A reader who has been trained only to ask “what does this say” walks into a craft and structure item and answers the wrong question. The choices in front of them describe what the writer is doing (qualifying a claim, setting up a contrast, signaling a shift in the argument), yet the student reaches for the option that paraphrases what the writer literally said. That single reflex, content over function, is the most expensive habit a reader brings to this part of the Digital SAT, and it is the habit this guide exists to retrain. Master one reframing move, hold it across every sentence you meet, and a fuzzy category of items collapses into something concrete and fast.

Craft and structure items live in the Reading and Writing section and test how and why a passage is built rather than the surface information it carries. They reward a meta-reading habit: standing one step back from the words to watch the moves a writer makes. The reframing question that powers the whole domain is short. Stop asking what the writer is saying. Start asking what the writer is doing. The first phrasing pulls you toward summary; the second pulls you toward function, and function is exactly what these prompts measure. By the end of this guide you will be able to name the job of any sentence, classify the architecture of a short text, read the effect of a chosen word, identify the intended reader, and answer a “primarily serves to” item without rereading the whole excerpt three times. You will leave with the InsightCrunch doing-not-saying frame and a function menu you can apply to your own practice.

Where craft and structure sits in the Reading and Writing section

The College Board groups the Reading and Writing content into four broad areas, and craft and structure is one of them. On the current digital exam its companions are Information and Ideas, Standard English Conventions, and Expression of Ideas. Craft and structure itself splits into three named families: words in context, text structure and purpose, and cross-text connections. This guide concentrates on text structure and purpose, on the function of sentences and paragraphs, and on the way word choice and intended audience shape a writer’s craft, because those are the facets students misread most often. Cross-text connections, where two short passages are set against each other, earns its own focused treatment in the cross-text connections guide; words in context, the connotation and precise-meaning family, is mapped in the vocabulary core. The work here is the architecture of a single piece and the reasoning about why a writer arranged it the way they did.

Every Reading and Writing item on the digital form follows the same outward shape. You meet a short, self-contained excerpt, usually a single dense paragraph drawn from literature, history, social science, or the natural sciences, followed by one multiple-choice prompt with four options. The excerpt is brief by design, often under one hundred and fifty words, which means a craft and structure item gives you very little real estate to work with and rewards precision over patience. There is no penalty for a wrong selection, so every prompt deserves an answer, but the brevity of each text is the strategic fact to internalize: you are reading a paragraph, not a chapter, and the job of every clause inside it is usually findable in a single careful pass.

How often do these items appear? Frame the frequency the way the test is built rather than as a fixed tally. Craft and structure is a standing portion of every Reading and Writing form, distributed across both modules, and the text-structure and purpose subtype shows up reliably enough that no student can treat it as a rare edge case. Because the section is module-adaptive, your performance in the first module routes you into an easier or harder second module, and craft and structure items appear in both routes. The harder route tends to carry items where the function is more subtly disguised, where a sentence does two jobs at once, or where two answer choices both describe plausible moves and only one matches the actual logic of the excerpt. Understanding the adaptive module design helps you read why a craft item in your second module may feel slipperier than the same category felt in your first.

The deeper orientation point is conceptual. Information and Ideas asks what a text claims and what its evidence supports; it lives at the level of content. Craft and structure asks how the writer assembled that content and to what end; it lives one level above, at the level of construction. A reader who never separates these two levels will keep importing a content answer into a function prompt and keep missing points that, with the right frame, are among the most learnable in the whole section. The good news buried in that difficulty is that craft and structure is a small, finite set of moves. Writers do a limited number of things: they inform, argue, compare, refute, propose, analyze, describe, and narrate. They build texts in a limited number of patterns. Sentences perform a limited number of jobs. Once you have the menu, the reading becomes recognition rather than invention.

The mechanics: what each craft and structure prompt is actually measuring

To answer these items reliably you need a precise model of the five things the test probes under this banner. Each is a distinct skill, each has its own tell in the answer choices, and each rewards a slightly different reading move.

What are craft and structure questions on the SAT?

They are the items that test how and why a passage is built rather than its surface content. They ask you to identify a writer’s overall aim, the organizational pattern of a text, the job a particular sentence or section performs, the effect of a chosen word, or the reader a piece addresses. The unifying skill is reading for function.

The first probe is overall purpose. A purpose prompt asks for the aim of the whole excerpt, usually phrased as “which choice best states the main purpose of the text.” The correct option names a verb of intent that fits the entire piece, not one clause of it. The menu of intents is short and worth memorizing as a working vocabulary: a writer can be arguing for a position, informing a reader of a development, comparing two things, refuting a competing claim, proposing a course of action, analyzing how something works, describing a scene or object, or narrating a sequence of events. Each verb implies a different shape, and the wrong choices in a purpose item almost always swap one of these verbs for a near neighbor: they call an analysis a description, or an argument a report, betting that you will not check the verb against the actual moves in the text.

The second probe is text structure, the organizational architecture of the piece. A structure prompt asks how the excerpt is arranged from first sentence to last, and the recurring patterns are few enough to hold in mind. A text may run cause and effect, tracing how one event produces another; compare and contrast, setting two things side by side; problem and solution, naming a difficulty and then a response; chronological, marching through time; or claim, evidence, and counterargument, the argumentative spine in which a position is stated, supported, and then tested against an objection. Recognizing the pattern is faster than narrating it, and the test rewards the reader who can label the architecture after one pass.

What does word choice tell you on a craft and structure item?

Word choice items ask what a specific chosen word accomplishes beyond its dictionary meaning. The right answer ties the word’s connotation to the tone or emphasis the writer is building. A neutral synonym would not carry the same charge, so the prompt is really asking why this word and not a flatter one, and the correct option names the effect rather than restating the definition.

The third probe, then, is the function of a sentence or a paragraph, and it is the heart of the domain. A function prompt isolates one sentence or one section and asks what job it does inside the larger structure. The verbs of function are a tighter menu than the verbs of purpose, and learning them as a fixed set is the single highest-yield move in this guide. A sentence can introduce an idea, support a prior statement with evidence or example, raise a counterargument, transition from one part to another, qualify or limit a claim, illustrate an abstract point with a concrete case, or conclude a line of reasoning. The “primarily serves to” phrasing is the classic signal, and it is asking for one of these jobs, not for a summary of the sentence’s content.

The fourth probe is word choice and its effect on tone and emphasis. Here a particular word is highlighted, and you must say what it does for the writer’s stance. A connotation-loaded verb can make a description admiring or skeptical; an unexpected adjective can sharpen emphasis or undercut a claim. The trap is to answer with the word’s denotation, its plain dictionary sense, when the prompt wants its effect, the coloring it adds. The fifth probe is intended audience and register. Some items ask who the writer is addressing, and the evidence sits in the diction and the assumed background knowledge: a piece thick with specialized terminology and unexplained references addresses experts, while a piece that defines its terms and builds from the ground up addresses newcomers. Register, the level of formality, is the companion clue.

