TELPAS Writing Rubric: The 12-Point and 4-Point Rubrics, Writing Samples and Prompts

Last updated: September 28, 2026

Estimated reading time: 12 minutes

The TELPAS writing rubric depends on the grade. In grades 4-12, TEA scores constructed responses on a twelve-point rubric with three traits, vocabulary, usage and completeness, each scored from 1 to 4. In grades 2-3, constructed responses are scored on a four-point rubric, and at every grade from 2 to 12, sentence-rewrite questions are scored 0 or 1. Writing is part of the online reading and writing test, and TEA publishes a writing scoring guide for every grade cluster, with real prompts and scored student samples. This guide covers the rubrics, the samples and prompts, how responses are scored, and when a writing collection is still required.

Executive Summary

Since spring 2023, TELPAS reading and writing in grades 2-12 is one combined online assessment, taken in the spring window. Writing is assessed through two question types: sentence rewrites, in which students correct the grammar and syntax of a short text, and constructed responses, in which students draft their own writing from a picture or a prompt. For grades 4-12, an automated scoring engine scores every response first and at least 25 percent go to trained human scorers; in grades 2-3, writing is 100 percent human scored.

Writing collections, a set of classroom writing samples rated holistically, now apply only to grades 2-12 students who qualify for a special holistic administration. The most useful free resources for everyone else are TEA’s writing scoring guides, which show what each score point looks like on real prompts.

Key Takeaways

• Grades 4-12: a twelve-point rubric, vocabulary, usage and completeness, each scored 1 to 4.
• Grades 2-3: a four-point rubric for constructed responses.
• Sentence rewrites: 0 or 1 point, grades 2-12.
• Scoring: automated engine plus human scorers in grades 4-12; human only in grades 2-3.
• Writing samples: TEA’s 2025 and 2023 scoring guides, by grade cluster, with annotated student responses.
• Writing collections: only for special holistic administrations, with at least five samples dated from February.

The Twelve-Point Writing Rubric, Grades 4-12

Quick answer: TEA’s twelve-point writing rubric for grades 4 through 12 scores three traits, vocabulary, usage and completeness, on a one-to-four scale each, for a maximum of twelve points.

The three-trait structure tells a teacher what to teach. The rubric is not scoring whether the student had a good idea. It is scoring whether the English carried it.

The descriptors anchor each level in the effect on comprehensibility rather than in error counts. A scorer is not counting mistakes; they are asking how hard the reader has to work. Two students with the same number of errors can sit at different levels if one set of errors obstructs meaning and the other does not.

The Four-Point Rubric for Grades 2-3

In grades 2 and 3, constructed responses are scored on a single four-point rubric that covers the same ground: vocabulary range, spelling, grammar and usage, sentence complexity and how clearly the message comes through. At score point 1, TEA describes high-frequency words with widespread spelling errors and significant grammar errors; at score point 4, infrequent spelling errors, correct use of simple and complex tenses, and grammar comparable to native-speaking peers.

Sentence rewrites use the same rubric at every grade: 1 point if “the response is one or more complete sentences and corrects all the errors in the given text,” 0 otherwise.

TELPAS Writing Samples and Prompts

Quick answer: TEA publishes a writing scoring guide for each grade cluster with its released reading and writing tests. Each guide reproduces the real prompts and shows scored student responses at every score point, with an annotation explaining each score.

For 2023, TEA also publishes sentence-rewrite scoring guides for elementary, middle and high school. All sets are listed on TEA’s TELPAS released test questions page, and the full tests are covered in our TELPAS practice test guide.

What the prompts look like, from the 2025 guides:

• Sentence rewrite, grade 2: “How I can find my classroom?” The student rewrites the text so that it is correct.
• Sentence rewrite, grades 4-5: “Most childrens is riding the bus.” TEA’s annotation for a full-credit answer notes that the writer replaces “childrens is” with “children are,” and that “the lack of punctuation does not impact the score.”
• Picture-based, grades 4-5: “Describe this picture. Then write three facts that you can learn from this picture,” followed by tasks to explain each stage in the growth of an orange tree and what students can do to help plants grow.
• Imaginative, grades 4-5: “Imagine that you invented a robot that could do small jobs inside and outside your home. Write about three jobs the robot could do.”
• Content-based, grade 3: a food chain picture, with tasks to tell what each animal eats and “explain what would happen if there were no grasshoppers.”

