"If you read carefully, a poem will teach you how to read it."
This comment came from Jake York, a poet and teacher who was facilitating a summer institute workshop offered by the Denver Writing Project. I immediately copied this statement into my writer's notebook. Jake had been helping us unravel the meaning behind a tricky poem. He kept encouraging us to stick with it and assured us we would be rewarded with greater understanding. Because my work as a district literacy coordinator means I lead teachers in carefully examining and assessing student writing, I began to wonder whether it is possible to read student writing in the same way Jake suggested we read poetry.
Can student writing "teach" us how to read it if we look carefully at what the writer has to say, where the writer shines in saying it, and what instruction would increase the writer's polish? And if so, can we use this technique of careful reading as a kind of formative assessment to guide instruction? And can the rubrics now so common in evaluating student writing help teachers read more deeply?
Moving Beyond the Score
Teachers use rubrics, scoring guides that define expectations at various performance levels, to guide their reading of student work. When any group of teachers meets to assess student writing, their first step is often to examine the rubric. Rubrics are meant to make the scoring of writing less subjective, but although this is a laudable goal, I have seen teachers in many scoring sessions wrestle to reach agreement on a particular score.
In my school district, which is heavily involved in using rubrics to examine student work, teachers often spend many hours scoring and cross-scoring samples of writing, determining which pieces best reflect various performance levels. Some good comes from this: We have raised our expectations for student writing, and we are more transparent about what we expect. But when our thinking doesn't move beyond reaching a score in these meetings, we are not taking advantage of what the writing can teach us.
Last spring, I worked with a group of about 20 teachers from grades K–12 in my district. We were meeting to align our district's benchmarks document with actual student writing samples. During one particular session, we decided to discuss one piece in a large group to clarify our own thinking. We chose the following 4th grade piece: My favorite place to play is my bedroom. My bedroom is painted extremely light pink. Also I have a peanut-colored carpet. It is full of meowing. It never stops! Most of the time I play by myself. My friend Megan comes over sometimes. When Megan is in my room we play Pokemon Stadium. Another thing we do is play with my multicolored cat. Her name is Maxie. We throw a lime green toy mouse for Maxie and she sprints after it. Megan and I also throw a ball for Maxie to get. We also play Conga sometimes. I play in my room at 3:30 pm with Megan. I like to play in my room because it is very fun and it is mine. I love to play in my room!
We spent about 15 minutes debating whether the student scored a 4, 3, or 2 based on our state-designed writing rubric (see fig. 1). Several teachers scored the paper a 2, eight teachers scored it a 3, and the rest gave it a 4. That means this student could either be marked as an advanced, proficient, or partially proficient writer. The inconsistency concerned me because if a student consistently scores below proficient on standardized assessments in my district, then the school must develop a learning plan, hold a parent conference, and discuss interventions for the student. So if we intend for rubrics to clarify, how did this group of teachers end up in a potentially high-stakes debate on this student? And why did we spend 15 minutes talking about a score without thinking about how this piece might inform our instruction for this writer? If rubrics are to be used successfully for formative assessment, we must look beyond the search for a score.
Figure 1. Holistic Writing Rubric for the Short Constructed-Response Task — Grades 4–10
What Student Writing Can Teach Us - table 1
Score Level | Content and Organization | Style and Fluency |
|---|---|---|
| 4 | Supporting details are relevant and provide important information about the topic. The writing has balance; the main idea stands out from the details. The writer seems in control and develops the topic in a logical, organized way. The writer connects ideas to the specified purpose. | The writer selects words that are accurate, specific, and appropriate for the specified purpose. The writer may experiment with words and/or use figurative language and/or imagery. The writer uses a variety of sentence structures. The writing is readable, neat, and nearly error-free. |
| 3 | The writer has defined but not thoroughly developed the topic, idea, or story line. Some supporting details are relevant but limited or overly general or less important. The writer makes general observations without using specific details or does not delineate the main idea from the details. The writer attempts to develop the topic in an organized way but may falter in either logic or organization. The writer connects ideas with the specified topic implicitly rather than explicitly. | The writer mostly selects words that are accurate, specific, and appropriate for the purpose of the writing. The writer uses age-appropriate words that are accurate but may lack precision. The writer uses simple but accurate sentence structures. Errors in language usage, spelling, and mechanics do not impede communication. |
| 2 | The writer has defined but not thoroughly developed the topic, idea, or story line; response may be unclear or sketchy or may read like a collection of thoughts from which no central idea emerges. Supporting details are minimal or irrelevant or no distinction is made between main ideas and details. The writer does not develop the topic in an organized way; response may be a list rather than a developed paragraph. Ideas are not connected to the specified purpose. | The writer sometimes selects words that are not accurate, specific, or appropriate for the purpose of the writing. Writing may be choppy or repetitive. Portions of the writing are unreadable or messy; errors may impede communication in some portions of the response. |
| 1 | The writer has not defined the topic, idea, or story line. Supporting details are absent. Organization is not evident; may be a brief list. Ideas are fragmented and unconnected with the specified purpose. | Much of the writing is unreadable or messy. Word choice is inaccurate or there are many repetitions. Vocabulary is age-inappropriate. The writer uses simple, repetitive sentence structures or many sentence fragments. Errors severely impede communication. |
| 0 | The response is off-topic or unreadable. | The response is off-topic or unreadable. |
Courtesy of Unit of Assessment, Colorado Department of Education. This material also appeared in When Writing Workshop Isn't Working, by Mark Overmeyer, 2005, Portland, ME: Stenhouse. Used with permission.
