There are three broad categories of grading criteria that have the potential to unite all teachers in the effort to grade more equitably, effectively, and efficiently. Bold claim—I know—yet I’m confident there’s something for everyone. In fact, I’m dying to hear what could be missing, so leave a comment if you think of a fourth “p” or something that doesn’t fall under one of the others. Guskey’s three categories were lurking in a 1996 article (“Reporting On Student Learning…”). He opens with a quote that sounded like it could’ve been written by a contemporary scholar, yet on the next page reveals that it was from 1933! Confirming my own experience with reading studies dating back to the early 1900s, Guskey saw consistent findings 60 years before his article, which now is approaching 30 years old. We’re talking about nearly a century of consensus on some things. One of those things is that everything teachers grade can fall under the following three broad categories of criteria:

Product – Grading what students know and can do at a certain time
Process – Grading how students get there
Progress – Grading how much students gain

These categories support my use of—and advocation for—grading process, and I’ve had an interest in grading progress, or what I’ve been calling “growth.” I’ve avoided grading product entirely. Why? My experience has been that learner differences seem far too amplified in a second language class for us to grade language ability in that way. In addition, recent discussions about grading performance & competency (i.e., product) do suggest there’s litte reason to do so. Regardless, we don’t have to go ahead and nix grading product altogether, especially if that’s what most teachers need to hold onto to get on board. Therefore, let’s look into how grading any one, or all three categories of product, process, or progress could unite teachers in a common pursuit of equitable, effective, and efficient grading (or ungrading)…

Implementation
I can imagine a grading policy that has all teachers use just these three categories (i.e., standards) in the gradebook, and one that gives teachers autonomy to prioritize them. For example, one teacher might have more of a high-stakes class (but why?!), choosing to weigh Product 50% of the grade, Process 25%, and Progress 25%. Another teacher running a course with students learning at their own pace might not have a need to grade Process at all! In their view, as long as the student achieves (Product) and shows growth (Progress), everything’s golden. Their gradebook setup could place equal weight on both, or emphasize one over the other. For example, weighing Product 25% and Progress 75% shows that it’s more important how much the student grows than it is how well they perform on individual assignments/assessments. Maybe there’s a teacher who chooses to grade only a single category, like Process at 100%. It all depends on what’s valued, and what makes sense for your content area.

Of course, establishing the actual criteria for what denotes process and progress is the real challenge. Product is the current default. There’s a host of ways out there to evaluate what students know and can do at a certain time. Guskey notes, however, that it’s VERY tricky for the teacher to evaluate process criteria, with the result often being biased. Perhaps the best way to address this is to develop a simple statement or two that describes the process of learning, and what it means to progress in it (i.e., criteria students are expected to meet). Then, have students self-assess and self-grade. As long as a student can back up their grade with evidence of learning, the teacher can rest assured. That gets rid of most bias AND makes the “grading” process like 100 times more efficient. Add in some ungrading (i.e., self-grade just once at the end of the progress report, or quarter, etc.), and you’re good.

Averaging & Zeros
So maybe these ideas feel too hippy dippy. Well, there was also a very important point about averaging and zeros tucked away in this article. While most educators have already hopped on that “no zeros” train that left the station decades ago, Guskey articulated the root of that problem. It’s not zeros, per se. It’s when combined with averaging that the effects of zeros are disastrous. For example, take the following scores:

0, 65, 85, 85, 0, 95, 95, 95

Traditionally, those scores suggest that a student didn’t do two assignments/assessments, and averaging all eight together results in a grade of 65. First of all, note how there’s only a SINGLE time when the student performed at the 65 level, so the same averaged number representing everything else doesn’t make any sense. It says nothing about what was done at the 85s, and nothing about what was accomplished with those 95s. Secondly, the scores tell another story, namely, that the student struggled at first, yet made a steady improvement. The zeros are probably flukes. Or, there’s some reason for them, especially if they merely represent not doing work (i.e., if the student had done those assignments, would they be grades of 55? 65? Would they be 95s?!). Excluding zeros from the data set gets us closer to something that makes more sense. That is, averaging the scores of actual performance, and not averaging those missing data (i.e., zeros) gives an average of student work that’s been evaluated. Excluding the zeros would result in an 87. Again, the student performed basically at that level only twice, but that number does seem to represent the rest of the scores a lot better, right?

So, the first step anyone can take to improve grading systems would be to exclude zeros from any calculation. Of course, that now begs the question of why we might use them in the first place…and I wouldn’t have a very good answer. The more responsible alternative is marking assignments/assessments as “missing,” and that’s that. But let’s say you just can’t use codes for some reason. Fine. Zeros—provided they’re not averaged into grade with the other actual scores—could tell a story that doesn’t have to mess with a kid’s GPA. Let’s look at another set of scores:

0, 95, 0, 0, 85, 0, 0, 0

Above all, there’s not enough data to draw any conclusions. Two actual scores are not enough. Averaging those scores results in a 90, but misses the whole rest of the story about why there are so many outstanding assignments/assessments. Still, that would be at least a more reasonable move than averaging everything together, in this case resulting in the impossible-to-overcome grade of 23. No, I’m not saying put a 90 in the gradebook and move on. This student’s data is incomplete. In fact, I’d advocate for actually using that Incomplete (I) option come grading term—after discussions with the student to see what’s going on, of course. Incompletes are seldom-used, but probably more necessary than we think. After all, slapping a zero on an assignment is about the easiest way out for both teacher and student. It’s kind of giving up. It’s the teacher saying “well, I’m not giving you any more opportunities,” and it’s the student saying “well, I guess I can’t do anything about it now.” Where’s the teaching and learning?! If you truly want to hold students accountable for doing work, zeros are not the way, and averaging those zeros is definitely not the way.

PPP, Averaging & Zeros
Tying these ideas together, then, we could imagine a grading policy for which teachers determine emphasis on one to three of the Ps, develop simple criteria for each, stop using zeros (or at least stop having them calculate into the grade when “missing” codes can’t be used), and have students self-assess and self-grade at the end of the grading term. Sometimes, the best improvements to education are so simple it hurts. This is definitely one of them.