Love this! And it couldn't have been more timely, as I am giving a workshop tomorrow entitled "Ungrading for Not Beginners: Is This as Great as I Thought?" (which follows "Ungrading for Beginners: The Joys of Doing Everything Wrong" 30 minutes before it.)
It's funny—every time I am interviewed about ungrading, I get asked some variation of "And can you now explain how ungrading reduces inequities in the classroom." And I always respond, "No." Which produces awkward silence. But just like ungrading is helping us to wrestle with untested assumptions about traditional grading, whatever is beyond ungrading has to help us do the same with ungrading. And assuming that ungrading does anything for equity at this point in the life cycle of the tool is just that, an assumption.
I *hope* the thing we're (re)learning from all of this is that there is no golden ticket or magic pill in teaching, no matter how hard we look for one. It will never not be a process of trying, reflection, refining, and trying again, no matter what we label our various machinations.
Jan 8Liked by Robert Talbert, David Clark, Jayme Dyer
This is such important and thoughtful information--thank you, Jayme. It's a definite issue that many of us are drawn to ungrading BECAUSE of equity issues, but new equity issues can certainly emerge. Being mindful of the full picture is crucial!
This is a very important reflection, thank you. It’s certainly true that grades play an oversized role in undergraduate students’ lives because of the preponderant role of grades in establishing perceived merit (with emphasis on “perceived”) for a wide variety of opportunities. Given the need for faculty to provide final grades, I absolutely agree that collaborative grading is perilous because students who have been conditioned to think that they are inherently less qualified or accomplished (your excellent example of the student who characterized herself as a “B student”) will often underestimate their accomplishments and grade themselves lower than students who have experienced more privileged status / encouragement. I agree (and I recall colleagues who had started collaborative grading before me alerting me to this) that students who are BIPOC, women, disabled, poor, and/or some intersection of historically marginalized identities will often “grade themselves down” unfairly because their perception of their own strengths has been stifled.
I appreciate your extensive discussion of other equitable approaches to assessment and I know I’ll learn from them. However, while I don’t think any of us can truly evade our implicit biases, I do think there are ways to minimize negative consequences of students’ tendency to “downgrade” themselves, though I think they require embracing the idea that grades are never an objective or equitable measure instrument and as a “necessary evil” they always deserve to be approached with harm mitigation strategies.
How explicitly an instructor might do that certainly depends on their privilege and power in the institution. Some of us can “get away” with giving every student who completes a minimal amount of work in the class an “A”, as long as we’re comfortable with their ability to succeed in subsequent classes for which our course is a prerequisite. More subtle is the choice to “round up” by default when we as a faculty member are on the fence, and/or if we perceive that the accomplishments that we have objectively perceived in a student’s work are being minimized when the student self-reports. This latter approach requires there to be a “paper trail” of student work that the instructor has engaged with and on which they have provided feedback, etc — something in my view entirely compatible with student self-assessment and/or collaborative assessment.
This is not to minimize in any way your observations and solutions — rather to propose that if grades are a problem, finding a way to minimize the problem should not exclude considering grades inherently flawed and therefore working in ways that might be perceived as “grade inflation” or in any case assigning the high grades students need to succeed — as long as the instructor has the privilege and authority not to face disciplinary action by an academic hierarchy that regrettably continues to depend on the construct of grade-based meritocracy.
I hear you on the idea of subversively undermining the power of grades by rounding up by default. I've considered that for my own classes, but rejected it because I don't feel comfortable giving *everyone* an A. So, I'm still making a call about which students get rounded up to an A, and which don't. I know my implicit bias can "act out" when I make that call, so I don't want to do it. Instead, I use a more structured grade-allocation system that has built-in flexibility (but many fewer implicit-bias-prone judgement calls).
Speaking as a community college writing professor, I appreciate the author's concern with equity and efforts to employ fairer grading practices (specifically with regard to multiple grading systems).
That being said, I find myself differing both with the arguments offered and the conclusions reached. Given the seriousness of the author's claims, I feel compelled to respond.
