Vicki Spandel, who trains teachers in performance assessment and improving student writing, often asks her workshop participants to do a familiar task. She provides them with a student's paper and asks them to grade it (but without giving them common criteria as a basis to score the paper).
Invariably, the papers will receive marks ranging from `A' to `D' or `F,' says Spandel, because teachers frequently differ in their view of what dimensions of the paper are more or less important. Some teachers, for example, place a heavy emphasis on ideas and creativity; others are impressed by mastery of writing mechanics.
The grading exercise has important implications, says Spandel, a senior research associate with the Center for Classroom Assessment at the Northwest Regional Educational Laboratory. If teachers themselves are unclear about what constitutes quality work, students are likely to be, too. And students who are unsure of what is considered quality work don't know what to aim for and can't be sure their products measure up. Moreover, if they subsequently receive a low letter grade but little specific feedback, they are unlikely to know how to improve their paper or presentation to get an `A' the next time.
This dilemma is contributing to a burgeoning interest in rubrics. Although experts vary in how they define the term, most see rubrics as providing criteria that describe student performance at various levels of proficiency. A well-written rubric can help teachers score students' work more accurately and fairly; it can also give students a better idea of what qualities their work should exhibit. When teachers use criteria spelled out in the rubrics to give students specific feedback on their performance, students have a far better picture of the merits and demerits of their work than grades alone would provide, experts say.
Rubrics "give students an understanding of the meaning behind the grade," Spandel says. "There is only one grade that's significant to a student, and that's an `A.' If they don't get the `A,' they're disappointed, but they don't know what to do about it. The criteria [incorporated in the rubric] tell you what to do about it."
The current interest in rubrics comes as educators throughout the United States try to define learning outcomes, set standards for student achievement, and expand the use of performance assessments to evaluate what students know and can do. For about a decade, educators have been "zeroing in a whole lot more on sharply defining the achievement targets that we want kids to hit," says Rick Stiggins, director of the Assessment Training Institute in Portland, Ore. "One way to do that is through the design of specific performance criteria for use in a performance assessment context, and that's what rubrics do." Stiggins sees the interest in rubrics as "a manifestation of a broader interest in being more focused on what it is that we expect students to do."
Of course, individual teachers have always kept in mind their own criteria for evaluating student work, says Bena Kallick, an education consultant. But the use of rubrics requires teachers to be much more precise about what these criteria are. "We're asking people to refine how they're going about doing this," she says. Using rubrics requires "being a lot more systematic about what formerly has been a more intuitive, less rigorous process," adds Joan Herman, associate director of the National Center for Research on Evaluation, Standards, and Student Testing at the UCLA Graduate School of Education.
As the use of performance assessment spreads, moreover, educators are increasingly developing common scoring rubrics that are used in many classrooms and schools. As a result, experts say, the criteria for what makes a good essay or project are becoming a shared standard, rather than an individual teacher's decision about what constitutes quality work.
Complexity Varies
Although all rubrics contain criteria to describe student performance, they may differ in important ways. A rubric may be little more than a checklist (i.e., the teacher checks whether a student did or didn't display a behavior), or it may contain many facets of a performance, each with its own criteria and levels of attainment.
An example of the latter is the rubric that Spandel uses to assess student writing. Under the rubric, student papers are assessed on six dimensions: ideas and content, organization, voice, word choice, sentence fluency, and conventions. Each paper is assessed on a five-point scale for each dimension.
On the dimension of organization, for example, a "5" paper is organized in a manner that "enhances and showcases the central idea or storyline. The order, structure, and presentation of information is compelling and moves the reader through the text." The rubric contains even more specific language about how this is achieved. For example, the paper will probably have an inviting introduction and thoughtful transitions, as well as a satisfying conclusion that leaves the reader with a sense of resolution.
By contrast, a paper earning a "1" in organization contains writing that "lacks a clear sense of direction. Ideas, details, or events seem strung together in a loose or random fashion—or else there is no identifiable internal structure." The rubric further elaborates that such papers may have awkward pacing, confusing or missing connections between ideas, and no real "lead" or conclusion.
When teachers are trained to use such criteria in evaluating student writing, their scoring becomes highly reliable, says Spandel. That is, teachers who might vary considerably in their evaluation of a paper without common scoring criteria are much more consistent in their ratings when using the rubric. "For teachers who are able to assimilate and accept the criteria, we have tremendously high reliability—lots of consistency," Spandel says.
Language Is Crucial
Developing rubrics is usually hard, time-consuming work, experts say. The best way to craft rubrics is for teachers and others to meet regularly to decide on the relevant dimensions of a performance and draft possible criteria, try them out with their students, and revise them as needed. In discussing the best criteria to evaluate a particular performance, teachers should draw upon prior students' work, standards within the disciplines, and other sources. (See "How to Begin" for ideas on how to start developing a rubric.)
Particularly challenging is the task of finding just the right language to describe qualities of student performance that distinguish between mediocre and excellent work. "It's often very hard to articulate in words what we've kept in our heads for many years," comments Doris Sperling, who recently retired from her job as classroom assessment specialist in Ann Arbor, Mich.
The rubric Spandel uses to evaluate writing has been through 12 revisions. Originally, criteria for sentence fluency had teachers counting the number of complex and compound sentences in student work. "Hemingway wouldn't have gotten very far" with such criteria, Spandel points out. Now the criteria refer to such elements as variety of sentence beginnings and sentence length, and how the sentence plays to the ear. The criteria "are constantly being refined and reshaped just a little bit to more closely resemble what we actually see in student writers at work," Spandel says.