How is craft and structure different from information and ideas?

Information and ideas items test what a passage says: its central claim, the detail that supports a conclusion, the inference the evidence allows. Craft and structure items test how the passage is built and why: the function of its parts, its organizing pattern, the effect of its diction. One reads for content; the other reads for construction. The same excerpt can host either kind of prompt.

The reason this distinction matters mechanically is that the two families reward opposite reading instincts. On an information item, the content of a sentence is the answer. On a craft item, the content of the sentence is the distractor, and its job is the answer. A student who has drilled inference and main-idea work without ever flipping that switch will read a “primarily serves to” prompt, locate the highlighted sentence, paraphrase what it says, find the choice that matches that paraphrase, and select a wrong answer that the test wrote precisely to catch them. The whole skill set in this guide is built to flip that switch on command, and the way you flip it is the reframing move from the opening: in front of every craft sentence, ask what the writer is doing here, not what the writer is saying.

The function menu: a “what is the author doing” reference

The findable artifact for this guide is a compact menu of the jobs a sentence or section can perform, each paired with a cue phrase you can scan for. Call it the InsightCrunch function menu. When a “primarily serves to” prompt appears, you are not searching the whole space of possible meanings; you are matching the highlighted line to one of these limited roles. Hold the menu in working memory and the prompt becomes a recognition task.

Function (what the writer is doing)	Cue phrase in the text	What the correct choice will say
Introduces	opens a topic, names a subject for the first time	“introduces the idea that…”
Supports	for example, for instance, in fact, evidence such as	“provides evidence for the preceding claim”
Illustrates	a concrete case after an abstract statement	“offers a specific example of a general point”
Raises a counterargument	however, critics contend, some argue, yet others	“presents an objection the writer will address”
Transitions	meanwhile, turning to, having established	“shifts the discussion from one part to the next”
Qualifies	although, to a degree, in most cases, with exceptions	“limits the scope of an earlier assertion”
Concludes	therefore, thus, in light of this	“states the conclusion the reasoning has built toward”
Defines	that is, in other words, is the term for	“clarifies a term for the reader”

Two disciplines make the menu work. First, decide the job before you read the four choices, so the options cannot lead you toward a content paraphrase. Second, test your candidate job against the sentences on either side of the target, because function is relational: a sentence qualifies something, supports something, transitions between things. The neighbors are the evidence. A line that begins with “although” almost certainly qualifies the claim that precedes it; a line that opens with “for example” almost certainly illustrates it. The cue phrases are not infallible, but they point you at the right region of the menu fast, and the surrounding sentences confirm or correct the guess.

Keep this menu distinct from a tone vocabulary. Tone words name a writer’s attitude (skeptical, admiring, wry), and they belong to the tone and attitude guide; function words name a writer’s move. A sentence can perform the function of raising a counterargument while carrying a neutral tone, so do not let the emotional coloring of a line decide its job. The menu above answers the structural question; tone answers a different one.

Eight worked examples, one per subtype

The fastest way to internalize the doing-not-saying frame is to watch it run on the actual item types. Each example below is solved as a tutor would narrate it: read the excerpt, name the move, then check the move against the choices. Every solution ends with the principle that carries to the next item of its kind. The excerpts are written in the style and length of the digital exam.

Worked example one: a primary-purpose item

Read this excerpt. Botanists once treated the Venus flytrap’s snapping leaves as a curiosity with no clear advantage, since the plant grows in soil rich enough to sustain it. Recent field studies suggest a different account. The traps capture insects not for ordinary nourishment but for nitrogen, an element scarce in the boggy ground the species favors. Far from a quirk, the mechanism appears to be a precise adaptation to a specific shortage.

The prompt: Which choice best states the main purpose of the text? The choices offer four intent verbs: to describe the appearance of a plant, to argue that an earlier view was mistaken, to compare two species of carnivorous plant, and to narrate the discovery of a new species.

Run the frame. What is the writer doing across the whole excerpt, not in any single line? The opening reports an old reading of the trap, the middle introduces newer evidence, and the close reverses the old reading in favor of the new one. That arc, stating a prior account and then overturning it with evidence, is the shape of a refutation. The aim is to argue that the earlier dismissal was wrong. Test the distractors: there is description in the excerpt, but description is not the point the piece builds toward, so the first choice mistakes a part for the whole. No second species appears, so the comparison option fails on a factual check. The text reports a revised explanation, not the discovery of a new organism, so the narration option misnames the content. The argument choice survives. The principle: a purpose answer must fit the entire arc, and the easiest way to find it is to track how the excerpt’s stance shifts from first sentence to last.

Worked example two: a text-structure identification

Read this excerpt. Urban heat islands form because dark pavement and dense building stock absorb sunlight by day and release it slowly after dusk. The trapped warmth raises nighttime temperatures, strains power grids as residents run cooling systems longer, and worsens air quality. Cities have begun to respond. Reflective roofing, expanded tree canopy, and lighter road surfaces all lower surface temperatures, and several pilot programs report measurable nighttime cooling within a single season.

The prompt: Which choice best describes the overall structure of the text? The choices: it presents a problem and then describes responses to it; it lists events in the order they occurred; it compares two competing theories; it defines a term and then gives its history.

Name the architecture. The first half identifies a difficulty, the heat island and its harms, and the second half turns to measures that ease that difficulty. That is the problem-and-solution pattern in clean form, and the pivot sentence, “Cities have begun to respond,” is the hinge between the two halves. The chronological option is tempting only if you mistake the logical sequence for a time sequence, but the piece is not marching through dated events; it is moving from a difficulty to its remedies. No two theories are weighed, so the comparison choice fails. The definition-and-history option misreads the opening, which explains a cause rather than fixing a term. The problem-solution choice is correct. The principle: locate the hinge sentence, the line where the text turns, and the pattern usually announces itself at that pivot.

Worked example three: a sentence-function item

Read this excerpt. Coral reefs support a quarter of all marine species despite covering a tiny fraction of the ocean floor. (1) This concentration of life depends on a fragile partnership between coral polyps and the algae living in their tissues. (2) When water warms beyond a narrow band, the polyps expel the algae and lose their main food source, a process called bleaching. (3) Yet some reefs in naturally warm lagoons show resistance that puzzles researchers. Studying these outliers may reveal which traits let coral endure.

The prompt: Which choice best describes the function of sentence 3 in the text? The choices: it summarizes the central finding of the passage; it introduces an apparent exception to the pattern just described; it defines a scientific term; it restates the opening claim in different words.

