Teachers all use rubrics. Of course we do.
And those rubrics are probably OK—serviceable and likely supportive of learning. However, many are also likely flawed.
The practice of using rubrics is well-established. In a 2000 article, Heidi Andrade notes that "[a]t their very best, rubrics are … teaching tools that support student learning." As educators, we've seen myriad examples of rubrics. We've written them, borrowed them, modified them, and added to them. In theory, rubrics are useful because they inform students of their level of achievement, providing detailed descriptions of a range of evaluation criteria. We know it's best practice to continually update and tweak our rubrics in response to student performance and feedback. But the question remains: do students actually use these detailed rubrics? My suspicion is that they don't, due to some basic flaws in how traditional rubrics are designed and used.
The Problem with Rubrics
Traditional rubrics are too often a patchwork quilt of teacher-talk—a wall of words. By "traditional," I'm referring to analytic rubrics where performance or product criteria are listed in one column, and levels of achievement (often numbered 0–4) are laid out in a row across the top. This creates the familiar grid filled with cells describing various levels of student achievement. In my experience, students read these grids about as often and in as much detail as you or I read the user agreement for our latest phone app—not very well, if at all. Andadre (2000) describes good rubrics as "concise and digestible," but I've witnessed lab-report rubrics with 40 different cells. Yes, a small subsection of students may conscientiously look at the rubric when given a new assignment, but the majority of students give traditional rubrics only a cursory glance. And there is no joy in creating a protracted, repetitive document that your students are not going to use.
By "use," I'm referring to the act of students interacting with a rubric before and during the learning process. A student briefly looking at their grade and perhaps noting some of the descriptors that have been circled doesn't qualify as using a rubric. If we want to place the student experience at the center of the learning process, we need rubrics that are written for students, rather than for teachers.
In addition to being overly long and detailed, many rubrics—maybe most—rely heavily on negative descriptors, or things we don't want students to do. A Google search for the term "Oral Presentation Rubric" returns innumerable examples heavily laden with words such as partial, incomplete, missing, unable, incorrect, flawed, and not present. One rubric labeled "kid friendly" included the phrase "audience seemed disinterested." In a sample rubric shown in a recent article (Alonso, 2017) more than half the descriptors are negative, identifying what students should not do; negative descriptions even appear in the "Satisfactory" column.
In my experience, students read rubrics about as often and in as much detail as you or I read the user agreement for a smartphone app.
Matthew Leisen
Rubric language is also often oddly imprecise. Although most guidance on how to craft a traditional rubric recommends making teachers' expectations very clear, example rubrics include phrases such as "essay marches along, but doesn't dance," "there is … no hint of a real person in it," and "I'm trying too hard to impress'' (Andrade, 2000). These types of negative or unclear descriptors are demotivating and confusing for our students.
Why So Many Negative Descriptors?
One potential reason for using negative descriptors is to eliminate the need to write negative feedback. It's quicker to circle the phrase Missing Evidence, than to write out the specific feedback. Another reason is that when creating rubrics, the starting point is often Competent (or 3, B, or whatever is deemed to basically meet the standard). After composing the descriptors to achieve competency, we work backwards to fill in the descriptors for levels 2, 1, and sometimes even 0. This results in a rubric that has two or three columns describing varying levels of low-quality work. Such descriptors reinforce deficit thinking and provide nothing for students to aspire to.
The effect of these two features—too many words and negative or unclear descriptors—is that traditional rubrics become autopsy reports that are only engaged with after the assignment has been submitted. When I taught middle and high school English, it was rare to see any of my students reading and engaging with my rubrics before or during the learning process—precisely when the rubric might have had a positive effect on their achievement. I now cringe when I consider the time I spent "going over" the rubrics for a project or paper, only to find that students weren't using them in the course of their work and that many missed the mark in the end. The fault, I realized, was mine and I needed to change if I wanted my rubrics to be used differently.
Creating a Rubric with Positive Language
The first step to creating a rubric students will use as a tool in the learning process is: simplify. Start with three boxes labeled, Basic, Competent, and Excellent, as shown in Figure 1. That's it—just three boxes, not 12 or 16 or 40. I also strongly recommend turning your rubric vertical and stacking the three boxes, but a horizontal arrangement would also work (see Leisen 2019a, 2019b). Other words can be used for the main descriptors; however, for this example I'll use Basic, Competent, and Excellent.
If you employ a 4-point grading scale, each box will also have a corresponding number—for example, 2 (Basic), 3 (Competent), and 4 (Excellent). The question "What about the 1 box?" often arises. I recommend having only three boxes because on a 4-point scale, a 1 does not meet basic requirements. A 2 is the first passing grade. If a student does not meet one of the descriptors in the 2-Basic box, then they will score a "1-Limited or Incomplete"—but filling a fourth box with negative descriptors isn't necessary or helpful. As long as the Basic box is very clear about what students need to do or show, if something is missing, a teacher can assign a grade of 1, "Not Yet," or "Work in Progress."
First, focus on the 2-Basic box and create a list of tasks or assessment descriptors that will demonstrate a basic level of achievement for which you could comfortably assign a passing score. It's imperative that students aren't assessed on whether or not they understand an overly complicated rubric, so teachers must write clear, positive indicators of success, as shown in Figure 2. Tell students exactly what they need to do, create, or show, not what we don't want them to do. If you want your students' lab reports to include an introduction that clearly states the topic they are investigating, then don't waste time and space with descriptors that say, "Missing introduction" or "Introduction does not clearly identify the topic." Instead, state that to meet the "Basic" requirements, their report needs to "Include an introduction that clearly states the topic of the lab." The clear message to students is, Do this.