According to the author, "there is an inherent problem to determining final course grades holistically," that is, without assigning intermediate grades, because grades determined in this way are "liable to be influenced by implicit bias." Through the influence of this bias, the author argues, ungrading "perpetuate[s] systemic inequities" in a way that is ethically unsupportable, such that she "cannot and will not do it any longer."
In my view, there are several problems with this argument.
1. For the author's claims to have force, there would need to be a compelling argument that assigning intermediate grades significantly reduces the impact of implicit bias relative to non-grading alternatives.
The problem is that no such argument is made. There's no serious discussion of the non-grading alternatives to intermediate grading, which are dismissed as if they amounted to little more than surfing on vibes. And there's no serious discussion of how the practice of assigning intermediate grades mitigates the problem of implicit bias. It's simply presumed that it does, or at least, that implicit bias is not a problem for grading in the same way that it's a problem for ungrading. Critiques of the assumptions at work here may be found elsewhere on this site.
2. The author's account of bias focuses almost exclusively on *implicit* bias, which is understood to be:
(a) "by its very nature, not something instructors are actively aware of"
(b) something that "cannot be erased, or consciously changed"
(c) something that is personal in nature: "we all have it (I do! you do!)"
(d) something that is ethically determinative: "the equitable – the ethical – thing to do is avoid situations where your implicit bias can 'act out' in the world by perpetuating systemic inequities"
The problem is that implicit bias is only one form of bias. While it is a significant problem, making it the exclusive focus of analysis comes with significant costs to our understanding, particularly if it leads us to neglect
(a) meaningful possibilities for awareness and action
(b) problems of structural as opposed to personal bias
(c) the ethical complexities of our work as teachers
This brings me to the crux of my argument.
3. As I have noted, the author identifies implicit bias as something that is ethically determinative, such that "the equitable – the ethical – thing to do is avoid situations where your implicit bias can 'act out' in the world by perpetuating systemic inequities."
What would this avoidance entail? As the author suggests, implicit bias potentially comes into play in any situation where (a) student work is assessed "non-anonymously," and (b) assessment involves a "judgment call," or subjective evaluation. If we accept the author's argument, it would be our ethical responsibility to avoid such situations.
But the problem is that such an avoidance would come at a heavy cost. The more we prioritize anonymity, the less space we leave for personal and inter-personal engagement, which are necessarily non-anonymous. The more we prioritize objectivity (or "objectivity"), the less space we leave for higher-order thinking, which necessarily involves judgment calls. In short, the more we avoid situations where implicit bias comes into play, the less space we leave for things that we know are essential to meaningful, effective, *and* equitable education.
This is where the ethical complexity comes in. While recognizing the problem of implicit bias, we need to understand that the space where implicit bias comes into play is the same space where everything we have been learning about learning requires us to go. We can't *avoid* that space without sacrificing quality *and* equity. (At best, avoidance effectively involves trading personal implicit bias for the structural biases of a depersonalized education system, and that's seldom a trade that works out well.) Instead, we need to enter that space responsibly - and among other things, that involves treating it as a space of learning for ourselves as well as for our students.
Have you tried indexing students' self-evaluation to a course rubric aligned with your learning outcomes and practicing that work through exercises in assessment norming? How did that work out for you?
What I think you're suggesting is that to determine the final grade in an ungraded course, the student's work is compared to a rubric, or some other set of standards. That helps determine the validity of the student's self-evaluation. Is that a correct interpretation?
I've thought about doing this, because it seems like a good solution: have a clear measuring stick to compare student performance to determine grades. But honestly, if there is a clear set of standards, why make the students self-assess at all? Metacognition is great and I'm all for teaching it, but I think we should use self-evaluation as a tool to help students develop a better sense of self-awareness, not as a tool to determine their actual grade. The instructor, not the student, is experienced in comparing student performance with standards, and therefore the instructor should be the one to make the call about the grade.
If there is a clear set of standards against which the student's work is compared in order to determine the grade, I'm not sure the class would be considered "ungraded," though I'm open to being wrong about that!
