The GRE Analytical Writing measure opens every test with the Analyse an Issue task, and the 30-minute clock starts the moment the prompt appears. ETS frames the task as a response to a single claim of policy or value, scored on a 0–6 scale by one human reader and the e-rater engine, with the two scores averaged when they agree. Most candidates preparing for graduate school admissions know the task exists, but fewer have a clear picture of what actually separates a 4.0 from a 5.0 once the timer is running. This article is the tutor-level read: the paragraph architecture, the line budgets, and the essay archetypes that the rubric quietly rewards.
What the Analyse an Issue task is actually asking
Every Issue prompt presents a one- or two-sentence claim, asks whether you agree or disagree, and invites you to explain your reasoning with examples. The two-claim prompts look more intimidating, but the structure is identical: pick a position, defend it, anticipate the counter-argument, and close. ETS does not grade whether you picked the 'right' side. A well-argued defence of a weaker position still beats a muddled agreement with the stronger one. This is the single fact most candidates miss in the first week of preparation.
ETS publishes a pool of official prompts and sample essays at each score band, and reading three or four 6.0 responses is the fastest way to recalibrate your mental model. Notice how the 6.0 essays commit to a position in the opening sentence rather than introducing a both-sides framing. Hedging language ('this is a complex issue', 'there are valid points on both sides') correlates with scores in the 3.0–3.5 range, because it signals that the writer has not decided what to argue.
The four-paragraph skeleton shared by most 4.0+ essays
Almost every high-scoring Issue response follows the same architectural shape, and practising the skeleton is more productive than improvising each essay. The structure is not a formula, it is a container that frees you to think about argument quality rather than paragraph count.
- Paragraph 1 (around 90 words): state your position, frame the reasoning, and outline the two reasons you will defend. End with a transition that names Reason 1.
- Paragraph 2 (around 120 words): develop Reason 1 with a worked example, ideally drawn from a domain you can explain in one sentence. A specific example outscores a general one every time.
- Paragraph 3 (around 120 words): develop Reason 2 with a different example, ideally from a different domain so the e-rater registers breadth of evidence.
- Paragraph 4 (around 70 words): acknowledge the strongest counter-argument, then explain why your position still holds. Close with a one-sentence synthesis.
The total comes to roughly 400 words, which leaves about 90 seconds per paragraph after you factor in planning, rereading the prompt, and proofreading. For most candidates I work with, the single highest-leverage change is enforcing this word count by paragraph rather than by total. A 500-word essay with a 30-word conclusion almost always reads as rushed at the close, which is where graders form their final impression.
Three essay archetypes the rubric rewards
Once the skeleton is internalised, the next decision is which archetype you are writing. Issue prompts cluster into a small number of shapes, and the archetype determines which two reasons you choose.
| Archetype | Prompt type | Strong Reason 1 | Strong Reason 2 |
|---|---|---|---|
| Policy defence | Should X be done? | Concrete outcome if implemented | Failure mode of the alternative |
| Value hierarchy | X is more important than Y | Case where the trade-off resolves cleanly | Cost of refusing to rank them |
| Causal claim | X causes Y | Mechanism, not just correlation | Boundary condition where the link breaks |
For a policy prompt about whether governments should fund space exploration, a 5.0 essay typically pairs a concrete spinoff argument (GPS, water filtration) with a counter on opportunity cost that the writer then qualifies. The reason both reasons score well is that each one is bounded: the writer specifies a domain, names an example, and limits the claim. Unbounded reasons ('it inspires people', 'it educates the public') read as filler because they could attach to almost any prompt. The e-rater is trained to flag exactly this kind of language.
Common pitfalls and how to avoid them
The same five errors appear in roughly two-thirds of the Issue essays I review, and each one drops the score by half a band before the grader has finished reading paragraph 1.
- The both-sides intro. 'There are merits to both sides of this argument' is a sentence that buys you nothing. Replace it with a direct claim: 'The argument fails because its central assumption ignores how X actually behaves in practice.'
- The personal anecdote. One sentence of lived experience is fine as a hook. A full paragraph about your summer internship reads as filler, because the e-rater cannot evaluate its generality.
- The list of three examples without development. Three thin examples score lower than one fully developed one. Pick two, and spend the third paragraph on the counter-argument instead.
