Digital SAT Craft and Structure module: how to triage Cross-Text Comparisons, Rhetorical Synthesis, and word-in-context stems in 75 seconds or less per item.
The Digital SAT Reading and Writing section dedicates one of its four content domains to Craft and Structure, and in the standard two-module adaptive routing this domain is the single largest share of items that decide the 600-to-700 spread. Students who treat Craft and Structure as a 'vibe check' — read the passage, trust the gut, pick the elegant answer — leave roughly 20 percent of the available points on the table. The fix is procedural: triage by stem shape, lock the text anchor before reading the choices, and learn to spot the four or five cross-text devices the College Board reuses across administrations. This article walks through the specific triage sequence that turns Craft and Structure from a soft skill into a clock-stable, score-stable domain.
What Craft and Structure actually measures on the Digital SAT
The College Board defines Craft and Structure as the domain where students interpret the meaning and purpose of words, phrases, and entire texts in context. In practical terms the module contains three item families, each with its own stem shape, its own timing budget, and its own failure pattern:
- Words in Context — usually 4 to 6 items per test, asking which choice best fits a specific sentence or pair of sentences.
- Text Structure and Purpose — usually 5 to 7 items per test, asking about the role of a sentence, paragraph, or the passage as a whole.
- Cross-Text Comparisons — usually 2 to 3 items per test, asking how two short passages relate.
In total a candidate sees roughly 13 to 15 Craft and Structure items across Modules 1 and 2 combined. Because each item contributes one raw point and the adaptive engine keys on raw accuracy to route Module 2 difficulty, missing 3 of these is roughly the difference between landing in the easy versus hard Module 2 branch on a typical sitting. That is why 50 scaled points can disappear without a single 'hard' item appearing on screen — the routing decision itself is the penalty.
The 75-second triage sequence for any Craft and Structure item
Craft and Structure rewards a fixed micro-routine. Train the routine in untimed practice until it is automatic, then run it on test day. The sequence is five steps:
- Read the stem, not the passage. The stem tells you which item family you are facing and which sentence or sentences you must anchor on.
- Locate the anchor. For Words in Context, the anchor is the highlighted word or phrase. For Text Structure and Purpose, the anchor is the sentence or paragraph named in the stem. For Cross-Text Comparisons, the anchor is the line or claim in Passage A that the stem names.
- Read 8 to 12 words around the anchor, never the whole passage. Craft and Structure items almost never depend on global comprehension.
- Pre-decide the function. Before looking at choices, name in one phrase what the anchor is doing — defining, contrasting, qualifying, conceding, exemplifying.
- Scan the four choices for the one that matches the function. If two survive, pick the one whose diction matches the surrounding register.
For most candidates reading this, the single biggest gain comes from step 4. The pre-decision compresses the four choices from a reading task into a matching task, and the matching collapses the time budget from 110 seconds to roughly 75.
Cross-Text Comparisons: the only Craft and Structure family that demands two passes
Cross-Text Comparisons appear in both Module 1 and Module 2, and they are the family most likely to push past the 90-second budget. The reason is structural: the item asks you to compare two passages that often argue in opposite directions, and the four choices will include one 'they agree', one 'they disagree', and two 'they agree on X but disagree on Y' formulations. The trap is to commit to a binary read after the first passage and then read the second passage through that lens.
The clean fix is a two-pass protocol. On pass one, underline the explicit claim in each passage — one sentence, written in the margin. On pass two, compare claims, not passages. Most candidates reading this skip pass one and rely on tone, and that is exactly how a 3-point swing on Cross-Text Comparisons disappears in a single sitting.
Words in Context: the four signals that mark a hard stem
Words in Context items look the easiest on the surface and quietly cost the most marks in the 600-to-700 band. Four stem signals separate a routine item from a hard one, and recognising them in advance lets you budget time accordingly:
- The stem quotes a clause rather than a single word — the answer must preserve the relationship, not the literal meaning.
- Two choices are near-synonyms and two are not — the decision turns on a two-letter prefix or suffix.
- The stem says 'most nearly' rather than 'best' — a one-word shift that signals the answer is approximate, not exact.
- The anchor sits at a sentence boundary — the word is doing transition work, not noun work.
