← GRE Prep

GRE Analytical Writing rubric: every score level explained

The GRE Analytical Writing rubric is ETS's six-level scoring guide for the one Analyze an Issue essay. Raters score the whole essay from 6 (outstanding) to 1 (fundamentally deficient), or 0 if it is off topic, judging your position, reasons and examples, organization, language and grammar together. Scores are reported from 0 to 6 in half points.

By Test Ninjas · Published November 14, 2026

Score your own essay

Write a practice Issue essay in 30 minutes, then rate it honestly on each dimension below. Pick the description that fits your essay as written, not as you meant it. You get an estimated band and the specific gaps between you and the next level.

1. Position and thesis
2. Reasons and examples
3. Focus and organization
4. Language and sentence variety
5. Grammar and mechanics

Rate all five dimensions to see your estimate (0 of 5 done).

A study estimate, not an ETS score. The estimate weights development and position most, rounds to the nearest half point, and lets one weak dimension cap the result. Last verified against the ETS Analytical Writing scoring guide: November 14, 2026.

Practice the rest of the GRE Free Verbal and Quant practice with explanations.
Create a free account

What does the GRE writing rubric measure?

The rubric measures how well you build and communicate an argument, not what you know about the topic. ETS describes the measure as testing whether you can articulate and support complex ideas, construct arguments and sustain a focused, coherent discussion, and it states that no specific content knowledge is assessed. Since September 22, 2023, the section has one task: Analyze an Issue, with 30 minutes. The prompt gives an opinion on an issue plus specific instructions on how to respond, and you take and defend a position.

Scoring is holistic. A rater reads the essay once and assigns the level whose description fits it best overall; there is no checklist where each trait earns a fixed number of points. Still, the Issue guide describes every level through the same five threads, which is why the tool above asks about them separately:

DimensionWhat raters look forWhere it shows up
PositionA clear stance that follows the exact task instructions; at the top, an insightful oneFirst paragraph and how consistently the essay holds that line
Reasons and examplesSupport that is relevant (4), well chosen (5), or compelling and persuasive (6)Body paragraphs: whether each example is explained, not just named
Focus and organizationIdeas that connect logically and stay on the thesisTopic sentences, transitions, the order of points
LanguageClear, then precise and fluent, with sentence varietyWord choice, sentence structure
ConventionsGrammar, usage and mechanics that do not get in the wayErrors that start to affect meaning

Two details from the ETS directions matter more than students expect. First, the instructions line is part of the task: an essay that ignores it is scored as "vague or limited" in addressing the task even if the writing is fluent. Second, you type in a basic ETS word processor with cut, paste and undo, but no spell checker or grammar checker.

GRE Issue essay rubric: each score level explained

Each level is a description of a typical essay, and the bottom half works differently from the top half. Levels 6, 5 and 4 describe essays that are strong across the board to different degrees. Levels 3, 2 and 1 are triggered by "one or more" flaws, so a single serious weakness is enough to pull an otherwise decent essay down. Here is each level in plain language; read the official ETS scoring guide for the exact wording.

6
Outstanding

The position is clear and insightful, and it answers the exact instructions. Every main point is fully developed with compelling reasons or persuasive examples. The analysis stays focused and the ideas connect logically from paragraph to paragraph. The writing is fluent and precise, with effective word choice and varied sentences. Small errors can appear, but the essay shows superior control of grammar and usage.

5
Strong

A clear, well-considered position, developed with logically sound reasons or well-chosen examples. The essay is focused and generally well organized, and ideas connect appropriately. Language is clear, with suitable vocabulary and some sentence variety. Minor errors do not get in the way. The difference from a 6 is depth: the reasoning is thoughtful rather than insightful, and examples are well chosen rather than highly persuasive.

4
Adequate

A clear position supported by relevant reasons or examples, adequately focused and organized. The language is clear enough to follow, and grammar is generally under control, though there may be some errors. This is a competent essay: it does the job without much depth. Typical gaps are generic examples, little explanation of why an example proves the point, and no engagement with the other side.

3
Limited

Some competence, but obviously flawed in at least one way: the position or the response to the instructions is vague, the support relies largely on unsupported claims, the focus or organization is limited, sentences are unclear, or errors are frequent enough to interfere with meaning. One of these flaws is enough to place an essay here.

2
Seriously flawed

The response largely ignores the specific instructions or shows serious weaknesses: an unclear position, few or no relevant reasons or examples, poor focus or organization, or language and grammar problems that often obscure what the writer means.

