The Classroom Experiment Part III: How should students be graded? Avoiding self-judgment and activating self-improvement
One of the big questions I had as a researcher turned teacher was, is it possible to use exclusively formative assessments as both feedback and grades? In my last blog post, I started off by stating that the reason it was not feasible to use only formative assessments all the time was that it was not possible to track 79 students across six dimensions… and then I figured out a way to do that.
So did I then use only formative assessment?
Yes and no. First, like any good researcher, I have become a little more precise with the question I am asking. I was previously mentally grouping “formative” with “qualitative” and “summative” with “quantitative”, but of course, in a traditional school setting… everything is quantitative.
By default, the Canvas app shows students their numerical grade only. To see teacher comments, they need to click on a small link that says “Submission and Rubric” (it doesn’t even mention comments!). The problem is, even if you choose to grade things complete/incomplete to avoid numbers, the students still don’t read the comments. It is as if the design of the app purposely focuses their attention on the grade rather than the feedback. I, and many other teachers, have spent hours writing comments that we know that no one ever read.
I realized that it didn’t really matter whether I framed the assignment as formative or summative; if there was a grade, the students perceived it as evaluative. And with Canvas, there was always a grade.
So my focus shifted to the value of purely qualitative feedback.
I developed a system that gives students exclusively (qualitative) feedback, sidestepping quantitative grades altogether. In that sense, it embodies the essence of formative assessment: it focuses their attention on what they have been doing well and what to improve on, rather than on judging themselves.
Students need a safe learning time/space where they don’t feel they need to perform.
As soon as students feel evaluated, even if you are just collecting the data from a game in class, they begin to focus on the evaluation rather than the learning, and worry about getting things wrong.
Learning, of course, starts with not knowing. If a student is judged at that point, or really at any other point in the learning process prior to mastery, they feel inadequate at best (and at worst, begin to fear how their parents will respond). They must then crawl out of this psychological hole for long enough to grapple with the content in order to actually learn it!
This grappling is hard work, and involves a lot of uncertainty, which is all the more difficult to navigate while judging yourself. As I know well from improv, taking risks requires a lot of positive thinking, and our inner critics are trying our undermine us at every step.
When thinking in this self-critical way, the brainpower we could be using to learn is being used to judge ourselves. Further, when we feel threatened, we become narrow-minded, not seeing possibilities and opportunities that we may naturally reach for if we feel safe. This is directly counterproductive to learning, and so I decided it must be addressed.
As I mentioned, I have set up a system where I am now giving students only qualitative feedback. That means, even for quizzes and tests, they receive a series of comments about their answers to each question.
When I first explained the rationale behind this, introduced the idea of quantitative vs. qualitative data (this is a science class, after all), and rolled out the personalized links for them to see their comments, each of these comments came with an E, M, P, or B: exceeding, meeting, progressing toward, or below expectations. But even that quantization proved to be too evaluative.
Students made it clear that they are looking for grades and trying to quantify the data. The most common response was, “How many Ps is bad?”. A few students counted up the comments and made their own pie charts… but then they didn’t know what to conclude from the charts. All they could do was show me and ask, “Is this good?”
Certainly, many students read the comments and learned something new from them. I had them complete a reflection using the following questions:
1. Which comment was most surprising?
2. Play with the Grouping and Sorting at the top of the page. What is your favorite way to view the data and why?
3. What grade would you give yourself for each of the 6 standards if you had to grade yourself today? Why? [The standards are Content Knowledge, Use of Information, Writing Skills, Oral Communication, Citizenship, and Academic Habits]
4. What do you want to focus on improving for the rest of the year?
5. What is at least one comment from this year that you would add to the list to give a better picture of your learning?
Many students had very nice reflections, with true surprises, thoughtful self assessments, and ambitious yet attainable goals! I think this first student interaction with their feedback was very worthwhile, on the whole.
One interaction really struck me from all the student reactions. This student said “I just wish I had all Ms”. I explained that, logically, you cannot master something without “progressing toward” mastery at some point. He sighed and quietly said “Yeah, I guess that‘s true’.” That’s when I knew that this system was still activating the self-judgment and performance mindset I was trying to avoid, at least for some of the students most entrenched in that way of thinking.
So today I took off the labels on the formative assignments’ comments and kept only those on the summative ones. We’ll see what happens on Monday! My hope is that the students will have to actually read the comments and use them to improve their learning, but anything could happen!
I think this sets up an interesting dynamic, where now even a quantitative comment such as “You got a 5/11 on the Kahoot” could potentially be seen as formative because they know that this score doesn’t actually count toward their final grade, but they also know they probably need to study whatever the Kahoot was about!
This is exciting because it gets to the heart of what many teachers are trying to address when they choose to grade the assignments in their classes as complete/incomplete, but still captures tons of useful qualitative and quantitative data and feedback.
I wonder if I will need to remove all grades, even from the summative assignments. Personally, when I play a board game, I don’t like when I have no idea how I’m doing until we count up the points. And so with the game of school, there may need to be some indicator of how “well” a student is doing to keep them oriented and motivated, given that they have a limited amount of time to spend on my class.
The trick is to not let that interfere with them learning to do even better.