I think you are wrong, esp. in your last paragraph. A clear set of standards according to which student work is prepared/assessed is a fundamental element of teaching unrelated to grades per se. The classroom question is: "do WHAT, HOW WELL, under WHAT CONDITIONS" -- performance standards exist to specify #2 and #3. If we're not communicating them, we are already failing students whether we give a grade or not.
To the point of your original post, I think you're accounting for two pieces of the literature while missing two others: backward design and 'hidden curriculum'. If we are practicing the former, a clear statement of how grades are determined and their significance is part of the package -- not "here are percentages" but "a W in my class means you can achieve X objectives to Y standards which enable you to Z". A rubric is the tool for communicating that to enable students to assess performance either before, or after, a grade is assigned. That is: these are tools we should all already have prepared and available IF we are teaching based on what we know helps students to learn.
Having these in class, however, also means that we're obligated to ensure students understand them -- minimally at the level of "what does it say?" and "how will it be used?", but (insofar as self-assessment is part of the goal) also and more importantly at the level of "how do I determine my own performance here?" Without this sort of norming behavior, we know that rubrics do not work the way that they are intended to.
Moreover, that norming falls exactly in line with research about eliminating a "hidden curriculum" -- i.e the specific cultural assumptions baked into school that are not equally accessible to students from diverse backgrounds. That is: if we're committed to fostering a diverse and welcoming environment, the best way to support students is by taking the time to teach them (a) our expectations, (b) the reasons for them, and (c) how to meet them. Insofar as assessment is an act of informed judgment, we are teaching them to become informed judges of their own work BEFORE it reaches us. This judgment is crucial to success, since it tells students what they're doing well, what they need to work on, and why it's important for them to do so. The better students are at doing it for themselves, the fewer issues they will have learning -- in our classes, surely, but also in general.
Which gets us to metacognition. If we are teaching students to do these things for the reasons above, then the ideal sign of their success should not just be an understanding of the material, but a capacity to evaluate and articulate that understanding in a way that (a) accurately interprets the "evidence" and (b) translates that "evidence" into a some evaluative statement. We are, in other words, already asking or expecting students to learn to grade themselves in all but the realest sense. If we don't want Dunning-Kruger to overtake them, we need to help them practice.
I'm perfectly capable of doing that evaluation for them, but that doesn't provide any opportunity to hone either the general academic skills of evidence/argument or the specific meta-cognitive skill of self-assessment. It especially doesn't teach them to view the work-products of their learning as "evidence", i.e. as a source capable of demonstrating their proficiency and growth over time. This not only leads to student under-investment in the classroom if they feel they aren't learning (itself a conclusion racially, etc. biased in ways that critical practice can help counter), but a general sense of detachment from the learning process (as opposed to its product). That undercuts the point of 'ungrading' and matters for our overall environment for this reason: https://hbr.org/2003/01/fair-process-managing-in-the-knowledge-economy?utm_source=ActiveCampaign&utm_medium=email&utm_content=This+can+make+or+break+your+company+s+best+execution+efforts&utm_campaign=HBR+fair+process
In my experience, working with students this way builds a better community, generates better performance, and ultimately takes more work off my plate as students become efficient at self-assessment and direction. The most utilitarian reason that matters is because no one will ever give you a "grade" at managing your life, but doing it well *really* matters. To do that, we have to be able to take a realistic and honest look at our efforts, no matter the context.
EDIT: I forgot to add -- rubrics also serve to hedge against the implicit bias that you're concerned about, insofar as they allow us to target *specific* behaviors, habits, skills, and capacities for evidence-based feedback. You don't get X level without evidence that you can do Y things. That's *the most* equitable way to approach things, insofar as both parties have and can discuss the same information. The instructor's assessment can be proven wrong... by evidence. The student's assessment can be proven wrong... by evidence. The grade isn't about them or the instructor, it's about the evidence on the page (or in the portfolio or...).