1
Fundamentally deficient

Little or no evidence of understanding the issue or of building an organized response. It may be disorganized or extremely brief, or have errors so pervasive that the writing becomes incoherent.

0
Off topic

No attempt to address the assigned topic: the response is off topic, merely copies the prompt, is in a foreign language, or consists only of keystrokes. A blank response is reported as NS (no score).

Where do half-point scores fit?

ETS's score-level descriptions group each half point with the whole number above it: 6 and 5.5 share one description, 5 and 4.5 another, and so on down to 1 and 0.5. So a 4.5 reads as a low "strong" essay, not a high "adequate" one, and a 3.5 sits with the 4s. In practice a half point usually means two readers placed your essay in adjacent levels.

How are the human and e-rater scores combined?

A trained human rater scores your essay, and ETS's e-rater engine checks that score. ETS's score pages say Analytical Writing responses are evaluated on a 6-point holistic scale that includes a score from e-rater, a program that identifies essay features related to writing proficiency, and that one score is reported for the measure.

ETS research describes how the two fit together in what it calls a check-score model, which it states is used for GRE Analytical Writing:

  1. One trained human rater scores the essay on the 1 to 6 scale.
  2. e-rater produces its own score for the same essay.
  3. If the two do not disagree, the human score stands.
  4. If they disagree, a second human reads the essay and the two human scores are averaged. If those humans differ by more than a point, another human score is added.
What this means for you: the final score always rests on human judgment of the whole essay, so you cannot write "for the machine". Padding with long words or extra length does not move a human rater, and e-rater mostly acts as a trigger for a second read. ETS does not publish the exact disagreement threshold it uses today, so treat any precise cutoff you read elsewhere with caution. ETS also says every essay is reviewed with its similarity-detection software and by content experts; the Information Bulletin explains when that can lead to cancelled scores.

Sample prompt: what a 6, a 4 and a 2 paragraph look like

The rubric is easier to apply once you see the same idea written at different levels. The prompt below is an original Test Ninjas practice prompt in the style of the Issue task, not one from the ETS pool. Each excerpt is a single body paragraph, written by us, with notes tied to the rubric dimensions.

The most valuable lessons are usually learned from failure rather than from success.

Discuss how far you agree with the statement. As you develop your position, describe at least one situation in which the statement would not hold and explain how that affects your view.

Level 6 paragraph (Outstanding)

Failure teaches most reliably when its cost is survivable and its cause can be traced. Commercial aviation offers a useful case: because crashes are rare and catastrophic, the industry built systems for reporting near misses, small errors that cost nothing but expose the same weaknesses a disaster would. The lesson is not that failure instructs by itself, but that a failure becomes instructive only when someone examines it before the stakes rise. Success, by contrast, rarely invites that scrutiny; a team that wins tends to credit its strategy even when luck did the work. The statement is right about where the richest information lies, yet it misses a condition. A failure no one is allowed to analyze teaches nothing, and a failure too large to recover from leaves no one to learn.

  • Position: agrees, but states the conditions under which the claim holds, which answers the instruction to consider when it would not.
  • Development: one specific example, explained by its mechanism (near-miss reporting) rather than just named.
  • Counterpoint: explains why success hides lessons, so the paragraph argues rather than asserts.
  • Language: varied sentence lengths, precise phrasing, and a closing sentence that returns to the thesis.
Level 4 paragraph (Adequate)

I agree that people learn more from failure than from success. For example, when I failed my first driving test I learned that I needed to check my mirrors more carefully, and I passed the second time. Failure makes people think about what went wrong, while success makes them feel comfortable. Also, many famous inventors failed many times before they succeeded, which shows that failure is important for learning. Therefore failure teaches more valuable lessons than success.

  • Position: clear, but unqualified, and it skips the instruction to describe a case where the statement would not hold.
  • Development: the driving test example is relevant but small; the inventors example is generic and unexplained.
  • Organization: adequate, held together by plain transitions (for example, also, therefore).
  • To reach a 5: develop one example in depth, explain why failure produced the lesson, and add a limit to the claim.
Level 2 paragraph (Seriously flawed)

Failure is very important in life and everybody fail sometimes. success is also good because it make people happy and when people are happy they work harder, so it depends. There are alot of things you can learn like in school or sports or jobs. In conclusion failure and success both have lessons and people should learn from them.