Apply the menu before reading too closely. Sentence 3 opens with “Yet,” a contrast cue, and it names reefs that behave against expectation after the prior line described the normal collapse. A line that opens with a contrast marker and presents a case running against the established pattern is raising an exception, which on the menu sits beside the counterargument role. The correct choice names exactly that move. The summary option fails because no finding is summarized; the close of the excerpt only proposes further study. The definition choice misfires because the defined term, bleaching, appeared in the previous sentence, not this one. The restatement option is wrong because sentence 3 introduces new information rather than echoing the opening. The exception choice is correct. The principle: a contrast cue at the head of a sentence is a strong signal that its function is to complicate or qualify what came before, so let the connective steer your first guess.

Worked example four: a paragraph-function item

This is the item the brief flags as the recurring shape, the “the third paragraph primarily serves to” prompt, so it deserves a full walk-through. Imagine a four-paragraph piece on the history of standardized timekeeping. The first section establishes that towns once kept their own local time by the sun. The second section explains how railways, running on fixed schedules across many towns, made local time chaotic and dangerous. The third section reads: Railroad managers, unwilling to wait for legislation, met in 1883 and divided the country into uniform zones of their own design, setting clocks by agreement rather than by statute. The fourth section notes that governments later ratified the scheme the companies had already imposed.

The prompt: The third paragraph primarily serves to. The choices: describe the natural method of timekeeping that preceded clocks; identify the practical pressure that made local time unworkable; present the specific action that resolved the problem the previous section raised; predict the future of global time coordination.

Walk the structure. The second section posed a difficulty: railways made scattered local times dangerous. The third section reports the railroads’ response, dividing the country into zones. So the third section’s job, relative to the second, is to supply the resolving action for the problem just established. The correct choice names that. The first distractor describes the content of the first section, not the third, a classic misplacement trap that rewards careless paragraph tracking. The second distractor describes the second section’s job. The prediction choice invents a forward-looking move the text never makes. The resolving-action choice is correct. The principle: a paragraph-function prompt is answered by reading the target section against its neighbors and asking what role it plays in the unfolding line of thought, and the wrong choices will frequently describe the true job of an adjacent section to bait a reader who tracks position loosely.

Worked example five: a word-choice effect item

Read this excerpt. The committee’s report did not merely question the proposal; it dismantled it, taking each projected saving apart until little of the original case remained standing.

The prompt: Which choice best describes the effect of the word “dismantled” as it is used in the text? The choices: it suggests the report was incomplete; it conveys the thoroughness with which the report took the proposal apart; it implies the committee acted carelessly; it indicates the report was brief.

This is the connotation move, and the trick is to read the chosen verb against its flatter alternatives. The writer could have written “criticized” or “questioned,” and indeed the sentence already used “question” in its first half before escalating with “did not merely.” The leap to “dismantled,” a verb of taking a structure apart piece by piece, is reinforced by the clause that follows, “taking each projected saving apart.” The effect the diction builds is methodical thoroughness, a deliberate demolition rather than a glancing complaint. The correct choice names that thoroughness. The incompleteness option contradicts the image of nothing left standing. The carelessness option reverses the meaning, since dismantling here is precise, not sloppy. The brevity option ignores the verb’s actual force. The thoroughness choice is correct. The principle: on a word-choice item, ask what flatter synonym the writer rejected and why, because the answer lives in the gap between the loaded word and its neutral cousin.

Worked example six: an intended-audience item

Read this excerpt. Practitioners weighing a switch to the newer assay should note that, while its limit of detection improves on the legacy method by roughly an order of magnitude, cross-reactivity with structurally similar analytes has not been fully characterized, and validation against the established reference panel remains incomplete.

The prompt: The text was most likely written for an audience of. The choices: general readers curious about science; students encountering laboratory work for the first time; specialists already familiar with diagnostic testing methods; policymakers deciding how to fund research.

The evidence is in the diction and the assumed background. The piece uses unexplained specialized terms, “limit of detection,” “cross-reactivity,” “analytes,” “reference panel,” and offers no glosses for any of them. It addresses “practitioners weighing a switch,” people already doing the work and choosing between methods. That assumed expertise is the tell. The correct choice names specialists. The general-reader option fails because no term is defined for a lay reader. The first-time-student option fails for the same reason; a beginner text would unpack the vocabulary. The policymaker option misreads the content, which is technical guidance on method selection rather than a funding case. The specialist choice is correct. The principle: intended audience is read from what the writer assumes the reader already knows, so scan for undefined technical terms and the level of background the prose takes for granted.

Worked example seven: a transition-function item

Read this excerpt. The early printing presses dramatically lowered the cost of reproducing a book. (1) Within decades, texts that had cost a year’s wages circulated for the price of a meal. (2) ____ literacy rates did not rise immediately, since reading instruction remained scarce and many households had no use for the printed page in daily life.

The prompt: Which choice completes the text with the most logical transition? The choices: For example; Therefore; Nevertheless; Similarly.

Decide the relationship before scanning the options, the discipline drilled in the transition-question method. The first sentence and its support describe a sharp fall in book prices, an apparent driver of reading. The sentence to be completed says literacy did not climb right away. The logical relationship between “prices fell” and “literacy stayed flat” is contrast: the expected consequence failed to follow. The transition must signal that reversal. “Nevertheless” marks exactly that contrast. “For example” would signal illustration, but the second clause is not an instance of cheap books; it is a countercurrent. “Therefore” would signal a result, the opposite of what the logic supports. “Similarly” would signal a parallel, but the two ideas pull against each other. The contrast transition is correct. The principle: name the logical relationship from the surrounding sentences first, then match the connective, because a transition item is a function item in miniature, asking what job the linking word must do.

Worked example eight: a qualify-the-claim item

Read this excerpt. Regular aerobic exercise improves cardiovascular health across nearly every population studied. (1) The benefits hold for young and old, for the sedentary beginning a routine and the already active maintaining one. (2) These gains, however, depend on consistency: occasional bursts of activity separated by long inactive stretches produce far smaller improvements than steady, moderate effort sustained over months.

The prompt: Which choice best describes the function of sentence 2 in the text? The choices: it provides an additional example of the benefits; it states the central claim of the passage; it limits the conditions under which the stated benefits occur; it introduces an opposing view that rejects exercise.

Sentence 2 opens with “however” and then narrows the broad claim of the first sentences by attaching a condition: the gains require consistency. Narrowing a general assertion by attaching conditions is the qualifying move on the function menu. The correct choice names that limitation. The additional-example option misreads a restriction as a reinforcement. The central-claim option is wrong because the central claim, that aerobic activity helps the heart, sits in the first sentence; the second sentence constrains it. The opposing-view option overstates the move; sentence 2 does not reject the benefit, it bounds it. The limiting choice is correct. The principle: a sentence that begins with a contrast marker and then attaches a condition is almost always qualifying, restricting the scope of a claim rather than denying it, and distinguishing qualification from rejection is the fine discrimination this subtype rewards.

Further worked examples across the harder subtypes

The eight examples above cover one clean instance of each item shape. The set below pushes into the variants a second module tends to favor, where the move is doubled, disguised, or separated from its obvious cue. Work them the same way: read with a target, name the job, then cut by mismatch.

