I’ve often emphasized the importance of transparency and fairness in teaching. The evaluation of written assignments is an inherently subjective activity, at least from the perspective of students. The grading of written assignments is most prone to the appearance of unfairness. When students think they’re being treated unfairly, they are not inclined to focus on learning.
Moreover, in the grading of written assignments we are most likely to be inadequately transparent and unfair. By using rubrics to grade writing, we can mitigate, or perhaps even eliminate, this problem.
Some folks don’t like using rubrics because they think that written assignments should be evaluated holistically or by gestalt. As experts in our field, we can tell apart a B paper from a C paper based on reading without the use of a rubric, and we can explain to students in our evaluation how this distinction is made without resorting to over-simplified categories. We can reward deep insight without being captive to a point-making system.
Even if the concepts in the preceding paragraph were factually correct, the choice to formulate is such an argument indicates a lack of focus on student learning. Rubrics should be used to grade written assignments not only because they lend themselves to the appearance of fairness in the eyes of students, they actually result in more fairness.
Grading written assignments without a rubric is unfair. Why is that? It’s very simple: when an assignment is graded without a rubric, students do not know the basis upon which their writing is to be evaluated. Fairness requires that students know in advance the basis upon which their grade is being assigned.
There are many different components to good writing, and presumably someone who grades holistically takes all of these into account in an integrated fashion and then assigns a grade. However, if the purpose of the assignment is to learn about writing, then the student needs to which components are important constituents of good writing. And then the student needs to receive credit for including these components, and not receive credit if not including these components.
If a professor wishes to reward students for making “deep insights,” then these deep insights can be placed as a category on the rubric. And, when handing out the rubric when assigning work to students, the professor can then explain in writing on the rubric what constitutes deep insights that are worthy of receiving points in the rubric.
Rubrics don’t rob professors of flexibility in grading written assignments; they only prevent professors from ambushing students with criticisms that the students would not have been able to anticipate. They also prevent professors from unfairly rewarding students who are able to perform feats that satisfy the professor’s personal tastes even though these feats are not a required part of the assignment.
Is bad grammar something that deserves points off? Put it on the rubric.
Should it be impossible to get an A without a clearly articulated thesis and well supported arguments? Build that into the rubric.
Does citation format matter to you? Put it on the rubric? Don’t care about citation format? Then don’t put it on the rubric.
When you’re grading, you should know what you are looking for. So, just put all of those things on the rubric, and assign the appropriate amount of points to them as necessary. Of course any evaluation of “clear thesis” and “well supported argument” is to some degree subjective. However, when students know that the clarity of their theses and the quality of their arguments are a big part of their grade, then they will be aware that they need to emphasize that up front, and focus on writing well. This point might be obvious to faculty, but it’s not necessarily obvious to all of the students. To be fair, every student needs to know these kinds of things up front and in an unbiased fashion.
There are several other reasons to use rubrics:
Rubrics help reduce the unconscious effects of cultural biases. Students who write like we do are more likely to come from similar cultural backgrounds as ourselves, and students who write well, but differently than we do, are likely to come from a different cultural background. If grading is holistic, then it is likely that professors will favor writing that reflects their own practices. Without the use of a rubric, professors are more likely to assign higher grades to students from cultural backgrounds similar to their own.
Rubrics save your time before grading. Students often are demanding about their professors’ time when they are anxious about whether they are doing the right thing. The more specific information students receive about what is expected of them, the more comfortable they are with fairness and transparency in grading, the less often instructors are bothered with annoying queries about the course, and the more often they’ll contact instructors about substantial matters pertaining to the course material.
Rubrics save your time while grading. If you grade holistically without using a rubric, and it takes you appreciably less time than it takes with a rubric, I humbly suggest that you’re not performing an adequate evaluation. The worse case scenario, with respect to time management while grading, is that a complete evaluation happens without a rubric, and then it takes only a few moments for the professor to then assign numbers on a rubric after being done with a holistic evaluation.
Rubrics save your time after grading. If students are unpleased with a grade on a written assignment, and all they have to go on is a holistic assessment and written comments – regardless of verbosity – they are far more likely to bother you to ask for clarification or more points. If they see exactly where on the rubric they lost points, they are far more likely to use their own time to figure out what they need to do to improve their performance rather than hassle you about it.
Most importantly, rubrics result in better writing practices from your students. It is a rare student who relishes receiving a draft of an assignment with massive annotations and verbose remarks about what can be done better. Those remarks are, of course, very useful, and students should get detailed remarks from us. When fixing the assignment, students will be focused on getting a higher grade than they received on their draft. The way to do promote success by students is to provide them specific categories on which they lost points. This kind of diagnosis, along with any written comments that professors wish to share, is more likely to result in a more constructive response and is less likely to terrify students who are unclear how to meet the expectations of a professor who gave a bad grade without providing a specific breakdown about how that bad grade was assigned. If a student wonders, “what can I do to produce excellent writing?” all they’ll need to do is look at where they lost points on the rubric. That’s a powerful diagnostic tool. If you think the use of a rubric in your course cannot be a great diagnostic tool, then you haven’t yet designed an adequate rubric.
