This page explores the characteristics of rubrics that allow teachers to define clearly what quality looks like, so that teachers are able to explain "quality" to all who need to know.
What makes a good rubric? How do we know one when we see it? If we type "rubric" into any Internet search engine, we get from 1,000,000 to13, 000,000 hits. So, how do we know which to choose? It's impossible to answer these questions without first considering how we want to use rubrics in the classroom. On this page, we noted the following ways to use rubrics:
You will notice that this list includes both assessment for and assessment of learning uses. Both are helpful in the classroom. To be able to fulfill these uses, rubrics must:
Teachers and researchers generally agree on the features of high-quality rubrics (Johnson, 1996; Moskal, 2000; Perlman, 2004; Popham, 2002' Rohrmann, 2003; Tierney & Simon, 2004). From our own experience (bolstered by these sources and work with teachers and students) we have developed a Rubric for Rubrics. Figure 2.1 shows a summary.
Below we have developed a section on Rubric for Rubrics to evaluate rubrics for use in the classroom, not for use with large-scale assessments such as state or provincial assessments. Although many features of quality would be the same for both uses, large-scale rubrics often end up with features that would be counterproductive in a rubric intended for classroom use. For example developers of rubrics for large- scale uses frequently emphasize a quick, overall picture of student performance- no detail Rubrics used in the classroom, on the other hand, often need to provide detailed diagnostic information to inform day-to-day instructional decisions.
The Rubric for Rubrics is intended to be used with general not task-specific, rubrics. We emphasize general rubrics because they help define those learning targets often least clear in curriculum documents (such as assessing reasoning proficiencies, performance skills, or products). Also, they can be given to students in advance of the assessment to practice with.
The Rubric for Rubrics has two criteria- Coverage/ Organization and Clarity.
The content of a classroom rubric defines what to look for in a student's product or performance to determine its quality: what will "count." Teachers and students use this content to determine what they must do to succeed. What students see is what you'll get.
Indicator 1A: Covers the right Content
The first thing to think about is the extent to which the rubric covers all the important features of work or performance that really do add to quality and leaves out features that do not relate to the learning target at hand, or that are not important. For example, the rubric for a lab report would include floor- the entire report relates to the original hypothesis and is not filled with irrelevant details- and the nature and sufficiency of the manner in which results are displayed. Such a rubric would not include length. Indicators of having the right content include the following:
Please note that you can accurately judge the soundness of a rubric's content only if you are well versed in the content domain.
Indicator 1B: Criteria Are Well Organized
The list of features that describe quality should be as concise as possible and organized into a useable form. This often involves identifying and grouping similar features into criteria and making sure that the relative importance given to each criterion represents its relative contribution to the quality of the product or performance as a whole.
Indicator 1C: Number of Levels Fits Targets and Uses
Finally, the number of levels needs to be appropriate for the intended learning target and your use of the rubric. Can users distinguish among the levels? Does the number of levels allow you to distinguish different levels of quality for that you (or the students themselves) can adequately track student progress?
A classroom rubric is clear to the extent that teachers, students, and others are likely to interpret the statements and terms in the rubric the same way. A rubric can be strong on the criterion of Coverage/ Organization but weak on the criterion of Clarity- the rubric seems to cover the important dimensions of performance, but doesn't describe them very well. Likewise, a rubric can be strong on the criterion of Clarity, but weak on the criterion of Coverage/ Organization- It's very clear what the rubric means, but it is not focused on the right criteria.
Indicator 2A: Levels Defined
The key with clarity is to define levels so transparently that students (and teachers) can see precisely what features of work cause people to agree that work is strong, medium, or weak. The instruction usefulness of any rubric depends on the clarity of level descriptions.
Read the question referring to Clarity in Figure 2.1. Teachers generally agree these things are important. However, our objection to simply counting number of frequency may require explanation. Are five weak references on a research report better than three strong ones? Does an introductory paragraph need specific number of sentences to be effective at drawing the reader in and setting up the topic? Quantity and quality are not always synonymous.
Counts are appropriate for certain criteria. For instance, students might monitor the frequency or consistency with which they adhere to various classroom rules, such as taking out needed materials or cleaning up a work area before being asked. Some academic learning targets might be monitored through use of counts, such as errors in oral reading or number of out-of-tune notes in a musical performance.
But, for most criteria, quality is not the same as quantity. If counts are used, make sure that counting is the best indicator of quality.
Indicator 2B: Levels Parallel
Rubrics should include a parallel feature of work on each level. For example, if you find that a rubric for playing the violin contains "Lackadaisical bowing" as one descriptor of a middle level performance, then a statement about the quality of the bowing must be included at the strong and weak levels as well. If the descriptor is not referred to at other levels, the levels are not parallel.
Whether a rubric should fit on one page depends on the use to which it will be put and the complexity of the learning target being captured. The 6-Trait writing rubric, for example, fits on seven pages- one for each trait. Each trait needs enough descriptive detail so that teachers can be consistent in judging quality and students can understand what the trait means. Could it be more concise? Maybe but we don't think it could be reduced to one page.
Keep the following two points in mind. First, you don't have to evaluate all criteria for every piece of work. This is especially true during learning. Evaluating criteria separately can often give you a better picture of student achievement and can give students focused feedback to use when improving their work.
Second, once you internalize the rubric you can put a brief version on a single page (or less) as a reminder, keeping the whole rubric handy in case questions arise.
Conciseness must not trump clarity in a rubric.
Consider two performance assessments:
1. A primary teacher is assessing student concepts of print by asking students to prepare a book cover.
2. A middle-school history teacher is assessing student ability to make a map by asking them to make a map.
Should the scoring guide for the book cover include features of the book cover itself, such as neatness and colorfulness? We think it should stick to concepts of print because the product- the book cover- is just the context for demonstrating achievement of the learning targets, in this case concepts of print. The scoring guide for the map, on the other hand, should consisted of features of the map itself- map key, accurate details, and proportional distances, for example- because these are the learning targets.
Include on the rubric features of the product to be created by students only when the product is the target of the assessment. Do not include on the rubric features of the product to be created by students if the task is merely the context for demonstrating mastery of the target. Here's why: If you were to records a score in your gradebook for "book cover" and this score included neatness and colorfulness as well as concepts of print, that score would lose its meaning vis a vis concepts of print.
Sometimes you may think, "If I specify exactly what a good product or performance looks like, won't I dampen individuality? Won't I cause all student work to look alike? How do I avoid restricting creativity?" Sometimes this manifests itself when you have to rate a performance is actually good (or vice versa). When either of these happens, it's time to revise the rubric. Rubrics should not restrict students into a single mold of quality.
As an example, an earlier version of the 6-trait writing rubric contained the trait Sentence Correctness- sentences needed to be complete (although being varied and tailored to the audience was also good). This trait, though important, sometimes led to an unjustifiable low rating when students used sentence fragments well. on review, the rubric's authors decided that being in control of sentence structure can craft sentences to support the ideas in the writing and select a structure that suits the audience, topic, and purpose for writing. Sometimes short, choppy, and fragmented works. Sometimes long, fluid, and complete is a better approach. Now the trait is called sentence Fluency.
Here's another example: consider a requirement that a poster have at least three colors to be judged effective. Why three? Can't a perfectly effective poster be black and white? Wouldn’t it better to say “the poster's presentation should enhance the ideas and catch, the viewer's attention"? There are many ways to accomplish this apart from number of colors.