Love this! And it couldn't have been more timely, as I am giving a workshop tomorrow entitled "Ungrading for Not Beginners: Is This as Great as I Thought?" (which follows "Ungrading for Beginners: The Joys of Doing Everything Wrong" 30 minutes before it.)
It's funny—every time I am interviewed about ungrading, I get asked some variation of "And can you now explain how ungrading reduces inequities in the classroom." And I always respond, "No." Which produces awkward silence. But just like ungrading is helping us to wrestle with untested assumptions about traditional grading, whatever is beyond ungrading has to help us do the same with ungrading. And assuming that ungrading does anything for equity at this point in the life cycle of the tool is just that, an assumption.
I *hope* the thing we're (re)learning from all of this is that there is no golden ticket or magic pill in teaching, no matter how hard we look for one. It will never not be a process of trying, reflection, refining, and trying again, no matter what we label our various machinations.
This is such important and thoughtful information--thank you, Jayme. It's a definite issue that many of us are drawn to ungrading BECAUSE of equity issues, but new equity issues can certainly emerge. Being mindful of the full picture is crucial!
This is a very important reflection, thank you. It’s certainly true that grades play an oversized role in undergraduate students’ lives because of the preponderant role of grades in establishing perceived merit (with emphasis on “perceived”) for a wide variety of opportunities. Given the need for faculty to provide final grades, I absolutely agree that collaborative grading is perilous because students who have been conditioned to think that they are inherently less qualified or accomplished (your excellent example of the student who characterized herself as a “B student”) will often underestimate their accomplishments and grade themselves lower than students who have experienced more privileged status / encouragement. I agree (and I recall colleagues who had started collaborative grading before me alerting me to this) that students who are BIPOC, women, disabled, poor, and/or some intersection of historically marginalized identities will often “grade themselves down” unfairly because their perception of their own strengths has been stifled.
I appreciate your extensive discussion of other equitable approaches to assessment and I know I’ll learn from them. However, while I don’t think any of us can truly evade our implicit biases, I do think there are ways to minimize negative consequences of students’ tendency to “downgrade” themselves, though I think they require embracing the idea that grades are never an objective or equitable measure instrument and as a “necessary evil” they always deserve to be approached with harm mitigation strategies.
How explicitly an instructor might do that certainly depends on their privilege and power in the institution. Some of us can “get away” with giving every student who completes a minimal amount of work in the class an “A”, as long as we’re comfortable with their ability to succeed in subsequent classes for which our course is a prerequisite. More subtle is the choice to “round up” by default when we as a faculty member are on the fence, and/or if we perceive that the accomplishments that we have objectively perceived in a student’s work are being minimized when the student self-reports. This latter approach requires there to be a “paper trail” of student work that the instructor has engaged with and on which they have provided feedback, etc — something in my view entirely compatible with student self-assessment and/or collaborative assessment.
This is not to minimize in any way your observations and solutions — rather to propose that if grades are a problem, finding a way to minimize the problem should not exclude considering grades inherently flawed and therefore working in ways that might be perceived as “grade inflation” or in any case assigning the high grades students need to succeed — as long as the instructor has the privilege and authority not to face disciplinary action by an academic hierarchy that regrettably continues to depend on the construct of grade-based meritocracy.
I hear you on the idea of subversively undermining the power of grades by rounding up by default. I've considered that for my own classes, but rejected it because I don't feel comfortable giving *everyone* an A. So, I'm still making a call about which students get rounded up to an A, and which don't. I know my implicit bias can "act out" when I make that call, so I don't want to do it. Instead, I use a more structured grade-allocation system that has built-in flexibility (but many fewer implicit-bias-prone judgement calls).
Thoughtful article.
I would argue the implicit bias present in ungraded courses is also present in graded courses ... and shouldn't be a reason NOT to ungrade!
Speaking as a community college writing professor, I appreciate the author's concern with equity and efforts to employ fairer grading practices (specifically with regard to multiple grading systems).
That being said, I find myself differing both with the arguments offered and the conclusions reached. Given the seriousness of the author's claims, I feel compelled to respond.