Interestingly, I adapted the "basic" descriptors in the rubric shown in Figure 2 from a traditional rubric, where they had been listed in columns 3 and 4. This is another effect of filling our rubrics with negative words like partial, incomplete, and unclear. Because we start with such a low standard and need to differentiate between the columns, we end up listing basic, sometimes minimal expectations to score a 3 or a 4. Previously, I might have awarded a 3 for correctly including a conclusion on a report, but that's really a basic requirement. Now, that requirement is where it belongs, in 2-Basic. In other words, eliminating negative descriptors enables us to increase rigor and expectations.
Once you've listed requirements in the Basic box, start populating the 3-Competent box. These descriptors usually mirror what's in the Basic box but have a higher level of complexity. For example, students might be required to write an introduction to earn a 2, but for a 3, that introduction needs to establish a broad topic before narrowing and identifying a specific topic.
Following completion of the 3-Competent box, list descriptors for the 4-Excellent box. These should be the most complex and rigorous tasks, again stated in positive terms. Figure 3 shows what these boxes for this rubric for a lab report might look like.
An equal number of descriptors in each box isn't necessary; the "4-Excellent" box should usually have fewer descriptors than the 2–Basic or 3-Competent boxes, to avoid cluttering our rubric with inauthentic, unnecessary, or redundant expectations. If a "basic" requirement is to have "a properly formatted APA heading," then no "competent" or "excellent" equivalent of that element is needed. Also, keep in mind that even if a teacher writes seemingly crystal-clear descriptors, students will always benefit from examples to aid the learning process. Seeing exactly what is meant by "data is presented in a specific, logical order" will help students see what meeting these indicators and demonstrating excellence looks like in practice. Ideally, an educator would take these exemplars from past student work, but as a teacher, I sometimes created these myself to provide clarity of expectations to my students.
Using Rubrics Before and During the Learning Process
A rubric filled with clear, positive descriptors is ready to be put into action before and during the learning process. As part of this, it's crucial to accept that your rubric will evolve. As students are working with your rubric, you may identify descriptors that should be reworded, redistributed into a different box, or deleted altogether. Give yourself permission to start with an imperfect rubric, knowing that adjusting, modifying, and improving is one of the things great teachers do. (We often ask our students to take something "good" and make it "great"—and we must do the same at times.)
One question I often hear about this approach is, "What if a student accomplishes everything in the 4-Excellent box and everything in the 3-Competent box, but they're missing something in the Basic box? That learner has demonstrated excellence, but what grade should they receive if they haven't met one of the basic requirements?" In most cases, with a well-constructed rubric, this shouldn't actually happen. The Basic requirements will closely align with the Competent and Excellent descriptors, with the higher-level requirements building on the foundation laid by the Basic level. A situation where a learner has met requirements at higher levels but not fulfilled something at a lower level is unlikely to arise, but it is possible.
That said, if a student has demonstrated excellence but missed a basic descriptor that is easily corrected, teachers could justifiably award the 4. Sometimes, however, one of the "2-Basic" requirements is absolutely required (e.g., I will include transitional phrases to introduce new topics), although a student might have done all the other work needed to attain Competent or Excellent except for that one task or action. When this happens, I will assign the student a "W" (Work in Progress), meaning "You aren't done yet, and you need to fix this"—with the expectation that the student will make the required revisions within a reasonable grace period. This could also be indicated by "NY" (Not Yet), "NG" (No Grade), or a similar abbreviation.
If the required revision is never completed, then a teacher may decide to change the "W" to a score of 1 (basic requirements have not been met), but this is a situation I work to avoid. I've found it's extremely rare for a student who has met some or all of the "excellent" descriptors to not fix a deficiency to demonstrate their learning. Often only a brief tweak or addition is needed, and most students are happy to have the opportunity to make the correction and turn in a better assignment. Keep in mind that these rubrics are designed to raise students' performance, not penalize them for technicalities. And it is imperative to clearly communicate how our rubrics will be used—during the learning process to identify and make needed revisions or improvements.
Indeed, when we don't initially assign a grade to an insufficient product, and instead ask the student to make revisions or improvements, the rubric becomes a tool for feedback, rather than an autopsy report. If you are exploring ways to provide reassessments or multiple opportunities for students without going through a complete redo, this procedure offers a way. An entirely new assessment, new project, or a "retake" isn't necessary; instead, you can highlight omissions and areas to be improved while providing a chance to make it right. If rubrics are to be tools of assessment during the learning process, then we need to incorporate such feedback and opportunities for continual improvement.
Finally, because we use positive descriptors, it's very clear to the students what they need to do. The rubric becomes a tool for self-assessment to be applied during the learning process. Students can also use the rubric to assess examples provided by the teacher, which can lead to rich conversations about achievement, expectations, and what exactly phrases like high-level thinking mean. Teachers might even leave space on a rubric for students to come up with descriptors: What do they think would show basic, competent, or excellent achievement?
The end result is a living rubric that provides feedback during the learning process—a process now centered around students.
