Rubrics: Friend or Foe?

☐    Do teachers in your school feel like students misunderstand assignments?
☐    Are you worried that some students are graded differently than others?
☐    Do your students (and their parents) often debate grades?
☐    Do your teachers complain about spending way too much time grading?

If you checked “yes” for any (or all) of these questions, your current rubrics might be working against you rather than for you. Let’s find out how to make rubrics more effective!

teachersatwork

An Overview

A rubric, at its most basic, is a tool to define the expectations of a particular assignment with ways to indicate different levels of effectiveness in meeting those expectations. It is this last aspect, the gradations of quality, that differentiates a rubric from its simpler counterpart, the checklist. Even within the category of a rubric, there is lots of variety. General rubrics describe expectations for a skill that can be generalized across assignments. Task-specific rubrics are, as their name suggests, geared toward a particular assignment.

The biggest distinction within rubrics, however, is between holistic and analytic rubrics. Holistic rubrics provide a single score to summarize a student’s performance on a given task, whereas analytic rubrics provide several scores for the task, one for each different category being evaluated. See examples of each kind here.

Anatomy of the Rubric

There are four main components of a rubric: task description, scale, dimensions, and descriptions of the dimensions (for more on these, see: Introduction to Rubrics). To illustrate, for an assignment from a high school literature classroom:

  • the task description is a ten-page research paper

  • the scale is excellent, competent, or needs work

  • the dimensions of the assignment include organization, clarity, argument, and grammar

  • the descriptions outline what a student needs to do to get a certain score for a certain dimension

It is with these components themselves that we already start to see how a rubric could be clear and helpful or murky and frustrating.

Sample Rubric

Screen Shot 2018-08-01 at 10.49.19 AM.png

When Rubrics Work

What good can rubrics really do? A rubric, after all, is a sheet of paper that is all too easy for students to glance at once and then shove to the bottom of an overstuffed backpack. However, rubrics are more important than they seem. Every time you make a rubric, you articulate your educational philosophy; you indicate where students are, where they can go, and how they will get there. As English Professor Emeritus Peter Elbow explains, by offering your students a rubric, you're telling them:

“You deserve to know more about my values as a particular reader: Here are the aspects of writing that I believe are most central to my idea of excellence.”
— Peter Elbow, Writing Assessment in the 21st Century: Essays in Honor of Edward M. White

Students Get More Feedback

Rubrics have so much potential that even minimal engagement with them can yield great benefits. As educator Heidi Goodrich Andrade explains, “If I were to simply circle boxes on a rubric and give it back with an assignment, I would still be providing more feedback about the strengths and weaknesses of the work than if I had just assigned a letter grade, and it would not have taken me any longer.” Offering more feedback to students yields even more benefits. In a 2015 study, rubrics that included individual advice on how to improve led to students with “significantly better performance in planning scientific experiments”; these students “perceived themselves as being more competent and were also more accurate in their self-evaluative performance judgments.” That’s exactly what The Graide Network hopes to achieve by pairing rubric-based scores with personalized, qualitative feedback about each student’s areas of strength and areas for growth. As educator and author Carol Jago points out, “feedback cools quickly,” so anything you can do to speed up the availability of feedback for students matters.

Students Develop Metacognition

At their best, rubrics can encourage students to think critically about their own work and scores. It’s a fairly common sentiment among students that teachers “give” certain grades rather than students “earning” them. These word choices reveal that students often don’t feel involved in the assessing process; they feel as if it is being done to them rather than with them or by them. Having a clear rubric lets students in on the decision making process. Students better understand the criteria of each assignment, so they can begin to look at their assignments with the critical eye of a teacher, allowing them to understand their past mistakes and fix their current ones. One teacher found that students who assessed themselves with a rubric as they worked on a reading comprehension assignment scored higher on a content knowledge quiz afterward than students who did not use a rubric. The metacognition, or awareness of one’s own learning strategies, that can result from familiarity with rubrics produces both more insightful learners in the long-term and less frustrated students when report cards come around.

Teachers Make Better Decisions

As Professor Timothy Brophy explains, rubrics are also helpful tools for teachers because they encourage “criterion-referenced rather than norm-referenced” scoring. In essence, criterion-referenced scoring means that while grading, teachers ask themselves whether a paper meets the criteria of a certain score rather than consciously or unconsciously assigning grades based on the quality relative to the other papers in that class. In a vacuum, norm-referenced scoring might not seem that bad, but it seriously hinders teachers’ ability to meaningfully compare scores across their own classes and across classes taught by several teachers. In addition, consistently comparing each paper to a fixed set of standards decreases the chance that a teacher will grade papers differently based on student preference, number of coffees consumed, decision fatigue, or any other factor that shouldn’t affect student grades. Lastly, important nationwide tests like the AP English Language and Literature exams use criterion-referenced scoring systems, which should encourage teachers to grade similarly.

Students Are Treated More Fairly

Rubrics create a common language to speak about academic work, make the learning goal clear, and level the educational playing field. In this era of increasing diversity, this last point is an important one. Researchers Kenneth Wolf and Ellen Stevens point out that first-generation students, as well as immigrant students and students from minority backgrounds, often don’t come into school with the same academic assumptions as their peers. The clearer the learning target for these students, the more likely they are to meet it. Since unconscious bias can also work against students most dissimilar to their teachers, comprehensive rubrics can also encourage fairness for all students.