According to the author, "there is an inherent problem to determining final course grades holistically," that is, without assigning intermediate grades, because grades determined in this way are "liable to be influenced by implicit bias." Through the influence of this bias, the author argues, ungrading "perpetuate[s] systemic inequities" in a way that is ethically unsupportable, such that she "cannot and will not do it any longer."
In my view, there are several problems with this argument.
1. For the author's claims to have force, there would need to be a compelling argument that assigning intermediate grades significantly reduces the impact of implicit bias relative to non-grading alternatives.
The problem is that no such argument is made. There's no serious discussion of the non-grading alternatives to intermediate grading, which are dismissed as if they amounted to little more than surfing on vibes. And there's no serious discussion of how the practice of assigning intermediate grades mitigates the problem of implicit bias. It's simply presumed that it does, or at least, that implicit bias is not a problem for grading in the same way that it's a problem for ungrading. Critiques of the assumptions at work here may be found elsewhere on this site.
2. The author's account of bias focuses almost exclusively on *implicit* bias, which is understood to be:
(a) "by its very nature, not something instructors are actively aware of"
(b) something that "cannot be erased, or consciously changed"
(c) something that is personal in nature: "we all have it (I do! you do!)"
(d) something that is ethically determinative: "the equitable – the ethical – thing to do is avoid situations where your implicit bias can 'act out' in the world by perpetuating systemic inequities"
The problem is that implicit bias is only one form of bias. While it is a significant problem, making it the exclusive focus of analysis comes with significant costs to our understanding, particularly if it leads us to neglect
(a) meaningful possibilities for awareness and action
(b) problems of structural as opposed to personal bias
(c) the ethical complexities of our work as teachers
This brings me to the crux of my argument.
3. As I have noted, the author identifies implicit bias as something that is ethically determinative, such that "the equitable – the ethical – thing to do is avoid situations where your implicit bias can 'act out' in the world by perpetuating systemic inequities."
What would this avoidance entail? As the author suggests, implicit bias potentially comes into play in any situation where (a) student work is assessed "non-anonymously," and (b) assessment involves a "judgment call," or subjective evaluation. If we accept the author's argument, it would be our ethical responsibility to avoid such situations.
But the problem is that such an avoidance would come at a heavy cost. The more we prioritize anonymity, the less space we leave for personal and inter-personal engagement, which are necessarily non-anonymous. The more we prioritize objectivity (or "objectivity"), the less space we leave for higher-order thinking, which necessarily involves judgment calls. In short, the more we avoid situations where implicit bias comes into play, the less space we leave for things that we know are essential to meaningful, effective, *and* equitable education.
This is where the ethical complexity comes in. While recognizing the problem of implicit bias, we need to understand that the space where implicit bias comes into play is the same space where everything we have been learning about learning requires us to go. We can't *avoid* that space without sacrificing quality *and* equity. (At best, avoidance effectively involves trading personal implicit bias for the structural biases of a depersonalized education system, and that's seldom a trade that works out well.) Instead, we need to enter that space responsibly - and among other things, that involves treating it as a space of learning for ourselves as well as for our students.
Have you tried indexing students' self-evaluation to a course rubric aligned with your learning outcomes and practicing that work through exercises in assessment norming? How did that work out for you?
What I think you're suggesting is that to determine the final grade in an ungraded course, the student's work is compared to a rubric, or some other set of standards. That helps determine the validity of the student's self-evaluation. Is that a correct interpretation?
I've thought about doing this, because it seems like a good solution: have a clear measuring stick to compare student performance to determine grades. But honestly, if there is a clear set of standards, why make the students self-assess at all? Metacognition is great and I'm all for teaching it, but I think we should use self-evaluation as a tool to help students develop a better sense of self-awareness, not as a tool to determine their actual grade. The instructor, not the student, is experienced in comparing student performance with standards, and therefore the instructor should be the one to make the call about the grade.