The prompts ask students to describe, explain, narrate and imagine, often with several parts. A response that answers only the first part rarely reaches the top of the completeness trait.

How TELPAS Writing Is Scored

TEA’s speaking and writing scoring process (March 2024) sets two models:

The writing items produce a raw score that TEA converts to a scale score and a writing rating, Beginning to Advanced High. How the four domain ratings form the composite is in TELPAS scores explained.

Why Completeness Is the Trait Districts Underestimate

Vocabulary and usage are what teachers expect to be scored. Completeness is where scores quietly go wrong. A student who writes three careful sentences with no errors has not demonstrated much. The rubric cannot credit range that was never attempted, so a cautious, accurate, very short response caps out low.

This produces a counterintuitive instructional message that is nonetheless correct: for TELPAS writing, a student attempting a longer, more detailed piece with some errors often demonstrates more than a student producing a short flawless one. TEA’s own descriptor for a 4 in completeness asks for “specificity and detail needed to address the task completely.” Encouraging Emergent Bilinguals to write at length, and to answer every part of a prompt, improves both the instruction and the score.

When a Writing Collection Is Still Required

A writing collection is no longer the way most Texas students are rated in writing. TEA’s paper and holistic administration information applies holistic writing ratings, based on a collection, to grades 2-12 students under a special administration. For those students, TEA sets these rules:

• At least five writing samples.
• At least one sample that elicits the past tense.
• At least two academic samples from mathematics, science or social studies.
• No samples dated before February of the current school year.
• No copied, heavily edited or teacher-corrected work, and no worksheets.

TEA also asks that the collection include tasks that let students show their capabilities and tasks that “stretch and push students’ limits.” The collection is kept in the student’s permanent record file for two years from the time of rating. The requirements are in TEA’s TELPAS writing collection document.

In kindergarten and grade 1, trained teachers rate writing holistically against the proficiency level descriptors, from classroom observation. For 2026-2027 and 2027-2028, raters use TEA’s eligible transition descriptors, covered in TELPAS proficiency level descriptors.

District Benchmark: One Scoring Guide Session per Campus

The scoring guides are the cheapest calibration tool a district has. Take a district with 40 campuses: one hour per campus, in which teachers of Emergent Bilinguals score five sample responses from the scoring guide for their grade cluster before reading TEA’s scores and annotations, costs 40 hours of meeting time. It gives every teacher a shared picture of what a 2 and a 3 look like on each trait, which is exactly what they need to give feedback on class writing all year.

The comparison worth making internally is with speaking. Writing practice scales with the class, so a district that organizes it can move the writing rating. Speaking practice does not scale that way, which is why the two domains diverge in most TELPAS reports.

Why Writing Outruns Speaking in Every District Report

Read a Texas district’s TELPAS data by domain and writing often sits above speaking. The cause is not that writing is easier to learn. It is the Speaking Time Gap: the shortfall between the responsive spoken practice an Emergent Bilingual needs and what one teacher can supply across a full class.

The math, run across the two productive domains. In a 45-minute block, all 25 students can write at once, so a single teacher can generate 25 pieces of practice simultaneously and collect 1,125 student-minutes of production. In that same block, spoken production happens one student at a time, so the ceiling is a couple of minutes each no matter how the lesson is run. Writing scales with the class; speaking divides by it.

That structural difference explains why writing responds to a district’s effort while speaking resists it. A district that decides to improve writing can do so inside the existing schedule. A district that decides to improve speaking has to find capacity the schedule does not contain.

Metrics Worth Tracking

• Extended writing tasks per Emergent Bilingual per week, in every subject.
• Share of responses that answer every part of a multi-part prompt, given the completeness trait.
• Agreement with TEA’s scores on scoring guide samples, by campus.
• Writing ratings by grade cluster, since grades 2-3 and 4-12 use different rubrics.
• The writing-to-speaking rating gap, tracked year over year as a capacity indicator.

Common Mistakes District Leaders Make

1. Building writing collections for every student. They apply only to special holistic administrations.
2. Using a September collection calendar for students who need one. Samples dated before February are not eligible.
3. Teaching one rubric for all grades. Grades 2-3 and 4-12 are scored differently.
4. Rewarding short and safe writing. The completeness trait cannot credit range that was not attempted.
5. Skipping the sentence rewrites. They are all-or-nothing: every error must be corrected.