- Quoting the prompt. The grader has just read it. Restating the claim in your own words shows you understood it; quoting it word for word suggests you are buying time.
- Ending without a synthesis. A conclusion that restates Reason 1 and Reason 2 wastes its 70-word budget. Use the space to name the condition under which the counter-argument would actually win. That sentence is the highest-signal line in the entire essay.
A useful self-check: read your conclusion aloud. If it could be transplanted into any other Issue essay on the same archetype, it is too generic. Force yourself to add the specific boundary condition that makes your position defensible only here.
How the e-rater and the human reader split the scoring
Each Issue response receives one human score and one e-rater score. When the two agree, the average is taken; when they diverge by more than one point, a second human reader is brought in, which is the only situation in which the e-rater essentially acts as a tiebreaker. The engine evaluates roughly 50 features including sentence variety, word choice, and topical relevance, but its biggest weight sits on argument structure. This is why the four-paragraph skeleton above matters more than vocabulary lists or template phrases: the e-rater is looking for predictable, defensible moves, and a recognisable architecture is the cheapest way to satisfy that check.
For most candidates, the practical consequence is that you should optimise the essay for a careful human reader first, then verify that your writing also contains the structural features the e-rater expects. Writing in long, varied sentences and avoiding bullet-style paragraph fragments are the two highest-leverage style choices; vocabulary games (thesaurus-swapping 'important' for 'crucial') register as noise rather than signal.
Building a two-week Issue prep plan
Two weeks is enough to lift a 3.0 to a 4.5 if you practise the right drills. Here is the schedule I usually assign.
- Days 1–3: read six official 6.0 sample essays and underline the first sentence of each body paragraph. You will see the same commitment pattern every time.
- Days 4–6: outline five new prompts without writing a full essay. Time each outline at seven minutes. The point is to internalise the two-reason selection.
- Days 7–9: write two timed essays (30 minutes each) and self-score against the ETS rubric. Identify the single feature that lost you the most points.
- Days 10–12: rewrite the weaker of the two essays with the skeleton enforced by paragraph, then compare.
- Days 13–14: one final timed essay, plus a 15-minute review of common transitional phrases that fit your archetype.
Two timed essays across two weeks is the right dose. More than that and you start rehearsing bad habits under fatigue; fewer and the time pressure stays unfamiliar. Most candidates I work with find that the outline drill, not the full-essay drill, is where their score actually moves, because it forces Reason 1 and Reason 2 to be chosen before the timer pressures them into filler.
What 'good writing' actually looks like on this task
ETS describes the 6.0 essay as 'insightful' and the 4.0 as 'competent'. In practice, the difference shows up in three concrete moves that a competent essay skips.
First, the 6.0 essay names the mechanism behind its claim. A 4.0 essay says 'student loan forgiveness would help the economy'; a 6.0 essay specifies the channel: 'by freeing the monthly cash flow that currently services debt, recipients shift roughly 8–12 per cent of disposable income into local consumption, which has a measurable multiplier in service-sector employment'. The candidate who can write that second sentence is the one who clears 5.5.
Second, the 6.0 essay uses the counter-argument as a stress test rather than a straw man. Acknowledging the strongest version of the opposing view, then explaining the specific condition under which it fails, is the move that separates a 5.0 from a 5.5 in most graders' eyes.
Third, the 6.0 essay varies its sentence length deliberately. Short sentences carry the claims; longer sentences carry the qualifications. Reading your draft aloud and counting words per sentence is a fast diagnostic: if every sentence is between 18 and 24 words, the e-rater is going to read you as formulaic, even if the argument is strong. Aim for at least two sentences under 12 words in the body, and at least one over 30.
Conclusion and next steps
The Analyse an Issue task rewards a recognisable architecture, a committed position, and two bounded reasons more than it rewards vocabulary or prose polish. Practising the four-paragraph skeleton, the three archetypes, and the counter-argument close is the highest-leverage use of the two weeks before test day. Candidates who build their prep around the outline drill rather than the full-essay drill typically gain more ground, because the outline is where the argumentative decisions actually live. TestPrep İstanbul's diagnostic assessment is a natural starting point for candidates building a sharper Issue-task preparation plan, since it scores a timed response against the same 0–6 scale ETS uses.