Rubric Creation and Sharing
What is a Rubric?
A rubric can be defined as “a scoring scale used to assess student performance along a task-specific set of criteria” (Mueller, 2010). A rubric is a criterion-referenced measure, meaning that mastery or success is determined by how performance aligns with a set of pre-selected skills and/or knowledge.
As a rule, “each rubric has at least two criteria and at least two levels of performance” (2010). A common rubric construction includes: pre-set criteria (skills and/or knowledge) listed on the left side of the grid, standards for what merits a certain level of performance in each grid cell, and levels of performance across the top of the grid.
General rubric layout:
|Lowest level of performance
|Progressively higher level
|Progressively higher level
|Highest level of performance
|Standards for criteria #1 at lowest level of performance
|Standards for criteria #1 at second lowest level of performance
|Standards for criteria #1 at second highest level of performance
|Standards for criteria #1 at highest level of performance
A very basic example rubric:
|Frequent grammatical and punctuation errors
|Some grammatical and punctuation errors
|Few grammatical and punctuation errors
|Used fewer than 4 references
|Used 4 references
|Used more than 4 references
Rubrics are formative by nature; they can be strictly formative or a combination of formative and summative if grading is involved. When used for grading, the total number of points can be determined in a variety of ways, including:
- All criteria can have the same number of levels of performance and thus the same score range (i.e. 1–4).
- Each criterion can have the same number of levels of performance but different points assigned to each level (e.g., one criteria gets 0 points for “poor,” 1 point for “good,” and 2 points for “excellent,” while another gets 0 points for “poor,” 2 points for “good,” and 3 points for “excellent”).
- It may be desirable for one criterion to factor more heavily into the total score. In this case, a weight can be assigned to the score given for this criterion (e.g. the total score is multiplied by 2).
Levels of Performance
Levels of performance in a rubric are used to determine how a student performed on a given skill or knowledge area. In the example rubric, the three levels of performance are “poor,” “good,” and “excellent.” They provide:
- “Clearer expectations,” both for students and for teachers, especially if the rubric is distributed prior to completion of the assignment
- “More consistent and objective assessment,” which is particularly important in courses where different instructors teach different sections
- “Better feedback,” which is beneficial for student learning and improvement of performance (2010)
- The number of levels of performance is completely up to the rubric creator. However, the more levels are used, the more difficult it is to truly distinguish between levels and to generate descriptors for each level. Thus, it is rarely necessary to use more than 4 levels.
- Particularly with holistic rubrics, you can use levels as you would with a checklist, such as “satisfactory” or “unsatisfactory” or “check” or “no check” (2010).
- While not required, descriptors are useful for giving information on what constitutes a certain level of performance for each knowledge/skill. In the “Research Report” example, “frequent grammatical and punctuation errors” is the descriptor for what constitutes a “poor” level of performance on the “composition” skill.
- Descriptors are a key part of using rubrics as formative assessment tools. By including these, students gain a better idea not only of where they stand on various criteria, but also what they need to do to improve their performance.
Types of Rubrics
Analytic rubric (refer to "Research Report" rubric for an example)
- Analytic rubrics “articulate levels of performance for each criterion” so that performance can be assessed “on each criterion.”
- Better in situations where one utilizes multiple criteria and/or weighting.
- The greater amount of detail that is offered is better for formative assessment.
- Holistic rubrics assign “a level of performance by assessing performance across multiple criteria as a whole.” They are most useful when thorough review is not necessary or when it is more logical to review the assignment as a whole rather than by each criterion. The following is an example of a holistic rubric:
A more detailed example of and guidelines on holistic rubrics is Facione and Facione’s "Holistic Critical Thinking Scoring Rubric," which can be found at http://www.insightassessment.com/content/download/1078/6171/file/Rubric%2520HCTSR.pdf.
After the Rubric is Completed
- Review the rubric: Will this measure what it is intended to measure? Will it do so validly? If possible, have the rubric reviewed by someone familiar with rubric construction.
- Consider distributing the rubric to your students prior to the assignment due date. This helps students understand what constitutes good performance, which in turn allows them to learn specific ways to improve their performance.
- Remember that as with assessment as a whole, rubrics are not static projects. After using your rubric to evaluate an assignment, reflect on whether it needs revising. Reflection points include: Do the number and labels for levels of performance make sense? Did the descriptors accurately fit each level?
- Share your rubric with other faculty members! Rubric creation can be an arduous process; having an example rubric can help greatly when determining style and content. Additionally, your rubric may work perfectly for someone else’s assignment, which eliminates redundant work. Finally, rubric sharing is perfect for courses that have different sections taught by different faculty members. The use of a common rubric(s) helps to ensure better standardization of grading across course sections.
Mueller, J. (2010). Rubrics. Retrieved from http://jonathan.mueller.faculty.noctrl.edu/toolbox/rubrics.htm.