Years back, I wrote about how a standards-based model to learning and grading (SBG) fell short of the bar. This was true of a particular kind of SBG—the kind with 10-30 standards being tested and graded every single quarter, scheduling multiple reassessments for each one, and still using scores of 1-4 in a way that keeps focus on points (not learning). The good news is that not all models are like that. The bad news is that a LOT of them are, which in turn give standards and accompanying practices a bad name. Teachers end up hating SBG, and admin scrap plans for any schoolwide change.

To be clear, I’m more of a “burn it to the ground” kind of guy, advocating for little to no grading whatsoever, but I’ve also found that a basic understanding of standards is crucial to ungrading. In fact, I’m not sure you can do it without standards…

1) To grasp standards—and actually use them to your benefit—take a recent unscored assignment that you’ve collected.

2) Now, sort the assignment first into two piles: students who get it, and and students who don’t quite get it.

3) After you’ve got two piles, go to the “got it” pile, and sort them again, this time into “reaaaaaally get it,” and “get it.”

4) Move back to the “didn’t quite get it” pile, and sort them again, this time into “reaaaaaally don’t get it,” and “don’t quite it.”

What you’ve just done, without any complex detailed rubric or list of criteria whatsoever, is assessed students according to some standard: the “got it,” and you now have 4 levels of that standard, which *if truly needed a score/grade,* I would say correspond to 95(A), 85(B), 75(C), and 65(D). Why no 100(A+) or 55(F)?! In short, students get a 4.0 added to their GPA for any grade above 93 in most school systems. The 100 is psychological and symbolic, there for exceptional work or kids who would freak out without it. The 55s are reserved for the lowest of low performers seriously struggling.

So, whatever students “got” could be articulated as that assignment’s standard. Give assignments that assess this standard 3-5 more times throughout the quarter, and you’ve now got enough data to grade that standard.

That’s it, but cavē! Giving each assignment its own standard has the potential to end up with 10+ standards. Under that model, you’ll be gathering waaaaay too much learning evidence for each quarter. With 3-5 pieces of data needed for each one, that can get out of hand. My rec is to go with 1-5 standards. If you’ve got 10 of them, combine a few of them. Is there some other global umbrella they could fit under that gives you more flexibility while still reporting on what students know and can do? Make it happen. If you’ve got 20 standards you can’t live without, consider focusing on only several at a time, like gathering evidence for a DIFFERENT set of 5 each quarter.

What’s the catch?

The trick is selecting or developing standards at a Goldilocks level of not too narrow (e.g., single-assignment-specific), and not too broad (e.g., “Standard #1= The Student Understands Math”). Once you have a few, like the reaaaaaaally good ones that apply to everything you expect students to learn all year long, all it takes is some simple criteria to make it official. Think about it…what was the “got it” you had in mind during the sorting exercise? Formalize that into a sentence or two, and you’re good. If it can apply to what you’ll mostly be asking students to do all year long, bonus! You’ll be eating some Michelin star porridge in no time.

Now that you’ve got a handle on standards, you can go ahead and follow a more traditional SBG model with something like giving scores of 1-4 based on meeting the standard (or not), and start thinking about how new learning evidence can overwrite old evidence (i.e., do NOT average all those scores). Or, you can take an ungrading approach, and give those standards scores/grades just once at the end of the quarter. Beyond that, there are a lot of upgrades and other grading practices to explore.

For example, in K-12 teaching, I’m a big fan of student-led assessment and self-grading at the end of each quarter, but also during he first progress report! That’s because in my experience, people get nervous about not seeing a grade until November. And that makes sense given the history of grading and what grades have been used for. So, do the first self-grade earlier on so students always have a grade in your course showing up in the gradebook. Just remember to overwrite anything entered at progress reports with what the data show at the quarter (i.e., do NOT average!).

OK, what are your questions about standards?