Recommended Actions

Immediate (this month): Share the rubric and the scoring guide for each grade cluster with every teacher of Emergent Bilinguals, and run one scoring session per campus.

Medium-term (this year): Make multi-part, extended writing a weekly routine in content classes, and plan February collections for students in special administrations.

Long-term (strategy): Track the writing-to-speaking gap as a standing indicator of instructional capacity, and resource the speaking side, which the schedule cannot supply on its own.

Questions District Leaders Should Ask

• Do our teachers know which rubric applies to their grade?
• Which students need a writing collection this year, and do we start it in February?
• How closely do our campuses agree with TEA’s scores on the scoring guide samples?
• Are our Emergent Bilinguals writing at length, or safely and briefly?
• How wide is the gap between our writing and speaking ratings, and is it narrowing?

Frequently Asked Questions

What is the TELPAS writing rubric?

For grades 4-12, a twelve-point rubric that scores vocabulary, usage and completeness from 1 to 4 each. For grades 2-3, a four-point rubric. Sentence rewrites are scored 0 or 1 at every grade.

What three traits does the TELPAS writing rubric score?

Vocabulary, usage and completeness, each from 1 to 4, for a maximum of 12 points. The descriptors are anchored in how far errors interfere with comprehensibility rather than in how many errors appear.

Where can I find TELPAS writing samples?

In TEA’s writing scoring guides, published for grades 2, 3, 4-5, 6-7, 8-9 and 10-12 with the released reading and writing tests. Each shows the real prompts and scored student responses with annotations. The links are in the table above.

What are TELPAS writing prompts like?

Picture-based, imaginative and content-based tasks, often with several parts: describe a picture and write three facts, explain the stages of a process, imagine a robot and write about three jobs it could do. Sentence rewrites ask students to correct a short text such as “Most childrens is riding the bus.”

Is a TELPAS writing collection still required?

Only for grades 2-12 students rated holistically under a special administration. Those collections need at least five samples, including one that elicits the past tense and two from mathematics, science or social studies, none dated before February of the current school year.

Is TELPAS writing scored by a computer?

In grades 4-12, partly: an automated engine scores every response first and at least 25 percent go to trained human scorers. In grades 2-3, writing is 100 percent human scored.

Does length matter in TELPAS writing?

Yes, through the completeness trait. A short flawless response cannot demonstrate range, so it caps out lower than a longer, more detailed response that answers every part of the prompt.

Related Resources

• TELPAS Practice Test
• TELPAS Speaking
• TELPAS Scores Explained
• TELPAS Proficiency Level Descriptors
• TELPAS Testing Window 2027
• TELPAS Calibration

Conclusion

TELPAS writing is scored on the online test with published rubrics, and TEA shows what every score point looks like in its scoring guides. Districts that teach the right rubric for each grade, use the scoring guides to align feedback, and push students to answer every part of a prompt at length will see the writing rating move. The more informative number is what that leaves behind: the distance between writing and speaking.

Sources: TEA, TELPAS Twelve-Point Writing Rubric for Grades 4 through 12; TEA, TELPAS Released Test Questions; TEA, 2025 TELPAS Grades 4-5 Writing Scoring Guide; TEA, 2025 TELPAS Grade 3 Writing Scoring Guide; TEA, TELPAS Speaking and Writing Scoring Process; TEA, TELPAS Paper and Holistic Test Administration Information; TEA, TELPAS Writing Collection. Verified September 28, 2026.

The Domain That Answers, and the Domain That Does Not

A Texas district that takes writing seriously usually sees it. Teach the rubric, score the TEA samples together, push students to write at length, and the writing ratings climb. It is one of the few places in EB program management where effort and outcome line up cleanly.

Which makes the contrast painful. The same district applies the same seriousness to speaking and the ratings barely move, because writing can be practiced by 25 students at once and speech cannot. The gap is arithmetic, not commitment.

Some Texas districts add speaking practice that does not wait for the teacher’s turn. Telo AI is one example: five-minute sessions three times a week that adapt to each student’s level and push past one-word answers, with progress reported on a dashboard mapped to the Texas ELPS. See how that works across a school year.

A rubric describes what good work looks like. It cannot create the conditions in which a student produces it. Writing gets those conditions from a normal school day; speaking has to be given them deliberately.