Worked example nine: a compare-and-contrast structure with a buried pivot

Read this excerpt. Solar and wind generation share an obvious appeal, since both draw on resources that never run out and emit nothing as they operate. Their rhythms, though, run opposite. A solar array peaks at midday and falls to nothing after dark, while a wind farm often produces most heavily at night and through the stormier months, when demand for light and heat climbs. A grid that pairs the two can smooth the gaps that either leaves alone.

The prompt: Which choice best describes the overall structure of the text? The choices: it traces the historical development of two technologies; it presents a shared trait of two sources and then contrasts their behavior; it argues that one source is superior to the other; it defines a technical term and lists its uses.

The opening names a likeness, both sources are renewable and clean, and the word “though” at the head of the third sentence pivots to a difference, their opposite daily and seasonal rhythms. That is compare-and-contrast in its standard two-move form: establish common ground, then mark the divergence. The buried pivot is the connective “though,” easy to skim past, and noticing it is what fixes the pattern. The historical option fails because no development over time appears. The superiority option misreads the close, which proposes pairing the two rather than ranking them. The definition option has no anchor in the text. The shared-trait-then-contrast choice is correct. The principle: in a compare item the pivot connective (“though,” “by contrast,” “on the other hand”) marks the seam between likeness and difference, and the structure is read off that seam.

Worked example ten: a claim, evidence, and counterargument spine

Read this excerpt. The case for four-day school weeks rests on a plausible logic: a shorter week should cut commuting costs and lift teacher morale. Districts that adopted the model do report savings and higher staff retention. Test scores, however, have moved little in most of these districts, and some families struggle to arrange care on the freed weekday, costs the savings figures rarely count.

The prompt: Which choice best describes the function of the third sentence in the text? The choices: it provides further evidence for the benefits of the model; it restates the central claim; it raises considerations that complicate the favorable case just built; it defines the four-day week.

The first two sentences build a favorable case: a logic for the model and reported savings and retention. The third sentence opens with “however” and introduces flat test scores and uncounted care costs, evidence that pulls against the case. Its job is to complicate the argument the prior sentences assembled, the counterargument move on the spine of claim, evidence, and counterargument. The further-evidence option reverses the direction, treating a complication as reinforcement. The restatement option is wrong because new, opposing information appears. The definition option has no basis. The complicating choice is correct. The principle: in a claim-evidence-counterargument structure, the line after the supporting evidence that opens with a contrast cue is doing the counterargument work, and its function is to test the case, not to extend it.

Worked example eleven: a “primarily serves to” with a double-duty sentence

Read this excerpt. Early cartographers filled the unknown edges of their maps with sea monsters and warnings. Such flourishes were not mere decoration. (target) Having sketched this older practice, we can now see how the blank spaces on a modern map carry their own quiet message, one of measured confidence rather than dread. The empty ocean today says “unremarkable,” where once it said “beware.”

The prompt: The sentence beginning “Having sketched this older practice” primarily serves to. The choices: provide an example of a sea monster; transition from the historical practice to its modern counterpart; define the term cartographer; argue that old maps were more accurate.

This sentence does two visible things. It refers back to the older practice (a backward glance) and it opens the discussion of modern maps (a forward turn). The phrase “Having sketched this older practice, we can now see how the blank spaces on a modern map…” is built as a bridge, and its primary structural job is to carry the discussion from the historical material to its present-day counterpart. The transition choice names that. The example option mistakes the content for the function and points at the wrong sentence entirely. The definition option is unanchored. The accuracy-argument option invents a claim the text never makes. The transition choice is correct. The principle: when a sentence both looks back and moves forward, its primary job is usually to transition, because the backward glance exists to launch the forward turn, and “primarily” asks for the move that serves the whole arc.

Worked example twelve: a transition item where the elegant word is the trap

Read this excerpt. The new alloy resists corrosion far better than the steel it replaces, extending the service life of bridges built from it. (1) It also weighs less, easing the load on supports and foundations. (2) ____ it costs several times as much per ton, which has slowed its adoption to a handful of showcase projects.

The prompt: Which choice completes the text with the most logical transition? The choices: Furthermore; Consequently; However; For instance.

The first two sentences list advantages: better corrosion resistance and lower weight. The sentence to complete names a drawback, high cost slowing adoption. The relationship between “two advantages” and “one serious drawback” is contrast, so the transition must signal a reversal. “However” marks it. “Furthermore” is the elegant trap: it reads smoothly and would fit if the third idea were another advantage, but the logic is opposition, not addition, so the polished-sounding word points the wrong way. “Consequently” signals a result the logic does not support; the cost is not a consequence of the lighter weight. “For instance” signals an example, but the cost is not an instance of the advantages. The contrast choice is correct. The principle: never let a sophisticated connective win on sound, because the test routinely offers a smooth-but-wrong addition word where the actual relationship is contrast; name the logic first, then match.

Worked example thirteen: a register-and-audience discrimination

Read this excerpt. So you have decided to start running. Good. Forget the gadgets and the training plans for now. The only thing that matters in week one is that you lace up three times, go easy enough to hold a conversation, and come home wanting to do it again. Everything else can wait.

The prompt: The register of the text suggests it is addressed primarily to. The choices: competitive athletes refining their technique; complete beginners taking up the activity; medical professionals advising patients; readers of a formal scientific report.

The diction is plain, warm, and direct, with short imperative sentences and a conversational opener, “So you have decided to start running. Good.” The advice is foundational, lace up three times, keep it easy, build the wish to return, with no specialized terminology and no assumed background. That informal register and ground-floor content address a newcomer, not an expert or a clinician. The beginner choice is correct. The competitive-athlete option fails because the advice is introductory, not technical. The medical option fails because the register is encouraging rather than clinical, and no patient relationship is implied. The formal-report option contradicts the casual tone outright. The principle: register, the level of formality, is read from sentence length, diction, and directness, and it works alongside assumed background knowledge to fix the intended reader; informal, foundational, jargon-free prose points to a beginner.

Worked example fourteen: a “primarily serves to” that introduces rather than concludes

Read this excerpt. (target) Consider what happens when a city bans cars from a single downtown street. Merchants there often predict ruin, certain that drivers are their lifeblood. Yet foot traffic frequently rises once the noise and danger recede, and several pedestrianized blocks have posted higher sales than before the change.

The prompt: The first sentence primarily serves to. The choices: state the conclusion the passage reaches; introduce the scenario the passage will examine; provide statistical evidence; raise an objection to a claim.

The opening, “Consider what happens when a city bans cars from a single downtown street,” sets up the case the rest of the excerpt will explore; it names the situation before any evidence or conclusion appears. Its job is to introduce the scenario, the opening move on the function menu. The conclusion option is wrong because the conclusion, that foot traffic and sales often rise, arrives at the end, not the start. The evidence option fails because the first sentence offers no data; it frames the topic. The objection option misreads an invitation to consider a case as an opposing claim. The introduction choice is correct. The principle: a sentence’s position in the arc constrains its likely job, an opening line tends to introduce and a closing line tends to conclude, so use position as a first filter before confirming with content and connectives.

