How is the AASA writing prompt scored in Arizona?

What does the AASA writing task actually ask students to do?

Arizona’s Academic Standards Assessment, or AASA, is the statewide ELA and math test for Grades 3 through 8, delivered through Pearson. ELA runs three units at every grade: Writing plus two Reading and Language units, with an Oral Reading Fluency unit at Grade 3. Writing is one prompt, an extended essay, worth 10 points of a 52 to 56 point form, in 60 to 90 minutes.

The prompt is source-based. Each one gives students at least two or three stimuli, usually research articles with charts and graphs, and asks them to synthesize across the passages and cite details. Students answer no comprehension questions about those passages. The stimulus reading level is held to the easy Lexile band for the grade, so weak readers are not blocked from showing what they can write.

Genre is the part most departments get wrong. Each student receives one prompt: informative or explanatory at every grade, opinion in Grades 3 through 5, argumentative in Grades 6 through 8. Both types are administered operationally at every grade level every year, and nobody can predict which a student will see. A released Grade 8 sample asks for a multiparagraph argumentative essay on whether the Nebra Sky Disk is Bronze Age or Iron Age, with requirements including counterclaims and not overly relying on one source.

• Writing is 13 to 19 percent of ELA points at Grades 3-5 and 17 to 19 percent at Grades 6-8.
• DOK 4 accounts for 16 to 19 percent of the blueprint, all of it in Writing.
• AASA stops at Grade 8. Grade 9 takes ACT Aspire, Grade 11 takes the ACT, and Writing is required there because it is used for accountability.

Which three domains does the AASA writing rubric score?

The rubric has three domains and a maximum of 10 points. Purpose, Focus, and Organization scores 1 to 4. Evidence and Elaboration scores 1 to 4. Conventions scores 0 to 2. Four versions exist by grade band and genre. Scores come back per domain, which is why the state’s annotated samples are labeled 4/4/2 or 2/1/1, not a single number out of ten.

Purpose, Focus, and Organization is not just structure. In the Grades 6-8 Argumentative version it covers focus for the purpose, audience and task, the claim, whether it is maintained without loose material, transitions, progression of ideas, introduction and conclusion, and style and tone. Evidence and Elaboration carries nine key elements, among them use of sources, evidence integration, elaborative techniques, vocabulary and sentence structure.

Conventions is the domain teachers most often overweight. It is holistic and worth 2 points, not 4. The state frames it as one question: do the errors obscure the meaning? Arizona is also blunt about the instrument: the rubric is not a checklist, it requires some subjectivity, and a paper receives the score it most represents.

One naming quirk is worth knowing. The technical report labels the same traits Statement of Purpose, Focus and Organization, Evidence and Elaboration, and Conventions and Editing. The teacher-facing rubric guides use the shorter names. Both are official; use the rubric-guide names with students.

Why is Grade 6 not scored on counterclaims?

The Grades 6-8 Argumentative rubric is one shared document, but an element inside it is switched off for one grade. A footnote against the element about clearly addressing alternate or opposing claims says it is not applicable to Grade 6. Grade 6 writers are not scored on counterclaims. Grade 7 and Grade 8 writers are.

The state ships student materials that match the split: a Grade 6 Argumentative guide and a separate Grades 7-8 guide. Hand one scoring sheet to the whole band and Grade 6 papers get judged against something the state does not count. Teach counterclaim in Grade 6 because it makes better arguments. Do not score it as though the state does.

Who actually scores the essays now?

Arizona’s own technical report is direct about this. Pearson’s automated scoring engine, the Intelligent Essay Assessor, is the default option for scoring the AASA ELA writing prompts. As of the 2025 administration, all ELA prompts were scored at least in part by that engine, and 10 percent of responses got a second reliability score from human scorers. Where both exist, the machine score is reported if it is high confidence, and the human score stands if it is not.

Two mechanisms sit behind that. Continuous Flow means the engine keeps learning from human scorers in real time during the live scoring window. Smart Routing means responses the engine finds difficult are flagged low confidence and sent to people. Grade 3 spoken work is handled the same way, by Versant, Pearson’s engine for speech.

Arizona also publishes the numeric bar the engine must clear before it scores live, which most states do not. Machine-to-human correlation must reach 0.70 and sit within 0.1 of the human-to-human figure, and quadratic weighted kappa must clear the same test. The primary criterion is exact agreement of at least 65 percent, also within 5.25 percent of human-to-human agreement.

None of this appears in teacher-facing material; it lives in a technical appendix. The rubric is what the scoring is calibrated against.

How should this change the way you grade writing during the year?

Report classroom essays as three numbers, not one. If a student sees 3/2/1 in October and knows the middle number is evidence and elaboration, the May score report is legible to them. Cap conventions at two points in your own grading, and ask the state’s question rather than counting errors: does this obscure meaning? Teachers who flag every comma are weighting the smallest domain like the largest.

Build classroom tasks the way the test builds them: two or three sources, no comprehension questions attached, and an instruction to cite across them. Grade the use of sources and call out over-reliance on one, because released prompts name it. Students should also know the state puts a Writing Guide beside the prompt during the test, reminding them of key rubric pieces as they plan and revise. Practice with that guide, the annotated sample papers and any study guides that match the genre, and alternate the genre through the year, since both types run at every grade every year.

This is where a drafting tool earns its place, if you use one. Set the rubric up as the three AASA domains with the state’s own key elements rather than a generic six-trait grid, and have Jeddle draft against that, so the language students read in October matches what the scoring process models in April. JeddAI proposes domain scores and comments, and you edit before anything reaches a student. A moderation cycle can use expert essay grading as a second read on a sample, checking that two teachers reading the same Grade 7 paper land on the same three numbers.

What has changed in Arizona assessment recently?

The test’s name has moved twice and the content has not. The State Board adopted AzMERIT in November 2014, renamed it AzM2 in 2019-2020, and renamed that AASA in 2021-2022, aligned to the 2016 academic content standards and carrying the same cut scores. Arizona adopted its own English Language Arts standards in December 2016, and the current rubric guides date from October 2022.

Two real changes landed recently. Grade 3 Oral Reading Fluency, field tested in 2022, 2023 and 2024 without counting, became operational in spring 2025 and now contributes to student scores. And beginning with 2025-2026, identified Essential Standards in Grades 3 through 8 carry a higher proportion of items while staying within the current blueprint, with test length unchanged and every standard still assessable.

On paper testing, the rules did not loosen. Statute allows written form where an IEP or 504 requires it, for special circumstances, for religious purposes, or on request by a parent or guardian. HB2033 in 2026 would have added a majority vote of the governing board as a fifth trigger, and Governor Katie Hobbs vetoed it on April 14, 2026. The same statute says a test for penmanship shall not be required. One thing to watch: the AASA and AzSCI elementary contract ends June 2027.