What is a rubric?
A rubric is a simple but powerful document used for assessing a student’s complex product and/or performance. In a nutshell, a rubric makes assignment expectations very clear by: 1) writing down the essential elements required by the assignment; and 2) describing the levels of quality for each element (from excellent to poor). A rubric gives assessors and students a common understanding of what level of quality defines excellence. It helps assessors make dependable judgments about the quality of students’ work.
A "good" rubric
A good rubric is more than just a way to ensure consistent grading: it’s also a planning tool for teachers. Because it provides a way to track student mastery of important learning objectives, it’s very useful in helping teachers plan next steps for instruction. A good rubric is also a teaching tool. By clearly defining the criteria for excellence, the very rubric used to assess students also helps students to self-assess, plan how to perform better next time, and track their own progress over time.
A good rubric effectively defines the standards to be met and communicates expectations clearly. It describes the task(s) or project(s) to be completed, the criteria for judging the task or project on each required element, and the scale from “perfect” to unsuccessful on each required element. It has enough detail without having too much detail.
Effective rubrics have the following characteristics in common:
- The elements for an excellent performance are identified: Each thing you identify as being important enough to assess should be considered an element. For each element, develop a scale with ratings that describe it.
- Each element is discrete: Discrete elements ensure that the rubric is assessing what you think it’s assessing. For example, an element called “Quality of writing” in a rubric for an essay could actually be a combination of elements that should be broken down more completely into, for example, “Mechanics” and “Organization.” You will know that you’ve inadvertently lumped discrete elements together for assessment if it becomes confusing to assign a rating or feels difficult to give specific performance feedback to the learner.
- Each element uses only enough descriptive labels to identify behaviors for the range of expected performances: The rating scale for each element should be distinct, descriptive, and cover the full range of expected performances. It should clearly distinguish one level of performance from another with the use of descriptions that include specific and observable behaviors. While some elements are best assessed using a simple two-rating scale (as with a simple yes/no distinction), other might require several distinct ratings. It rarely makes sense to use the same rating scale for every element.
- Everyone who will use the rubric understands it the same way: In order for a rubric to act in its intended fashion, as a baseline for reflection by both learners and teachers, it needs to communicate the same message to all users. Use specific “plain language” terms. Avoid jargon and words with weak or several meanings. Show your rubric to others and get feedback. Better yet, collaborate with others (educators and students) to create your rubric.
- Feedback is crucial: The true value of a rubric is less in the numeric score it provides, and more in the information it provides about how to improve performance on each element. Nonetheless, we often need to have a score on the assignment. The best we can do here is to assign a weighting system that allots numbers of points based on the importance of each element, and then combine the scores of different elements to get a total. But we can seriously improve upon the value of this final score when we also provide written feedback, which brings the attention back to what's most important: student performance and how to improve upon it. A good rubric includes a place for assessors to give written feedback.
Let's look at a sample rubric
Here's a typical rubric. How does it stack up against the common characteristics of a good rubric?
Adapted from an iRubric rubric found here
You can see that while there are clear descriptive labels that would be commonly understood, the elements are jumbled together in an artificial fashion; Not too many students would fall squarely into one of these performance levels. Additionally, because each level of performance has been assigned a somewhat random point spread, there is a great deal of room for interpretation on the part of the assessor. Last, there is no space for the assessor to give personalized feedback.
Let's reorganize this rubric and take another look:
Here's the very same set of descriptive labels, this time reorganized into discrete elements. Now it's possible to assign points on performance for each learning objective individually. The point spread is no longer random, which means that grading becomes consistent. Last, there is space for the assessor to give personalized feedback.
Source: iRubric rubric found here
When you design your own rubrics, keep the common characteristics of good rubrics in mind. The benefits of a good rubric are clear: assessors and students a common understanding of what level of quality defines excellence, and assessors are able to make dependable judgments about the quality of students’ work. Teachers use rubrics to inform next steps for instruction, and students use them to to self-assess, plan how to perform better next time, and track their own progress over time. A good rubric helps everyone!