Hazard icon.png

Where Rubrics Go Wrong

Although rubrics have many potential benefits for both students and teachers, a poorly constructed rubric can do more harm than good. Luckily, there are a number of predictable pitfalls that can be easily avoided if you know what to look out for.

Extreme Scales

Although the scale may seem like a fairly straightforward decision to make, there are many different options. One teacher might use a scale of A-F, another might use the scale 1 through 5, and yet another might choose to use the words excellent, competent, and needs work as the scale.

There is no right answer when it comes to using letters, numbers, or words for your scale, but there are a few choices that could lead to unproductive rubrics. Scales quickly become ineffective when there are too many or too few options. For example, except for very basic assignments, using just a pass/fail scale will not provide you or your students with very much helpful information. On the other hand, using a scale of 1 to 20 will make it harder to justify small differences, such as giving a student 13 rather than 14. Having three to five possible options on a scale is often effective and four is sometimes considered ideal. That being said, one of the largest working scales in effect right now is the College Board AP English essay holistic grading scale from 1 to 9. Even so, this scale is still essentially a 5-point scale broken down into pairs: 9/8, 7/6, 5, 4/3, 2/1.

Inconsistent Language

There are a number of considerations as well when it comes to descriptions of each dimension; parallelism and consistency are your friends here. For instance, let’s say that your current rubric breaks down the dimension of citation in the following way:

  • level 1 describes a paper with incorrect citations

  • level 2 is a paper with sometimes accurate citations

  • level 3 is a paper with many correct citations

  • level 4 is a paper with complete citations

These levels have obvious differences, but the language isn’t as clear as it could be. Choosing to describe either amount, frequency, or intensity across all parts of your scale will help to keep the language consistent automatically.

A revised version of the same rubric might look like this:

  • level 1 is a paper with few correct citations
  • level 2 is a paper with some correct citations

  • level 3 is a paper with many correct citations

  • level 4 is a paper with all correct citations

The parallel language (using “correct”) of each description shows how all the levels relate to each other, easing comprehension for anyone who encounters the rubric.

Negative Language

Another consideration when writing descriptions is valence. As researchers Tierney and Simon explain, using purely negative terms to describe lower levels on the scale can discourage students, particularly younger ones. Using words like “little,” “slightly,” or “seldom” will be less discouraging (and likely more accurate) than words like “none” and “never.”

Vague Language

Lastly, the criteria for each level should be observable and measurable in some way. If I claim that students are supposed to “know” something, it is important to indicate what precisely that knowledge will look like. For example, I could specify in the rubric description that a paper with an argument that falls under Level 4 (always strong) must have a thesis that is defendable, specific, and evidence-supported. A paper that falls under Level 3 for argument (usually strong) will have a thesis that is only defendable and evidence-supported. Without this kind of specificity, it would be easy for a student to argue that their paper meets Level 4 requirements rather than Level 3 requirements since “always” and “usually” can be vague and subjective terms.

If at First You Don’t Succeed...

Good rubrics are hard to write. Here are two simple strategies to make sure your rubrics are working well:

1) Calibrate with colleagues

An effort to construct effective rubrics is clearly important, but it can be difficult to tell how effective a new rubric is without discussing and revising it with colleagues. As Professor Timothy Brophy suggests, these sessions can be used to establish minimum scores for passing and develop benchmark papers for the rubric’s scale. Selecting six pieces—two high scores, two medium scores, and two low scores—is often a helpful way to show the students what papers look like on each part of your scale. College Board provides three anchor pieces (high, medium, and low) for all previously released AP essays as an example of what this might look like.

If you pass out the benchmark pieces to your department without scores and have the other teachers score them blindly, this meeting can turn into a productive discussion about how to increase grading reliability across classes. A similar brainstorming and revision session can be helpful if you are trying to adapt previously existing rubrics—whether from another teacher or from the The Graide Network’s rubric creator— for a new project.

2) Review actual scores

Discussing rubrics after their implementation can also help teachers analyze grade breakdown. If 75% of your students are receiving a two on a four-point scale for organization, that might be an indication that you are grading too harshly and/or the students need a review on how to arrange their thoughts. The Graide Network provides lots of this data automatically for teachers along with our qualitative feedback. Having this quantitative data makes it significantly easier to see class-wide patterns; without a rubric, you might not realize that you are writing comments about organization on 75% of your student papers!

 

To learn more about getting high quality feedback for your students and high quality data for your teachers, we'd be happy to talk. 

 

Conclusion

There are unlimited different ways to construct rubrics. However, not all rubrics are created equal. Rubrics with reasonable scales, consistent language, specific requirements, and positive descriptions are more helpful for students and teachers alike. You can take an effective rubric to the next level by involving other people; refine rubrics with other teachers before their implementation, and take a day to talk your students through a new rubric. You can even go further and grade example essays using a rubric during class, an invaluable way to clarify learning goals. By clarifying learning goals, we bring students closer to them, one rubric at a time.