When you spot any of these four signals, the routine is to read the full sentence twice. The second reading is not optional; it is the difference between the routine pick and the trap pick.
Text Structure and Purpose: the role-of-the-whole-passage question
Once per module the College Board asks a 'primary purpose' or 'central claim' item that anchors on the passage as a whole. These items punish students who anchor on the first sentence and ignore the qualifier in the second half. A practical rule: read the first and last sentence before reading the stem. If those two sentences agree in direction, the passage is doing one job; if they push in different directions, the passage is doing two jobs and the correct answer will name the second.
Common pitfalls and how to avoid them
Craft and Structure items have a small number of recurring failure modes. The four I see most often in a 600-to-700 scorer:
- Choosing the choice that is true in the real world but not true in this passage. The Digital SAT rewards text-internal logic, not background knowledge.
- Reading the whole passage before reading the stem. Reading order is stem first, anchor second, around-the-anchor third, choices last.
- Spending more than 90 seconds on a single item. The correct response to a stuck item is to mark and move; the routing engine is more sensitive to accuracy per minute than to a single perfect answer.
- Confusing Text Structure with Central Ideas. Text Structure asks what the author is doing; Central Ideas asks what the author is saying. The stem verb disambiguates, and a 2-second check on the verb saves the point.
Craft and Structure versus the other three Reading and Writing domains
Craft and Structure is the second of four content domains, and it overlaps at the edges with Information and Ideas, Expression of Ideas, and Standard English Conventions. The table below maps the boundary so you can route a stem to the right routine in under five seconds.
| Domain | Stem trigger | Anchoring rule | Time budget |
|---|---|---|---|
| Craft and Structure — Words in Context | 'best fits', 'most nearly' | Highlighted word, plus 8–12 words around it | 65–80 seconds |
| Craft and Structure — Text Structure and Purpose | 'primarily serves to', 'main idea of the passage' | Named sentence or first-and-last sentence | 75–90 seconds |
| Craft and Structure — Cross-Text Comparisons | 'both authors', 'differs from Passage B' | One claim per passage, written in margin | 90–110 seconds |
| Information and Ideas — Command of Evidence | 'which finding', 'most strongly supports' | Quantitative or textual data point | 70–85 seconds |
| Expression of Ideas — Rhetorical Synthesis | 'most effectively', 'transitions' | Two source sentences, target sentence | 80–100 seconds |
How Craft and Structure feeds the adaptive routing
The Digital SAT uses a multistage adaptive design: performance on Module 1 routes the candidate to an easier or harder Module 2, and Module 2 raw score is what scales. Craft and Structure contributes roughly 13 to 15 of the 27 raw points in a Reading and Writing module, which means a candidate who goes 11 of 15 on Craft and Structure in Module 1 has effectively locked in the hard branch even if they miss every other item. In my experience the cleanest indicator of routing success is the per-domain accuracy at the end of Module 1; a domain-by-domain scorecard tells the student which section to drill in the week before the sitting.
Building a one-week Craft and Structure drill plan
For most candidates a focused week is enough to move Craft and Structure by 30 to 50 scaled points. The plan has three blocks, each anchored on a specific item family:
- Days 1–2: 40 Words in Context items, untimed, with the pre-decision protocol. Score target: 36 of 40.
- Days 3–4: 30 Text Structure and Purpose items, timed at 80 seconds each. Score target: 24 of 30.
- Days 5–7: 20 Cross-Text Comparisons, with the two-pass protocol. Score target: 16 of 20.
The drill is not a question bank sweep. The drill is the same item re-run until the protocol is automatic. A student who can run the five-step routine without thinking about it will outscore a student who has memorised a thousand items but still reads the whole passage first.
Conclusion and next steps
Craft and Structure rewards a procedural, not intuitive, approach. The five-step triage sequence, the two-pass protocol for Cross-Text Comparisons, and the four Words in Context stem signals together convert a soft-skill domain into a clock-stable one. The next step is a domain-by-domain scorecard on the last two full-length practice tests, used to choose which of the three item families to drill first. SAT Courses' Digital SAT Reading and Writing Craft and Structure programme builds that scorecard against the official Bluebook interface and turns the per-domain routing decision into a concrete weekly plan.