Smarter Balanced Writing Rubrics, Explained with Examples

A seventh grader gets her essay back from a practice performance task. You've scored it with the Smarter Balanced rubric: a 3 for organization, a 3 for evidence, and a 1 for conventions. She looks at it and asks, "So is a 3 a B? And why is conventions only out of 2?"

If you teach in a Smarter Balanced state, the writing rubrics shape how your students' essays are judged at the end of the year. They're also useful all year round, if you know how they work. This guide explains them in plain language, trait by trait, with examples.

The short answer: Smarter Balanced scores each full essay on three traits. Organization/Purpose and Evidence/Elaboration are each scored from 1 to 4. Conventions is scored from 0 to 2. The first two scores are averaged and rounded up, so each essay ends up with two scores: one out of 4 and one out of 2. A 3 means the writing does the job: the rubric calls a 3 "evident" organization and "adequate" elaboration. A 4 is "effective" and "thorough". A 3 isn't a B. It's solid, grade-level writing.

What is Smarter Balanced?

Smarter Balanced is a group of states and territories that share one set of standards-based assessments in English language arts and math, built around the Common Core. Students take the tests in grades 3 to 8 and 11.

According to the Smarter Balanced partnerships page, current members are California, Connecticut, Delaware, the District of Columbia, Hawaii, Idaho, Michigan, Nevada, Oregon, South Dakota, Washington, the U.S. Virgin Islands, and schools run by the Department of Defense Education Activity. Washington, D.C. joined in October 2025. Montana and Vermont have moved to other tests.

States can use their own names for the test. In California it's part of CAASPP. Our CAASPP guide covers the California side in detail.

Where writing shows up on the test

Students write in two places.

1. The full write (the performance task). Students read a few sources on one topic, answer some research questions, and then write a full essay or story. This is the piece scored with the three-trait rubric, by trained human scorers.

2. Brief writes. In the main part of the test, older students may get a short writing item. These ask students to read a short text of one or two paragraphs and write a paragraph, for example a conclusion. Brief writes are scored 0 to 2. Not every state includes them, so check your state's test blueprint.

Which kind of full write a student gets depends on their grade:

Grade 11 students never get a narrative full write (Smarter Balanced scoring guide). Some states use fewer purposes than this table shows. Oregon's current blueprint, for example, lists only opinion/argumentative or informational/explanatory writing.

The three traits, explained

Every full write is scored on the same three traits. The wording changes a little by purpose, but the structure doesn't.

Trait 1: Organization/Purpose (1-4)

This trait asks: does the writing hold together, and does it do what the task asked? Scorers look at the introduction and conclusion, whether the focus stays on the claim or main idea, how ideas connect, and whether the writing moves logically from one part to the next.

Here's how the argumentative rubric opens each score point:

Trait 2: Evidence/Elaboration (1-4)

This trait asks: how well does the student back up their ideas? Scorers look at how sources are used and cited, how evidence is explained, and how precise the language is. In narrative writing, this trait is called Development/Elaboration, and it's about details, dialogue and description instead of sources.

The key words to notice are thorough, adequate, uneven and minimal. Elaboration isn't about length. A long essay that repeats itself can still be "cursory".

Trait 3: Conventions (0-2)

Conventions covers grammar, usage, punctuation, capitalization and spelling. It has only three score points:

• 2: "an adequate command of conventions"
• 1: "a partial command of conventions"
• 0: "little or no command of conventions"

Scorers don't count errors. The rubric tells them to weigh three things:

• Variety: the range of error types.
• Severity: "Basic errors are more heavily weighted than higher-level errors."
• Density: "The proportion of errors to the amount of writing done well."

So a long, ambitious essay with a few slips can still earn a 2. A short one full of basic errors may not. That's worth remembering in class, too. (We cover the classroom side in grammar feedback without grammar penalties.)

How the scores combine

The official scoring rules say it plainly: "Scores for the first two dimensions are averaged, and the average is rounded up." So:

Conventions stays separate. As the technical report puts it, essays are "scored on three traits, two of which are combined, resulting in two scores for each essay."

