Parts of Feldman’s assessment of the problems with traditional grading resonate deeply and are places to which I’d largely worked myself in my thinking about what needs to change: the way a single letter grade unhelpfully collapses information about all kinds of things into one metric, the deeply embedded notion that grades are for ranking students against each other and that we need some sort of bell curve for our grades to be legitimate. Isn’t our mandate to educate every student as best possible? If that means allowing revision and retakes, doesn’t that make sense? (Although I'd argue that there are practical limits to what's possible here, as Feldman does seem to acknowledge.) And of course, there’s the devastating math of the 0 for missing assignments averaged into grades, when the part of the scale in use these days really is the C to A range, with very rare Ds and virtually no Fs. I also like Feldman’s point about not averaging assignments equally across the year, because students come in with different levels of ability and so one might want to privilege the learning that happens in one's class instead of in prior years and weight later assignments more heavily.
Other parts of the book grated, because there was a self-righteousness about them that presumed only Grading for Equity devotees (1) use rubrics, (2) define for students the goals they’re trying to get them to meet, (3) think about the skills they want students to gain, (4) design curriculum intentionally, (5) vary their methods of assessment, (6) refrain from grading classroom behavior, (7) refrain from grading homework, (8) do their own gradebook calculations, rather rely on software to do it for them. I don’t know any teacher who lets the LMS calculate grades for them! While it also may be out of step with grading norms today, it's also not de facto "mathematically inaccurate" to use a traditional 100-point grading scale to indicate that mastering 60% of content is the acceptable minimum bar.
I also know very few teachers who have time or inclination to grade homework. To some extent I wonder how much Feldman is speaking about elementary and middle school classroom practices as opposed to high school at college prep schools much more likely to operate closer to a college model.
Toward that end, the part that’s most problematic for me is the standards-based grading. It’s mechanistic and assumes that all learning is a process of linear acquisition of explicit, discrete skills and content. I finally don’t think that creative and critical thinking, talking and writing about literature or other ideas is reducible to those kind of binary measurements: mastered/not mastered. I do think it is possible to teach students what we value in engaging with texts and ideas and to help them become better at that. But mastery is a lifelong process--and performance varies from day to day and text to text, even for the best in the field. Having taught writing at the freshman undergraduate level in a program devoted to thoughtful pedagogy, I do know that both those faculty and the faculty in the other two undergraduate humanities departments in which I have taught strongly agree with Feldman that what gets graded is not effort. But neither is it a checklist of discrete skills and whether a student has acquired them. It’s the quality of the piece on the page. There are chances for revision. There is iteration and investment in a common vocabulary for talking about the elements of the essay. There are explicit examinations of writing’s structure and organization. But at the end of the day, for a multiplicity of reasons, a given piece may excel, be passable, or just not work. Progress is rarely linear. As any novelist or moviemaker or reporter or essayist or scientist knows, they may have a stellar debut book, film, lead story, or scientific finding, but it’s no guarantee that the next one will shine as brightly. Students can learn to read, think, discuss, and write better. But there’s no moment when, check, they’ve definitively mastered those skills.
I also disagree that classroom behavior is an irrelevant soft skill. Learning to sustain a dialogue with others in which your ideas build on each other is, in fact, a core skill I’m teaching. I can happily work with introverts, anxious kids, quiet intense thinkers, exuberant talkers, students with ADHD, etc. But if they don’t listen to and respond to what others say in some form, and instead try to operate in a silo, making their learning about me and them alone and disregarding classmates, they have missed a core purpose of the class. I have a rubric for all the ways a student can legitimately contribute to a class conversation or other endeavor that don’t involve quantity of contributions or brilliance of ideas. Students like that rubric. But I firmly believe in grading this aspect of a high school class full of future scientists, businesspeople, non-profit leaders, community organizers, institution directors, and Congresspeople. It's about creating intellectual community and making the whole better. In that respect my class is about civics as much as anything else.
A quibble in that vein: Feldman stands firm against grading class participation, in large part because he says that doing so is more subject to teacher bias and subjectivity than other metrics. Yet at the end of the book, he argues that teachers are experts who can lay aside a student's performance on summative assessments if the student doesn't test well and decide to weigh their memory of the student's contributions to class discussion as better proof of student mastery of the content than tests. Um, what happened to not relying on subjective memories of students' class participation?! What is a class participation grade if not an assessment of the quality of student's understanding and ideas as shared in class?