the sequence they progress: if they complete all the lessons, they will have a song that meets all the requirements (and therefore demonstrates full mastery) of the unit, whereas if they fall short of completing the sequence, the song will be missing some components from later in the unit. More practically, every aspect of my class is structured to guide students linearly through the sequence, which is canonized in my gradebook - the first "grade" (i.e., 0 or 1) is Lesson 1, the second is Lesson 2, etc., and everything about my class (my gradebook, my LMS, and my actual teaching) is purpose-built to ensure my students can follow this path and to make it easy for me (and them, and also their parents) to tell how far along the path they are.

Of course, at the end of each unit I do grade the final product as a summative assessment, using a rubric, and by that time I have a very good sense of where each student is in their learning process; the quality my students' songs tends to hew closely to how far into the sequence they've advanced. More importantly, however, the overall quality of my students' songs has improved since I implemented this structure because the data in my gradebook give me the opportunity to intervene more frequently before we get to the end of the unit. As a young teacher, so much of what I did felt reactive, like I was constantly putting out fires that had already engulfed my classroom (and my students' learning, and my relationships with students, and their parents), and the transition to a gradebook that provides unambiguous data as to where my students actually are (not data based on a hunch or a vague sense of learning that I came up with myself) has allowed me to support my students proactively, which is one of those things young teachers are told they should do but rarely told how (if you're relying on your own observation to come up with the data, it's very easy to miss kids who are slipping through the cracks until it's too late). There are ways to improve our classes, but sometimes they look different from what we're accustomed to - certainly the idea of a progress tracking checklist feels different from the traditional conception of what a gradebook is - but if we open ourselves to change and adopt new practices (like upgrading our gradebooks and using progress rather than quality to measure achievement), we learn to teach, and we can do a better job of supporting our students.

Read More