It's also difficult to find the right language to describe the differences among student performances at varying levels of quality. Too often, rubrics suggest only that poor work has "less" of the same types of qualities as better work. "It's ultimately lazy just to use comparative language; it stems from a failure to keep asking for the unique features of performances," says Grant Wiggins, director of programs for CLASS, the Center on Learning, Assessment, and School Structure.
Wiggins suggests a rule of thumb for ensuring the quality of rubric criteria: If a student can achieve a high score on all the criteria and still not perform well at the task, you have the wrong criteria. For example, criteria for writing an analytical essay might focus on organization, mechanics, and accuracy. But if the finished piece doesn't have an impact on the reader through its novelty or insight, it hasn't really achieved its purpose, Wiggins points out.
Benefits to Teachers, Students
The benefits of well-written rubrics range from their usefulness as a staff development opportunity to their impact in the classroom.
The process of developing rubrics requires "one of the most important professional conversations that teachers can have," says Jay McTighe, director of the Maryland Assessment Consortium. Rubric development "requires that people think very clearly about what they mean by `quality work,'" he says. Bringing to the table their own notions of quality and experiences with past students, teachers begin to reach some common agreement about the attributes of good work. As a result, students benefit, because teachers' expectations and standards become more consistent and uniform.
Within individual classrooms, teachers using rubrics say the scoring criteria can be helpful in several ways. The most common benefit reported is that students become better attuned to the characteristics of quality work—and are therefore in a better position to produce it.
Pomperaug Regional School District 15 in Connecticut, which serves the towns of Middlebury and Southbury, began working with rubrics eight years ago and has since developed hundreds of what the district calls "classroom assessment lists." These lists explain, in language that students can understand, the characteristics of, for example, a good graph, says K. Michael Hibbard, assistant superintendent. As these lists become more widely used, "Teachers are amazed at how kids are finding things to improve in their work and making it better the first time," he says. "It has really helped us to focus in on the criteria for excellence before we get the product," agrees Linda Van Wagenen, who teaches 8th grade at Memorial Middle School in Middlebury.
Not only can students understand and internalize the criteria in a rubric, they can also help to develop rubric criteria, use the criteria to assist their peers in revising work, and assess their own work.
Julie Folkert, a teacher at Highmeadow Common Campus in Farmington Hills, Mich., says her 5th graders are capable of generating sound rubric criteria, although she often modifies or supplements their suggestions. For example, Folkert, who has her students respond to literature by writing in their journals, asked students to help develop the criteria for good journal work. Using models or exemplars that exhibit some of the characteristics of good entries, she pushed and prodded students to come up with the most relevant characteristics. Eventually, Folkert and her students came up with the criteria for an entry that would be scored on a scale of 1–5.
Rubrics have proven especially helpful in teaching writing. The prevailing theory on writing suggests that students should sometimes have the chance to critique one another's work before it is revised and turned in. But without precise language to describe facets of the writing, the feedback too often consists of comments such as "`you misspelled two words,' or `your margins are too narrow,'" says Nikki Elliott, a resource coordinator in Vancouver, Wash. Seeing the criteria spelled out "helps students to be more specific in their feedback," she says. As a result, "kids get real excited about and understand the revision process better."
Teachers sometimes ask students to use rubrics to evaluate their own work—either to check its quality before turning it in or as part of an official self-assessment that goes in the grade book. Folkert, for example, has her students rate their own work on major assignments before turning it in. She records both the student's self-score and her own evaluation in the grade book—and usually finds the two evaluations are consistent. "I haven't had a big problem with discrepancies," she says. If students give themselves an inflated score, a teacher can usually get them to revise it by asking them to re-examine their work and show evidence that they've met the relevant criteria, experts say.
Folkert says that by using rubrics to assess student performance, she and her pupils are more clear about how student work will be evaluated. In the past, she might have graded the same paper differently if given it at separate times, she admits. Now, she's much clearer up front about how she'll evaluate student work. As a result, students—perhaps for the first time—know what is required of them to do high-quality work. "I definitely see an impact on lower achieving students," she says, because they are able to see the concrete ways they can improve their work to meet a high standard. "This is an incentive for them to push themselves harder."
How to Begin
How to Begin
How do educators develop the criteria for evaluating student performance that make up a rubric? An ASCD book, A Practical Guide to Alternative Assessment, by Joan Herman, Pamela Aschbacher, and Lynn Winters, suggests a process.
Investigate how the assessed discipline defines quality performance.
Gather sample rubrics (which can be adapted) as models.
Gather samples of students' and experts' work that illustrate a range of quality.
Discuss the characteristics of the work that distinguish good from poor examples.
Write descriptors for the important characteristics.
Gather another sample of students' work.
See if the criteria help to make accurate judgments about students' work.
Revise the criteria.
Try the criteria again until the rubric score captures the “quality” of the work.
Copies of A Practical Guide to Alternative Assessment (stock no. 611-92140) are $10.95 each. Also, Assessing Student Outcomes: Performance Assessment Using the Dimensions of Learning Model (stock no. 611-93179) contains numerous examples of rubrics. It costs $13.95. Both books can be purchased from ASCD, 1250 N. Pitt St., Alexandria, VA 22314. Telephone: (703) 549-9110. Fax: (703) 836-7921.