Worked example fifteen: a cause-and-effect structure

Read this excerpt. When a volcano injects sulfur high into the stratosphere, the particles spread into a thin global haze that reflects a fraction of incoming sunlight back to space. Surface temperatures dip in the months that follow, sometimes by a measurable degree across whole continents. Harvests can falter, and historical records of cold, failed summers often trace back to a major eruption a year or two earlier.

The prompt: Which choice best describes the overall structure of the text? The choices: it weighs two explanations for a cooling trend; it follows a single causal chain from an eruption to its distant effects; it lists volcanic eruptions in chronological order; it defines stratospheric haze and gives examples.

The excerpt runs a chain: sulfur reaches the stratosphere, which creates a reflective haze, which lowers surface temperatures, which in turn harms harvests. Each link produces the next, the signature of a cause-and-effect structure traced through several steps. The two-explanations option fails because only one mechanism appears; nothing competing is weighed. The chronological option mistakes a causal sequence for a time line of separate events; the steps are linked by because, not merely by and then. The definition option misreads the opening, which describes a process rather than fixing a term. The causal-chain choice is correct. The principle: distinguish a causal chain from a chronological list by asking whether each step produces the next or merely follows it in time; the words of causation, “reflects,” “dip,” “can falter … trace back,” mark a chain, not a calendar.

Worked example sixteen: a defining-function sentence

Read this excerpt. The deep ocean hosts creatures that make their own light. (target) This ability, called bioluminescence, lets a fish lure prey, startle a predator, or signal a mate in water that sunlight never reaches. Some species carry the light in their own tissues; others borrow it from glowing bacteria they shelter.

The prompt: The sentence beginning “This ability, called bioluminescence” primarily serves to. The choices: introduce a counterargument; define a term and outline its uses; conclude the discussion; compare two species.

The target sentence names a term, bioluminescence, attaches it to the ability the opening described, and then lists what the ability does, luring prey, startling predators, signaling mates. Naming and clarifying a term for the reader is the defining function on the menu, here paired with a quick survey of uses. The counterargument option fails because nothing opposing appears; the sentence builds on the opening rather than resisting it. The conclusion option is wrong because the excerpt continues afterward with the two sources of the light, so this sentence sits mid-arc, not at the close. The comparison option misreads the sentence, which defines before any two species are set against each other. The defining choice is correct. The principle: a sentence that attaches a name to an idea, often signaled by “called,” “known as,” or “that is,” is performing the defining function, clarifying a term so the rest of the text can use it freely.

These additional cases round the worked set well past the eight required, and together they map the full range of moves the domain tests: every intent verb, every structural pattern, and every function on the menu, each shown in a clean form and then in a disguised one. Work them until the naming is automatic, because the gap between a reader who recognizes these moves instantly and one who reconstructs them from scratch each time is, in a timed module, the gap between a comfortable pace and a scramble.

Turning the frame into points: a repeatable method

Recognition of the subtypes is the foundation; a fixed solving routine is what converts that recognition into reliable marks under time pressure. The method below applies to every craft and structure item and takes only seconds once it is habitual.

Begin by reading the prompt before the excerpt when the item is a function or structure type. Knowing in advance that you are hunting for the job of sentence 3, or for the overall architecture, lets you read the short text with a target in mind rather than reading passively and then hunting backward. The excerpts are brief, so a single directed pass is usually enough. As you read, mark the connectives in your mind or, in the Bluebook application, with the highlighter: the “however,” “for example,” “therefore,” and “although” that signal the logical joints. Those joints are where function lives, and a reader who notices them on the first pass rarely needs a second.

How do you predict the answer before reading the choices?

State the function in your own plain words before you look at the options. Decide that sentence 3 “introduces an exception,” or that the piece “argues an old view was wrong,” then find the choice that matches your prediction. Predicting first stops the distractors from steering you toward a content paraphrase, which is their entire design.

The prediction step is the single most protective habit on these items, because the wrong choices are engineered to sound reasonable against the text. They borrow real words from the excerpt, they describe moves the passage genuinely makes somewhere, and they paraphrase content accurately. What they get wrong is the job of the specific line in question. If you walk into the four options with a function already named, you compare each option to your prediction rather than to the seductive surface of the text, and the engineered appeal of the distractors loses its grip. If you walk in without a prediction, the best-written wrong answer often reads more smoothly than the correct one, and smoothness is not correctness.

When you reach the choices, eliminate by mismatch rather than by attraction. A distractor on a function item usually fails in one of four diagnosable ways, and learning to name the failure speeds the cut. Some options describe the job of a different sentence or section, the adjacent-role trap from the timekeeping example. Some restate the content of the target line rather than its function, the content-not-function trap that the whole guide exists to defeat. Some name a move the text never makes at all, an invented function with no anchor in the prose. And some overstate or understate the move, calling a qualification a rejection or a counterargument a mere example, the degree trap from the exercise item. Run each surviving option against these four failure modes and the field narrows quickly.

How should you pace yourself on craft and structure items?

Treat them as medium-cost items. The excerpts are short, so reading is fast, but the function decision rewards a beat of thought, so do not rush the prediction. Aim to spend slightly less than your average per-item time on a clean structure prompt and slightly more on a subtle paragraph-function prompt in a harder module, banking the difference.

Pacing on the digital section is a whole-module calculation rather than a per-item stopwatch, and the Reading and Writing pacing guide lays out the full budget. The relevant principle for craft items is triage: a structure or purpose prompt with a clear hinge sentence is among the faster items on the form, so clear it quickly and move on, building a time cushion. A second-module paragraph-function prompt where two choices both look plausible is worth a few extra seconds of careful neighbor-checking, and the cushion you banked on the easy items funds that care. Never let a single subtle function item swallow the time of three clean ones; if two choices survive elimination and you cannot separate them, choose the one that better matches your pre-read prediction and move forward, because there is no penalty for the attempt and lingering costs you points elsewhere.

The connective-scanning habit deserves its own emphasis as a strategy, not just a reading tic. The logical joints of a short text are carried by a small, recurring set of words, and those words map almost directly onto the function menu. “For example” and “for instance” flag illustration. “However,” “yet,” “nevertheless,” and “still” flag contrast, which on a function item usually means a counterargument, an exception, or a qualification. “Therefore,” “thus,” and “consequently” flag conclusion or result. “Although,” “while,” and “despite” flag qualification or concession. “Furthermore,” “moreover,” and “in addition” flag continuation or support. Training your eye to catch these on the first pass turns the function decision into a near-automatic match, and the discipline transfers directly to the transition items, which test the same logical relationships from the opposite direction.