Using Rubrics Formatively
Holistic rubrics that give only one score—which evaluators often use for state and national testing—are particularly difficult to use for informing instruction because the assessor cannot separate out individual qualities of the writing. Each piece can only receive one score. Trait-based rubrics lend themselves better to differentiated comments, but assessors still must give a single score for each trait. So even if a piece has strengths, readers can't highlight strengths fairly in the push to give the entire paper one mark. But teachers can, and should, adapt rubrics to serve the purposes of guiding our instruction as well as the purposes of testing.
When the teacher group I lead in finding exemplars for our district finally moved the conversation away from scoring the 4th grader's paragraph, we engaged in discussion about what this writer did well and how she could improve. We all agreed that the student clearly stayed on the topic, provided sufficient supporting details, and showed some strong word choice. We soon began discussing instructional strategies, which is a crucial step in assessing student writing. Everyone agreed that although specificity and rich vocabulary were the writer's strong suit, she could benefit from thinking about how to organize details within a piece of writing. This experience helped us realize as a group that the goal of looking at student writing must be to develop strategies to help students write better.
Signaling Strengths and Weaknesses
Because our group began focusing more on how to help teachers see our thinking about each piece of writing, we had to devise a format that would show teachers that deeper thinking and give them a model for using a rubric to inform instruction. Our goal was to provide a more helpful document than a tidy, completed rubric. We created a chart on writing strengths and weaknesses (shown in fig. 2). This format draws from the state-created rubric (fig. 1) but shows how a teacher can annotate student work in such a way as to discover what to teach a student next—in other words, to hear what the writing has to teach.
Figure 2. Strengths and Weaknesses Found in a 4th Grader’s Essay
What Student Writing Can Teach Us - table 2
Meets or Exceeds These Benchmarks | Does Not Meet These Benchmarks | Suggested Next Instructional Strategies |
|---|---|---|
| Supporting details are relevant. The writer connects ideas to the specified purpose. The writer selects words that are accurate and specific. The writer uses a variety of sentence structures. The writing is nearly error-free. | The writer seems in control and develops the topic in an organized way. | This student may benefit from creating a bulleted list or a graphic organizer prior to writing. The teacher can ask the student to read through the piece aloud to clarify any confusing details. |
Sample chart annotating a piece of writing’s strengths and weaknesses in terms of benchmarks on a standardized rubric. Benchmarks are drawn from the Colorado Department of Education’s rubric for student writing.
Overmeyer, M. (2007/2008). What Student Writing Can Teach Us. Educational Leadership, 65(4) [Online].
Figure 2 shows how we pinpointed strengths and weaknesses for the 4th grader's piece on her room. It indicates which benchmarks from the rubric this writer meets (such as including relevant details), which benchmarks need more work (such as organization), and a few suggested next steps (such as creating a bulleted list). Comments on a chart like this can be directly aligned to the rubric, as shown here, or be based on teacher discussion. Either way, the goal is to read more deeply and go for teaching points.
We explained to administrators and curriculum leaders in language arts departments throughout the district our thinking about how student writing samples can be used to inform instruction and made this annotated format accessible on our district Web site. Teachers at all grade levels have used the format to work with student writing.
Like a well-written poem, each piece of student writing can teach us how to read it—and how to read the author's needs. Every piece of writing we collect from students can become part of formative assessment.
