Saturday, May 3, 2014

The Separation Of Learning And Grades Is Antithetical To SBG

Ilana Horn writes that amazing math programs make a clear distinction between doing school and doing math. By making explicit what aspects of "doing school" are necessary for students, these schools are able to separate their impact on doing and learning math.
In line with their goal of increased participation, the teachers were explicit that learning to be a student was an important part of their curriculum, and they came up with structures to support that learning. At the front of each classroom was a homework chart laid out much like a teacher’s roll book, with students’ names in a column along the side and the number of each homework assignment across the top. Although actual grades were not posted, completion of homework was represented by a dot...At the same time that they emphasized traditional student skills like doing homework, they did not confuse failure in class with students’ intelligence or ability.
In this way, teachers were able to help students do the things that are necessary for school, but not let that get entangled with their feelings about math or doing math.

How does SBG compare to Horn's model? SBG is premised on the idea that grades should represent how much a student knows. I'll link to Dan here because this was such an influential post for so many people:
The students record the new concepts and their scores on their Concept Checklist. (Template here.) They have now started a self-monitoring process which will continue throughout the year. This is one of the most beautiful parts about this system, that every student knows exactly what she does and doesn't know.
This runs completely against the separation of learning and grades that Horn advocates.*

*This is what I wrote in the original draft. Ilana Horn was kind enough to clarify that I've muddled her position. Check out her comment below, or her clarification on twitter.

I stopped using SBG because I found that it was suffusing every aspect of assessment and feedback with grades and their negative effects. We should be trying to push grades into a corner, out of sight, but SBG wants us to instead dangle them in front of our kids on a weekly basis.

(Score-keeping: This is a change from my previous position that "SBG is worth it." I no longer think that it is, both because the tendency to too-neatly divide math into standards is an unhelpful way to think about learning and because grades and learning need to be separated. I think that I'd rather have grades be determined by occasional tests and homework than by learning, but here I'm less sure of myself. What I currently do for quizzes is supported by my school's no-grades policy, and I'm sure that's influencing my direction here.)


  1. This is an interesting stance. So if you had to do grades, would use SBG? It sounds like you would advocate for something based on habits, or productivity like homework grades, or time-on-task to give these grades a non-intelligence based foundation. I know you're at a school without grades, as am I, but how would you set it up if you weren't?

    1. If I had to give grades, I would be trying to live by these principles:

      1. Formative assessment is too important to be sullied with grades.
      2. Grades shouldn't purport to reflect mathematical knowledge.
      3. High-stakes assessments can be corrosive to classroom culture.

      If I were giving grades, I think that I'd handle quizzes exactly as I handle them now. I'd tell kids that your grade isn't about how much math I think you know, but about how well you do school. Doing school means studying for tests -- much like the ones you'll end up taking at the end of the year -- and doing homework. I'd give 2-3 tests a semester, and I'd offer something like "the average of your two highest grades" as the grade. Maybe I'd offer occasional retests? I'd try to keep grades relatively high, maybe arbitrarily setting a floor of 75 for anyone who meets some minimum bar?

      (Maybe I'd drop the homework bit, just because I hate collecting and marking homework.)

      This is obviously a lot of blustery talk since I'm not giving grades now. But last year I was -- I don't think that I've totally lost my marbles since then.

  2. Michael, teacher educator me enjoys how much you think about things and how committed you are to growing as a professional. Researcher me thinks this is a great example about why learning to teach is so hard. I know you read that post and thought about it, yet the conclusion you reach is not one I would endorse. Isn't that fascinating?

    Anyway, now I'll just talk as me-me.

    The great thing about topic checklists is that they provide a self-monitoring tool for students. One really critical student skill is keeping track of what you know and how well you know it. I think that is a big plus of (what I understand of) SBG. The second way that SBG can help distinguish between doing school and doing math is by getting rid of a lot of noise in the way grades are often done in school. We know that an overemphasis on things like work completion and extra credit can lead to "good grades" without signalling actual learning.

    1. It takes a certain amount of chutzpah to stick to my guns after I've been told that I misinterpreted the subject of my post. But I got to this point by being aggressively suspicious of my own beliefs, and I'll dig in here in the hopes of exposing my views to even deeper criticism.

      The way you state it, the problem with traditional grading is that it doesn't properly reflect student learning. SBG takes a step in the right direction by putting the student learning back into grades.

      I've got to ask: how are we so sure that we're capturing student learning with an SBG grade?

      I mean, when it comes to evaluating teachers based on test scores we balk, don't we? There's all sorts of stuff that a teacher does that can't be assessed with a test. Really, any single evaluative metric will necessarily be incomplete. Besides, measuring learning, by any measure, is a tough business. Humility requires me to admit that I can't design quizzes that will accurately do this for 80 kids on a weekly basis.

      "Don't let the perfect be the enemy of the good!"

      But the problem is that if we're not measuring student learning with SBG quizzes, then we've put ourselves in a bit of a pickle. Because with SBG we're making the strong claim that we are measuring student learning, that this isn't just about doing well in school but that this is measuring your mathematical ability. And now we've completely mushed up grades and learning in a way that they weren't before.

      Again: the basic premise of SBG is that grades on individual skills will be empowering and motivating for student learning. But we also know that grading creates distortions and introduces status issues. By telling kids that their grade is reflective of their mathematical knowledge, we're introducing those distortions in a more explicit manner.

  3. There's virtue in length, there's virtue in brevity. Here's a condensed version of my argument:

    "The things that are most important for student learning shouldn't be graded. If quizzes are really important formative assessments, then they shouldn't be graded."

  4. I interpret formative assessment as a means to gather feedback about student learning during the learning process. As such if they are used to help make instructional decisions, formative assessments should not be graded.

    It's interesting that your students appear to translate progress towards a learning target as a grade. There's a terrific article the March issue of Middle School Journal about creating self directed learners in a math class. In it the author describes how students learn to become self directed when they weekly monitor and track their progress and reflect on what they did to meet their goal.

    I think there is a real danger in turning skills into minute checklists; there needs to be balance so students don't feel like school is being done to them.

    1. "It's interesting that your students appear to translate progress towards a learning target as a grade."

      I mean, isn't that the premise of Standards-Based Grading?

      Or am I misunderstanding your point? Apologies if that's the case.

    2. I guess I see SBG as more open ended in the sense that a grade is not issued until the student demonstrates a certain level of proficiency. I agree that eventually that level of proficiency gets translated into a grade, but until that time the conversation is about learning, not about grades. It's a small distinction and some would argue there's no difference, but for me it puts some distance between learning and grades.

      Does that make sense?

    3. That makes perfect sense. The crucial question, to me, is whether our students see that distance or not.

      Maybe I was doing SBG wrong? Maybe there was too tight of a connection between my rubric and grades?

      If this ends up with us discussing ways that we distance grades from our formative feedback within an SBG system, I would be very happy.