Practice is where the routine becomes reflex, and the place to drill it is on realistic Reading and Writing item sets with full worked solutions, so you can check not only whether you picked the right choice but whether you named the right function for the right reason. ReportMedic gives you free, unlimited Reading and Writing practice with worked answers and section-targeted sets, which means you can run a focused block of nothing but craft and structure items, immediately see the reasoning behind each correct option, and convert your reading of this guide into rehearsal the same afternoon. Doing a dozen function prompts in a row, predicting the job before each set of choices and then reading the explanation, builds the doing-not-saying reflex faster than any amount of passive review, because the feedback loop is tight and the rep count is high.

The four failure modes, examined in depth

Eliminating wrong choices is faster when you can name why a choice is wrong, and on craft and structure items the wrong options fail in four diagnosable ways. Learning to label the failure rather than merely sensing it sharpens your cut and protects you on the two-survivor sets that decide a high score.

The first failure mode is the adjacent-role trap: a choice accurately describes the job of a different sentence or section than the one named. This is the most common distractor on paragraph-function items, because a short, structured text hands the writer several real roles to choose among, and offering the true role of a neighboring section is far more tempting than offering an obvious nonsense answer. The timekeeping walk-through earlier turned on this: a choice described the job of the first section when the prompt asked about the third. Defeat it by confirming that your named role belongs to the line the prompt actually points at, not to the line beside it. When two choices both name plausible roles, ask which role belongs to the named element, and the adjacent-role distractor falls away.

The second failure mode is the content-not-function swap: a choice paraphrases what the target line says instead of what it does. This is the distractor built for the content reflex, and it is dangerous precisely because the paraphrase is usually accurate. A choice can be a flawless summary of a sentence and still be the wrong answer to a function prompt, because the prompt never asked for a summary. The protection is the prediction step: name the function in your own words before reading the options, so a content paraphrase, however faithful, no longer matches your prediction. If a choice tells you what the line reports rather than the move it makes, it is answering a question you were not asked.

The third failure mode is the invented-move trap: a choice names a function the text never performs at all. The named move sounds like the sort of thing a writer might do, but no line in the excerpt does it. A choice may say a sentence “predicts future developments” when the text makes no prediction, or “defines a key term” when nothing is defined. These are usually the easiest distractors to cut, because a quick check against the actual prose finds no anchor for the claimed move. The discipline is simply to verify that the function named in the choice actually occurs somewhere a careful reader can point to; if you cannot locate the move in the text, the choice is invented.

The fourth failure mode is the degree trap: a choice names a real move but at the wrong strength. It calls a qualification a rejection, a suggestion a proof, a counterargument a mere example, or an exception a contradiction. This is the failure mode behind most two-survivor sets in a harder module, where both surviving choices describe genuine moves and only the strength differs. The exercise-and-consistency item earlier hinged on it: sentence two qualified a claim rather than rejecting it, and the distractor overstated the move into a rejection. Defeat the degree trap by returning to the text and reading the strength of the language the writer actually used. A line that says benefits “depend on consistency” limits a claim; it does not deny it. Match the verb in the choice to the force of the prose, and the over- or understated option separates from the correct one.

Run a surviving choice against all four modes and the diagnosis is usually quick. Does it describe the wrong element, the content instead of the function, a move that never happens, or the right move at the wrong strength? A choice that survives all four checks is almost always the answer, and naming the specific flaw in each rejected option turns elimination from a feeling into a procedure you can trust under pressure.

Building a craft and structure study routine

Reading this guide gives you the frame; a short, deliberate routine turns the frame into a score gain. The most efficient practice is targeted, not scattered, and the error-analysis method for the whole section applies in concentrated form here.

Start by gathering craft and structure misses from your recent practice into one place and sorting them by subtype: purpose items in one group, structure items in another, then sentence-function, paragraph-function, word-choice effect, audience and register, and transitions. The sorting itself is diagnostic, because misses rarely spread evenly. Most students find a cluster, often paragraph-function items in a harder module or the degree-trap word-choice items, and that cluster is your highest-yield target. There is no value in drilling the subtypes you already answer cleanly; the points sit in the cluster.

Then drill the cluster in focused blocks rather than mixed sets, predicting the function before every set of choices and reading the worked solution after each item to confirm you named the right job for the right reason. A block of ten paragraph-function items in a row, each predicted and then checked, builds the doing-not-saying reflex faster than fifty mixed items, because the repetition is concentrated where you are weak and the feedback is immediate. ReportMedic’s Reading and Writing tool supports exactly this kind of section-targeted, worked-solution practice, so you can run a block of one subtype, see the reasoning behind each correct option, and tighten the specific decision you keep getting wrong. Track your accuracy on the cluster across a week, and when it climbs to match your accuracy on the subtypes you already own, move the routine to the next cluster.

The discipline that makes this routine work is honesty in the diagnosis. A miss is not “a hard reading question”; it is a paragraph-function item where you fell for the adjacent-role trap, or a word-choice item where you answered the meaning instead of the effect. Label the failure mode on every miss, and the pattern that emerges tells you precisely what to fix. That labeling habit, applied across the whole section, is the engine of real score improvement, and craft and structure is the domain where it pays the fastest, because the misses are so cleanly nameable once you have the function frame in hand.

The hard end: where craft and structure items get slippery

In an easier module the function of a line is usually marked by a clean connective and confirmed by an obvious neighbor. In a harder module the routing rewards the test with room to disguise the move, and the disguises follow predictable patterns worth rehearsing before test day.

The first hard variant is the double-duty sentence, a line that performs two jobs at once. A sentence can both raise a counterargument and transition to a new part of the discussion, or both supply an example and qualify the claim it illustrates. When a single line carries two functions, the four choices will often name both real jobs and ask for the primary one. The word “primarily” in the prompt is doing real work here. Decide which job the sentence does for the larger structure, not which job is locally visible. A line that gives an example while also turning the discussion toward a new topic is, at the structural level, usually transitioning, because the example is in service of the turn. Read the sentence’s effect on the whole arc, not just its internal grammar, and the primary job emerges.

The second hard variant is the two-survivor choice set, where elimination leaves two options that both describe plausible moves. This is the most common reason a strong reader misses a craft item in a second module. The separation almost always comes down to a subtle degree or scope difference: one choice says the sentence “rejects” a claim while the other says it “qualifies” a claim, or one says the paragraph “proves” a point while the other says it “suggests” a point. The text supports only one strength of verb. Return to the excerpt and ask whether the line truly overturns or merely limits, truly establishes or merely gestures. The exercise-and-consistency example earlier turned on exactly this distinction, qualification versus rejection, and the harder forms of these items live in that narrow gap. When two choices survive, the deciding evidence is the strength of the language the text itself uses.

Is a purpose item the same as a main-idea item?