When an essay gets no trait scores

Some responses get a condition code instead of scores. The main ones, from Washington's guide:

• Blank: no response.
• Insufficient: not enough to score, including copied source text.
• Non-scorable language: written in a language other than English.
• Off topic: unrelated to the task or the sources.
• Off purpose: "clearly not written to the purpose designated in the task."

Off purpose is the one that surprises teachers. If the task asks for an argument and a student writes a story, only conventions gets scored. Organization and evidence don't. Teach students to check the purpose before they start writing.

Examples: one paragraph at three score points

Rubric words like "adequate" and "cursory" make more sense when you see them. Here's one body paragraph from a made-up grade 7 argument task, written at three levels. The task asks: Should our school start later in the morning? Students have read three sources.

These are illustrations written for this guide, not official anchor papers. Real scorers judge the whole essay, not one paragraph. But the difference between the levels shows up clearly at paragraph level.

Score 2: "uneven, cursory elaboration"

Schools should start later because students are tired. Source 2 says teenagers need more sleep. Also many students are late in the morning. Starting later would help with this.

The claim is there, and the student mentions a source. But the evidence is just dropped in. There's no explanation of why tired students are a problem or how a later start would fix it. "Many students are late" isn't tied to any source.

Score 3: "adequate elaboration"

Schools should start later because teenagers aren't getting enough sleep. According to Source 2, teenagers' body clocks shift during puberty, so they fall asleep later at night. That means a 7:30 start forces many students to come to school tired. When students are tired, it's harder to pay attention in class.

Now the evidence is cited and explained. The student connects the source to the claim. This does the job. What it doesn't do yet is go further: it doesn't use a second source, show the effect on learning, or deal with the other side.

Score 4: "thorough and convincing elaboration"

Schools should start later because our current schedule works against how teenagers' bodies work. According to Source 2, body clocks shift during puberty, so most teens can't fall asleep until late at night. A 7:30 start means they're in class before their brains are fully awake. Source 3 backs this up: after one district moved its start time later, it reported fewer students arriving late. Some parents argue that a later start would cut into after-school sports (Source 1). But a few minutes of practice time is a small cost if students are actually awake for their classes.

Two sources work together, each one is explained, and the counterargument is named and answered. One detail worth knowing: the counterargument is actually scored under Organization/Purpose. From grade 7, a 4 on that trait expects "alternate and opposing argument(s)" to be "clearly acknowledged or addressed". The language is also more precise ("works against", "small cost").

What moved the score? Not length on its own. The 4 explains how each piece of evidence supports the claim, links sources together, and deals with the other side.

How the rubric changes by purpose

Organization and Conventions read almost the same across every purpose. The Evidence/Elaboration row is where the purposes differ:

The official rubrics for all five purposes are linked from the Smarter Balanced samples page.

Using the rubrics in class (without teaching to the test)

The best reason to use these rubrics isn't the test. It's that a clear rubric, shared with students, helps them write better.

What the research says

• Rubrics help students do better work. A 2023 meta-analysis led by Ernesto Panadero found rubrics had a moderate positive effect on student performance, with an effect size of 0.45.
• Students who check their own writing against a rubric write better. In a study of 116 third and fourth graders, Heidi Andrade and colleagues found that "using a rubric for self-assessment can help elementary school students produce more effective writing" (Educational Measurement: Issues and Practice, 2008). A follow-up with 162 middle schoolers found the same effect on total essay scores and on every criterion (Assessment in Education, 2010).
• Feedback on writing works, and teacher feedback works best. A 2015 meta-analysis by Steve Graham, Michael Hebert and Karen Harris found effect sizes of 0.87 for feedback from adults, 0.62 for self-assessment, 0.58 for peer feedback and 0.38 for computer feedback on students' writing in grades 1 to 8 (Elementary School Journal).

The takeaway: use the rubric as a teaching tool all year, and pair it with real feedback. Don't save it for test prep in April.