If there is a clear set of standards against which the student's work is compared in order to determine the grade, I'm not sure the class would be considered "ungraded," though I'm open to being wrong about that!
I think you are wrong, esp. in your last paragraph. A clear set of standards according to which student work is prepared/assessed is a fundamental element of teaching unrelated to grades per se. The classroom question is: "do WHAT, HOW WELL, under WHAT CONDITIONS" -- performance standards exist to specify #2 and #3. If we're not communicating them, we are already failing students whether we give a grade or not.
To the point of your original post, I think you're accounting for two pieces of the literature while missing two others: backward design and 'hidden curriculum'. If we are practicing the former, a clear statement of how grades are determined and their significance is part of the package -- not "here are percentages" but "a W in my class means you can achieve X objectives to Y standards which enable you to Z". A rubric is the tool for communicating that to enable students to assess performance either before, or after, a grade is assigned. That is: these are tools we should all already have prepared and available IF we are teaching based on what we know helps students to learn.
Having these in class, however, also means that we're obligated to ensure students understand them -- minimally at the level of "what does it say?" and "how will it be used?", but (insofar as self-assessment is part of the goal) also and more importantly at the level of "how do I determine my own performance here?" Without this sort of norming behavior, we know that rubrics do not work the way that they are intended to.
Moreover, that norming falls exactly in line with research about eliminating a "hidden curriculum" -- i.e the specific cultural assumptions baked into school that are not equally accessible to students from diverse backgrounds. That is: if we're committed to fostering a diverse and welcoming environment, the best way to support students is by taking the time to teach them (a) our expectations, (b) the reasons for them, and (c) how to meet them. Insofar as assessment is an act of informed judgment, we are teaching them to become informed judges of their own work BEFORE it reaches us. This judgment is crucial to success, since it tells students what they're doing well, what they need to work on, and why it's important for them to do so. The better students are at doing it for themselves, the fewer issues they will have learning -- in our classes, surely, but also in general.
Which gets us to metacognition. If we are teaching students to do these things for the reasons above, then the ideal sign of their success should not just be an understanding of the material, but a capacity to evaluate and articulate that understanding in a way that (a) accurately interprets the "evidence" and (b) translates that "evidence" into a some evaluative statement. We are, in other words, already asking or expecting students to learn to grade themselves in all but the realest sense. If we don't want Dunning-Kruger to overtake them, we need to help them practice.
I'm perfectly capable of doing that evaluation for them, but that doesn't provide any opportunity to hone either the general academic skills of evidence/argument or the specific meta-cognitive skill of self-assessment. It especially doesn't teach them to view the work-products of their learning as "evidence", i.e. as a source capable of demonstrating their proficiency and growth over time. This not only leads to student under-investment in the classroom if they feel they aren't learning (itself a conclusion racially, etc. biased in ways that critical practice can help counter), but a general sense of detachment from the learning process (as opposed to its product). That undercuts the point of 'ungrading' and matters for our overall environment for this reason: https://hbr.org/2003/01/fair-process-managing-in-the-knowledge-economy?utm_source=ActiveCampaign&utm_medium=email&utm_content=This+can+make+or+break+your+company+s+best+execution+efforts&utm_campaign=HBR+fair+process
In my experience, working with students this way builds a better community, generates better performance, and ultimately takes more work off my plate as students become efficient at self-assessment and direction. The most utilitarian reason that matters is because no one will ever give you a "grade" at managing your life, but doing it well *really* matters. To do that, we have to be able to take a realistic and honest look at our efforts, no matter the context.
EDIT: I forgot to add -- rubrics also serve to hedge against the implicit bias that you're concerned about, insofar as they allow us to target *specific* behaviors, habits, skills, and capacities for evidence-based feedback. You don't get X level without evidence that you can do Y things. That's *the most* equitable way to approach things, insofar as both parties have and can discuss the same information. The instructor's assessment can be proven wrong... by evidence. The student's assessment can be proven wrong... by evidence. The grade isn't about them or the instructor, it's about the evidence on the page (or in the portfolio or...).