  • Position: "it depends" with no explanation of what it depends on, so there is no position to evaluate.
  • Development: lists areas of life instead of giving any reason or example.
  • Instructions: ignored entirely.
  • Mechanics: agreement and capitalization errors are present, but the main reason for the 2 is the missing analysis, not the grammar.

Notice that the 6 is not longer or fancier than the 4 in any way that matters. It does three things the 4 does not: it limits its own claim, it explains how its example works, and it follows the instructions. For whole essays with paragraph-by-paragraph feedback, see our GRE essay examples.

Common reasons GRE essays drop a level

Most essays lose a level because of thin development or a skipped instruction, not grammar. These are the patterns that most often separate adjacent levels in the guide:

From → toWhat usually causes the dropFix
6 → 5Examples are well chosen but not explored in depth; the position is sound but not insightfulQualify the thesis and explain why the limit sits where it does
5 → 4Generic examples ("many studies", "throughout history") and no response to the other sideUse one specific case per paragraph and explain its mechanism
4 → 3The instructions line is ignored or the support is mostly assertionAnswer every part of the instructions; follow each claim with a reason
Any → 3One dimension is seriously weak, such as frequent errors that blur meaningFind your weakest dimension with the tool above and fix that first
3 → 2No clear position, or a list of topics instead of reasonsCommit to one side in the first paragraph and give it two reasons
Any → 0Writing a memorized essay on a different topic, or copying the promptPlan around the exact prompt you are given

If timing is what keeps your essay short or unfinished, the planning and pacing routine in our score 5 strategy guide covers it; this page stays with what the rubric rewards.

What do GRE writing score levels mean for percentiles?

Writing percentiles bunch up because most test takers land between 3.0 and 4.5. In ETS's 2026 interpretive data, based on everyone who tested from July 1, 2022 to June 30, 2025 (785,720 test takers for this measure), the mean Analytical Writing score was 3.46. A 4.0 was higher than 63 percent of test takers, a 4.5 higher than 85 percent, a 5.0 higher than 93 percent, and a 3.0 higher than 16 percent. Half a point in the middle of the scale moves your percentile more than anywhere else.

See the full ETS percentile table (Table 1C, PDF) or our AWA percentiles guide for every score and what programs tend to expect.

GRE writing rubric FAQ

It is the scoring guide ETS publishes for the Analyze an Issue task. It describes six score levels, from 6 (outstanding) to 1 (fundamentally deficient), plus 0 for responses that do not attempt the task. Raters judge the essay holistically, by its overall quality, rather than adding up points for separate traits.

Both. ETS says essays are scored on a 6-point holistic scale that includes a score from its e-rater scoring engine. ETS research describes a check-score model: a trained human rater scores the essay, e-rater checks that score, and a second human reads the essay when the two disagree.

ETS reports Analytical Writing on a 0 to 6 scale in half-point increments. When two human ratings are averaged, for example a 4 and a 5, the result is 4.5. ETS describes a 5.5 with the same language as a 6, a 4.5 like a 5, and so on.

Not for the current test. Since September 22, 2023, the GRE General Test has one Analytical Writing task, Analyze an Issue, with 30 minutes. ETS still posts the Argument scoring guide, labeled for tests given before that date.

The rubric sets no word count. Length matters only indirectly: a level 1 response can be described as extremely brief, and a short essay rarely has room for the developed reasons and examples the upper levels describe. Development, not length itself, is what raters reward.

Only when they start to affect meaning. The guide allows minor errors even in a 6. At level 3 errors can interfere with meaning, and at level 2 they frequently obscure it. The test word processor has no spell checker or grammar checker, so proofreading is up to you.

Most applicants should read it against the percentile table ETS publishes. In the 2026 interpretive data (test takers from July 1, 2022 to June 30, 2025), the mean was 3.46, a 4.0 was higher than 63 percent of test takers, and a 5.0 was higher than 93 percent. Program expectations vary.

No. It is a study aid that organizes your own judgment of a practice essay around the rubric's dimensions. Only ETS raters and e-rater produce an official score. For the most honest result, rate an essay you wrote under a 30-minute limit and ask someone else to rate it too.

Official sources

All ETS facts on this page were checked on November 14, 2026. Rubric descriptions are our paraphrases; the sample prompt and paragraphs are original Test Ninjas material.

Prepare for every GRE section

Practice Verbal and Quant questions with explanations and track your progress.