Of course, it’s okay to disagree with me about writing rubrics. If you do, I’d be really curious about what your students think. The last time I graded a written assignment (a take-home exam), I asked my students if they wanted to receive a copy of a grading rubric before I handed out the exam. They all wanted it, and they all used it. By choosing carefully what I put on the rubric, I was sure that their efforts were allocated in the best way possible.
We used rubrics for writing assignments in an Ecology lab and the feedback from students was uniformly positive. They liked having expectations clearly stated and they felt it really helped them improve as writers. One drawback from my perspective was that I sometimes felt constrained – even in science writing there can be a certain “je ne sais quoi” that differentiates good from great papers but is nevertheless difficult to articulate in a rubric. Another drawback – I once read a paper in which bird communities with high species richness and low abundances had similar diversity index values to those with low richness and high abundances. Rubrics can result in something similar if you are not careful about how you weight categories that relate to structure/mechanics versus those that relate to content. To a large degree, however, both of these issues relate to the quality of the rubric. We revisited ours after each assignment and at the end of the semester and the rubrics did a much better job of distinguishing good papers from bad the second time I taught the lab.
My rubrics have definitely evolved as I identified new shortcomings in my student papers. For example, I initially thought it was obvious to them that they should write some sort of concluding sentence rather than just ending the paper abruptly. Some didn’t do this, so now that’s another line on the rubric
I fought rubrics when I first started, but came convinced of their utility when I caught myself being snowed by clear, effortless writing that lacked certain essential components. On the other hand, rubrics also help keep me from slaughtering a poorly written paper that contains all the necessary bits (albeit well hidden in poor grammar, etc.)
Thanks, Terry. I started teaching with absolutely no formal training or experience in grad school, and initially I didn’t use rubrics. It took about one semester to realize my mistake. My rubrics seem to get larger and more specific over time, as I address new areas where students commonly make mistakes, to the point where I think they start to get unwieldy. I would love to see an example of a rubric that you think strikes a good balance between size and clarity. Any chance that you might want to share one of your rubrics?
Also, I’m torn about the issue of grammar and spelling. I feel like college students who make such mistakes are really sloppy, and there’s no excuse for it. So I specify that in all my rubrics. But they still turn in sloppy papers! So I end up in the situation where a student who has done the minimum amount to satisfy the requirements of the rubric gets a better grade than a student who wrote a more thorough and genuinely insightful paper, but failed to run a spelling and grammar check. That’s always depressing. On the other hand, I can’t imagine any situation in a professional setting where producing a document riddled with spelling and grammatical errors is acceptable.
The other thing I struggle with is how to include “writing like a scientist” in a rubric. In my freshman courses, we read a lot of (simple and straight-forward) primary literature, and then students conduct research projects and write them up in the style of a scientific publication. Many of them turn in a paper that reads like a high school chem lab report. A few actually absorb the style of science writing from the publications they read and replicate it in their papers. I find it hard to quantify the importance of style in science in a rubric.
I actually don’t have an example of a good rubric, at least one that isn’t highly specific for a very particular assignment. I think it’s really important, but I’m not quite at the point where I can model the behavior well. (I can just argue for it.)
I’ve found that in subjects with a large cohort and many markers (e.g. >550 students and 20 TAs) a rubric can make the marking process more fair, because markers grade against the rubric criteria rather than ranking students against each other. This made it easier on the markers to stay consistent and the students were happy that their markers had the same criteria and benchmark to grade all reports.
I was formally introduced to rubrics in my “night job” of teaching and evaluating improv theatre. It was relatively simple (7 categories with 5 pts; 3 categories with 7 pts; 1 category with 3 pts), but it was remarkably consistent within and among adjudicators, within and among years.
During the last year of my PhD, I was on the departmental grad studies committee, and convinced the rest of the faculty that we should use a rubric to evaluate the comprehensive exam (before then, it was a pass/redo/fail system that was like pornography – you knew it when you saw it). Now there’s a formal rubric for the written component, one for the oral presentation, and one for the Q&A. The student really appreciated knowing what aspects were being evaluated, and how those were weighted. I include grading rubrics in my sample syllabi for job applications, and they’re often my default when designing an evaluation tool.
One of the professors at my undergraduate university gave a talk on a rubric system he and a graduate student invented for their introductory biology course. As an experiment, they decided to independently score the same set of papers within the class using the rubric. While the results were mostly consistent with each other, there was a lot more variability in the allotted scores than they expected.
That isn’t to say that rubrics aren’t valuable, or that their use isn’t superior to their absence, but they still entail a fair amount of subjectivity.
Reading the comment above, the question becomes under what circumstances rubrics can be consistently applied, and under what circumstances they are more variable. Is it a matter of how a rubric is constructed, regardless of subject, or are some subjects easier to grade under a rubric system than others? How much does experience in grading under rubric systems play into things?
The first lines of the post indicate that grading is inherently subjective. So, yes, that is clear.
The use of a rubric to promote fairness goes beyond whether or not individual scores in each rubric category are graded consistently. The real value of a rubric for fairness, as indicated by other commenters as well as myself, is that the rubric will allow students to know the basis of their evaluation while they are working on an assignment. That way, they can’t be surprised by being graded using vague or unstated criteria, which is how many professors grade written assignments.
Ah, apologies. Am quite tired, so only skimmed the post.
Should be more lucid next time I comment. ;-)