What’s wrong with grading on a curve? Or what’s wrong with grading by straight percentages? Twelve years ago I wrote a post about why grading on a curve is destructive and counterproductive — and why grading by straight percentages isn’t actually that much better. I then described a third alternative, which is the method I used then and use today: making a scale that’s appropriate for the difficulty level of the particular questions on the assessment.

I still believe every word I wrote.

Since then, however, I’ve learned about a fourth possibility: standards-based grading (SBG, for short), which is used by some (perhaps many?) schools. I haven’t actually adopted this method yet — for reasons that I’ll explain below — but it’s very intriguing, and I would like to persuade myself to give it a try.

There are a lot of descriptions of standards-based grading floating around. I could quote or link to any of them, but I’m currently liking this three-page infographic from Sarah last-name-unspecified New Jersey Math Teacher (SNLUNJMT for short), who summarizes it all in one place. I’ll pause now so you can read it carefully and digest it slowly:

OK, all of that makes sense. But what does it really mean? For that, we’ll turn to SNLUNJMT’s meta-page, which contains a lot of links. Too much to go through, but at least it does provide all the material you need, all on one page. You can read the links, but it might be best just to try to absorb everything that’s in that infographic.

BTW, if you find all this focus on grades annoying, check out SNLUNJMT’s brief post on How To Learn, since that’s the real point after all, isn’t it? (Well, maybe not, say some; I realize that many of my students will claim that the real point of grading is that it helps them get into the right college. But of course I disagree.) If changing the grading system can make a significant positive difference in how students learn, I’m all for it.

As I’ve written in a number of earlier posts, the “math wars” encourage a false dichotomy between skills and concepts. It’s false because students clearly need to learn both. And the point is not to get 80% on a test or to scrape by with a minimum amount of knowledge; the point is to master each skill up to at least a certain level of competence, and to master each concept up to at least a certain level of competence. If you’re not there yet, the operative word is “yet.” You get the opportunity to try again until you get there. There should be a certain standard for each skill and each concept in a given course. Read the infographic again.

You may also be interested — at least I am — in some of the changes that SNLUNJMT has implemented. I’ll definitely keep them in mind if I follow in her footsteps.

So, what are my reservations? First of all, they’re not the ones expressed by Joe McKeown in George Lucas’s newsletter. He claims the SBG assumes…

• that teachers only teach skills and not content
• that homework is always skills practice
• that grades should be exclusively for reporting what students can do against standards at the end of learning

Do read McKeown’s argument, which includes some good points even though I disagree strongly with his conclusions. His principal flaw is that standards can perfectly well include what he calls content (what I’m calling skills). But you don’t need my response: just read the first two comments (by Arthur Chiaravalle and jbouley), which give a thorough rebuttal.

So if my reservations don’t lie with McKeown’s argument, what are they? The answer is simple. Unless I’m mistaken, it appears that SBG takes too much time! I’m going to seek some advice from SBG practitioners and see whether it can be implemented without spending an unreasonable amount of time. Stay tuned…

Categories: Teaching & Learning