No. A main-idea item asks what the passage is chiefly about, its content summarized; a purpose item asks what the writer is trying to accomplish, the aim behind the content. A text whose main idea is “coral reefs are fragile” may have the purpose of arguing that current protections are inadequate. The main-idea guide treats the content question; purpose is the intent question.

That distinction is the third hard variant, the purpose-versus-summary trap, and it catches readers on the broad “main purpose of the text” prompts. The distractors will offer an accurate summary of the content dressed up with an intent verb, betting you will accept a true statement about what the text says as an answer to what the text is for. Defeat it by checking the intent verb first. If the choice says “to describe” but the excerpt builds an argument, the verb is wrong even if the noun phrase after it summarizes the content correctly. The verb of intent, not the topic, is where purpose items are won and lost, which is why the short menu of intent verbs from the mechanics section repays memorization: argue, inform, compare, refute, propose, analyze, describe, narrate. Match the verb to the actual arc and the trap dissolves.

The fourth hard variant is the overlap with words in context. A word-choice effect item and a words-in-context item can look identical at a glance, since both highlight a single word. The difference is what they ask. A words-in-context item wants the word’s meaning in the sentence, the closest synonym, and it belongs to the family mapped in the vocabulary core. A craft word-choice item wants the word’s effect, what its connotation accomplishes for tone or emphasis. Read the prompt’s verb: “most nearly means” signals a meaning item, while “the effect of the word” or “the use of the word” signals a craft effect item. Answering the meaning when the prompt wants the effect, or the reverse, is a self-inflicted miss that careful prompt reading prevents.

A final note on the hardest structure prompts. In a demanding module the organizing pattern can be a hybrid: a text might open with a problem, narrate a chronological response, and close with an evaluation, so that no single pattern label fits cleanly. When the structure choices each capture part of the truth, choose the one that describes the dominant movement across the whole excerpt, the arc from first sentence to last, rather than the pattern of any one stretch. The hinge-sentence technique still applies; there may simply be more than one hinge, and the controlling structure is the one that governs the opening-to-closing shift. These hybrid items are uncommon, but recognizing that the test occasionally builds them keeps you from rejecting a correct answer because it does not match a textbook-pure pattern.

Why this skill carries beyond the section

The doing-not-saying frame is not a narrow trick for one item family; it is a reading habit that pays across the whole Reading and Writing section and, beyond the exam, into any work that demands close analysis of how a piece is built.

Inside the section, the same meta-reading move that answers a craft prompt also sharpens your work on information and ideas items. When you can name the function of each sentence, you locate a passage’s central claim faster, because the claim is usually the line that the other sentences support, illustrate, or qualify, and you have just learned to spot those supporting roles. The skill feeds the reading comprehension strategies that govern the harder content items, where finding the one sentence that carries the argument is half the battle. A reader who maps function reads structurally, and structural reading is faster and more accurate than the line-by-line content grind most students default to.

The frame also connects to the writing half of the section. Expression of ideas items ask you to add, revise, or arrange sentences to serve a stated rhetorical goal, and that is the same function reasoning viewed from the writer’s chair rather than the reader’s. When a rhetorical synthesis prompt asks which sentence best accomplishes a goal, you are choosing a function and matching it to an aim, exactly the move you make on a craft item in reverse. The rhetorical synthesis guide builds on the same foundation, and students who master craft and structure reading tend to find synthesis items easier, because both rest on understanding what a sentence is for.

Step back to the whole section and the strategic picture sharpens. The Reading and Writing portion is, at bottom, a test of whether you can read like an analyst rather than a consumer. A consumer reads for what a text tells them; an analyst reads for how the text works on them and why the writer made the choices they did. Craft and structure is the part of the exam that most directly rewards the analyst’s stance, which is why it is among the most learnable point sources in the section: the analyst’s stance is a trainable habit, not an innate gift. This is the series thesis in miniature. The test looks like a verdict on raw reading ability, but the craft and structure items in particular reward a specific, teachable way of attending to prose, and a student who practices the function frame improves in a way that surprises anyone who believes the score measures something fixed.

There is a broader payoff that outlasts the exam. The habit of asking what a writer is doing rather than only what they are saying is the core of critical reading in college and the workplace. It is how you read a persuasive editorial without being swept along, how you notice that a report’s structure is steering you toward a conclusion, how you separate an author’s evidence from an author’s framing. The complete preparation guide frames the whole test as a system of learnable skills, and craft and structure is the clearest case of an exam skill that is also a life skill. The few weeks you spend training the function frame for a Saturday morning will keep paying you long after the scores arrive.

For a sense of how this fits a target score, the craft and structure items are points that mid-band readers leave on the table most often, because the content-over-function reflex is so natural. A reader pushing from a solid score into the higher bands, the kind of move mapped in the 1400 to 1500 guide, frequently finds that their remaining missed items cluster in exactly this domain: the subtle paragraph-function prompts and the two-survivor choice sets of a harder module. That clustering is good news, because it means the remaining points sit in a small, named, drillable region rather than scattered randomly across the section. Diagnose your misses, find the craft items among them, and you have found a concentrated source of recoverable marks.

The mistake that costs the most, named and corrected

The single most expensive error on craft and structure items has a precise name, and naming it is the first step to never making it again. Students answer with content when the prompt asks for function. Confronted with “the third paragraph primarily serves to,” they locate the paragraph, paraphrase what it says, scan for the choice whose words most resemble that paraphrase, and select it. The selected choice is a faithful summary of the content and a wrong answer to the question, because the prompt asked what the paragraph does, not what it contains. This is the content-over-function reflex, and it is the habit the entire guide is built to break.

Why is the reflex so stubborn? Because most of a student’s reading life, in school and out, rewards reading for content. You read a chapter to learn what happened, an article to learn what is true, a problem to learn what is asked. Reading for function is an unusual stance that school rarely drills directly, so the test can rely on the content reflex firing automatically under time pressure. The correction is mechanical and reliable: before you read the choices, force yourself to say the job of the line in plain words, using a verb from the function menu. If the sentence you generate describes what the line says, you have answered the wrong question; rewrite it until it describes what the line does. “The paragraph explains that railroads created time zones” is a content sentence. “The paragraph presents the action that solved the problem the previous section raised” is a function sentence. Only the second kind matches the prompt.

A related and widespread myth deserves correction: that craft and structure items are subjective, a matter of interpretation where any defensible reading might earn the point. They are not. Each item has a single defensible answer anchored in the text, and the wrong choices fail for diagnosable reasons, the four failure modes named earlier. The feeling of subjectivity comes from answering with content, where several choices can feel partly right because several do summarize the text accurately. Switch to function and the subjectivity vanishes, because a line has one primary job relative to its structure, and the text’s own connectives and neighbors point to it. Students who believe these items are guesswork are usually students who have never made the function switch; once they do, the items become as determinate as a math problem.