Five ways to use the rubrics all year

1. Make a student version. Rewrite each score point as "I can" statements. "I explain how each piece of evidence supports my claim" is easier to act on than "thorough and convincing elaboration".
2. Score examples together. Show the class an anonymous essay and have them score it on one trait before you reveal your score. The Smarter Annotated Response Tool (SmART) has real student responses with scores and explanations, organized by purpose and grade band.
3. Teach one trait at a time. Spend a week on elaboration: finding evidence, explaining it, linking two sources. Then move to organization. Students can't fix everything at once.
4. Have students self-score before they hand in. Ask them to score their own draft on the rubric and underline the sentence that justifies each score. This is the self-assessment routine Andrade's studies tested.
5. Use the practice tests. Every member state links free practice and training tests, with scoring guides for the performance tasks. Smarter Balanced's Tools for Teachers has lessons built around the same skills.

A caution: don't let the Smarter Balanced rubric become the only rubric students ever see. It's designed for one kind of task, a timed essay from sources. Students also need to write research papers, lab reports, poems and personal essays, and those deserve rubrics built for them. Our guide to high school argumentative writing rubrics covers classroom rubric design.

Five scoring mistakes to avoid

1. Treating a 3 as a B. A 3 is solid, grade-level work. Converting rubric scores straight into percentages (3 out of 4 = 75%) makes good writing look average.
2. Rewarding length. A long essay that repeats itself is still "cursory". Score what the evidence does, not how much there is.
3. Counting errors for conventions. The rubric weighs variety, severity and density. A few small slips in a long, ambitious essay can still earn a 2.
4. Letting conventions drag down the other traits. A paragraph full of spelling errors can still have strong organization and evidence. Score each trait on its own.
5. Forgetting the purpose. An essay written to the wrong purpose loses its organization and evidence scores entirely. Make "check the purpose" a habit before students start writing.

Using Smarter Balanced-style rubrics in GradingPal

If you grade in GradingPal, you can grade practice essays against the same three traits all year. Here's how it works:

• Set up the rubric once. Start from the rubric library, or build the three traits as your own rubric: Organization/Purpose and Evidence/Elaboration at 1 to 4, and Conventions at 0 to 2. Using a library rubric makes your own copy, so you can reword criteria, change levels or adjust points without touching the original.
• Make a student version. Ask GradingPal can rewrite level descriptions in student-friendly language. It shows each proposed change, and you approve or reject it.
• Tell it what you look for. Add grading instructions in plain words, like "From grade 7, count acknowledging the opposing view under Organization/Purpose." Save the ones you reuse.
• Check every suggested score against the work. For each trait, the grading view shows a suggested level, the reasoning, and the quoted evidence from the student's essay. If you disagree, tap a different level.
• Keep conventions in their own row. Turn on the grammar check and spelling, punctuation and grammar slips are underlined in the student's work. Grammar findings never change a score. Conventions only counts through the Conventions row of your rubric, the same way it works on the test.
• Paper is fine. Timed practice essays written by hand can be scanned as one PDF for the whole class. GradingPal splits it into one submission per student. See handwritten and scanned work.
• Find the whole-class lesson. Assignment analytics name common strengths and weaknesses with counts, like "Weak explanation of evidence: 9 of 24 students". That's your next elaboration mini-lesson.

Two things to keep in mind. GradingPal isn't affiliated with or endorsed by Smarter Balanced, and a practice score isn't a prediction of a state test score. Also, your total in GradingPal follows your rubric's points. It won't apply the test's rule of averaging the first two traits, so look at the trait scores, which are the useful part anyway.

You can try it on the free plan, which includes 50 submissions a month. Rubric library access depends on your plan.

The bottom line

The Smarter Balanced writing rubrics come down to three questions. Is the writing organized for its purpose? Is the evidence explained, not just dropped in? Does the student control the conventions well enough for the reader to follow?

Teach those three questions all year, one at a time. Share the rubric in student-friendly words, score examples together, and have students score their own drafts. Remember that a 3 is solid work, that length isn't elaboration, and that writing to the right purpose matters.

Want to grade practice essays against the same traits? Start free with GradingPal and set up a Smarter Balanced-style rubric for your next writing task.