A third misconception is that you must read the whole surrounding passage to answer a function prompt. On the digital exam each item is a short, self-contained excerpt, so there is no larger passage to consult; the evidence for a sentence’s job sits in the few lines around it within the same brief text. Students who imagine a missing context and answer based on what they assume the absent material says are inventing evidence. Everything you need is on the screen. Read the short excerpt closely, use the neighbors of the target line, and resist the urge to reason from a passage that does not exist. The Digital SAT format guide lays out the short-text design in full, and internalizing that design stops the imagined-context error before it starts.

The last common mistake is letting tone override function. A line can carry a sharp or admiring tone while performing a perfectly neutral structural job, and students sometimes let the emotional coloring decide the answer, picking a choice that names an attitude when the prompt asked for a move. Keep the two analyses separate. Decide the function from the menu and the connectives; decide the tone, when a different item asks for it, from the diction and stance. Conflating them sends you toward a choice that describes how the sentence feels rather than what it does, and the prompt, on a craft and structure function item, never asked how it feels.

Where to take this next

Return to the reflex this guide opened with, because it is the whole of the skill compressed into a habit. In front of every craft and structure line, the unprepared reader asks what the writer is saying and reaches for content. You now ask what the writer is doing and reach for function, naming the job from a short menu before the choices can lead you astray. That one swap, content to function, converts the slipperiest item family in the Reading and Writing section into a recognition task with a single defensible answer.

The path from here is short and concrete. Take the function menu and run it on real items until the match is automatic: read the excerpt, mark the connectives, name the job in your own words, then find the choice that fits your prediction and confirm by eliminating the four failure modes. Drill a focused block of nothing but craft and structure prompts on ReportMedic’s Reading and Writing tool, check the worked solution after each one, and pay attention not only to whether you chose right but to whether you named the right function for the right reason. Then fold the habit into your full-section work using the reading comprehension strategies and the pacing plan, so the function frame runs on every item, not just the ones labeled craft and structure.

A writer is always doing something on the page, and the reader who can name the move owns the question. Train the eye to see the moves, and the points that once felt like guesswork become the most dependable marks on the form.

Frequently Asked Questions

What are craft and structure questions on the SAT?

They are the Reading and Writing items that test how and why a text is built rather than the surface information it reports. Instead of asking what a passage claims, they ask what the writer is doing: identifying the aim of a whole excerpt, the organizational pattern that holds it together, the job a single sentence or section performs, the effect of a chosen word, or the reader the piece addresses. On the digital exam the domain gathers three named families, words in context, text structure and purpose, and cross-text connections, and the unifying skill across all of them is reading for function rather than content. The defining habit is the reframing move at the center of this guide: in front of every line, ask what the writer is doing here, not what the writer is saying. That single switch turns a fuzzy, seemingly subjective category into a recognition task with one defensible answer anchored in the text.

How do I find the primary purpose of a text?

Track how the writer’s stance moves from the first sentence to the last, then name the intent verb that fits the entire arc. A purpose prompt is not satisfied by summarizing the content; it wants the aim behind the content. Read the opening to see what the writer sets up, the middle to see what they develop, and the close to see where they land, then choose from the short menu of intent verbs: argue, inform, compare, refute, propose, analyze, describe, narrate. The correct option names the verb that governs the whole piece, while the wrong options usually swap in a near neighbor, calling an argument a report or an analysis a description, hoping you will not check the verb against the actual moves. Always test the verb first. If the choice says “to describe” but the excerpt builds toward a claim, the verb is wrong even when the topic phrase after it is accurate. The verb of intent is where these items are won.

How do I identify the structure of a passage?

Find the hinge, the sentence where the text turns, and the organizing pattern usually announces itself there. A short excerpt is built on one dominant movement, and the recurring patterns are few: cause and effect, where one thing produces another; compare and contrast, where two things sit side by side; problem and solution, where a difficulty is named and then answered; chronological, where events run in time order; and claim, evidence, and counterargument, the argumentative spine that states a position, supports it, and tests it against an objection. Read the excerpt once with the patterns in mind, locate the pivot line, and label the architecture by how the halves relate across that pivot. When a harder item blends patterns, choose the choice that describes the dominant arc from opening to close rather than the pattern of any single stretch. The hinge technique still applies; you are simply identifying which turn controls the whole movement of the text.

What does “the third paragraph primarily serves to” ask?

It asks for the job that the named section performs inside the larger line of thought, not for a summary of what that section says. The word “primarily” signals that the section may do more than one thing, and you want its main structural role relative to its neighbors. Answer it by reading the target section against the sections before and after it and asking what it contributes to the unfolding reasoning: does it introduce a problem, supply the resolving action, raise an objection, offer an example, or qualify an earlier claim? The most common trap is a choice that accurately describes the job of an adjacent section, baiting a reader who tracks paragraph position loosely, so confirm that your chosen role belongs to the section actually named. The second most common trap is a choice that paraphrases the section’s content rather than its function. Name the role in plain words before reading the options, and match the choice to that role.

How do I tell what the author is doing rather than saying?

Before you read the answer choices, force yourself to state the line’s job in plain words using a function verb: introduces, supports, illustrates, raises a counterargument, transitions, qualifies, concludes, or defines. Then check the sentence you generated. If it describes what the line reports, “the sentence explains that time zones were created by railroads,” you have written a content sentence and answered the wrong question. Rewrite it until it describes a move, “the sentence presents the action that resolved the previously stated problem.” Only the function version matches a craft prompt. The reason the content reflex is so strong is that nearly all ordinary reading rewards reading for content, so the function stance has to be deliberately switched on. The reliable trigger for the switch is the prediction step: name the job first, in your own words, so the content-flavored distractors cannot steer you. Once the job is named, you compare each option to your prediction rather than to the seductive surface of the text.

How does word choice affect tone on the SAT?

A chosen word carries connotation beyond its dictionary meaning, and a word-choice item asks what that connotation accomplishes for the writer’s tone or emphasis. The technique is to ask what flatter synonym the writer rejected and why. If a report “dismantled” a proposal rather than merely “questioned” it, the loaded verb builds an impression of methodical, thorough demolition that a neutral word would not carry, and the reinforcing clauses around it confirm the effect. The trap is to answer with the word’s plain denotation when the prompt wants its coloring, the charge it adds. Read the prompt’s verb to be sure which item you have: “most nearly means” wants the meaning and belongs to the words-in-context family, while “the effect of the word” or “the use of the word” wants the connotation and its work on tone. Tie the connotation to the stance the writer is building, name the effect rather than the definition, and check that the surrounding lines support the coloring you have chosen.

How do I identify the intended audience of a passage?

Read what the writer assumes the reader already knows. Intended audience is encoded in diction and in the background the prose takes for granted. A piece thick with unexplained specialized terms, addressing people already doing the work and choosing between methods, is written for specialists; a piece that defines its terms, builds from the ground up, and avoids assumed jargon is written for newcomers or general readers. Scan for undefined technical vocabulary and for the level of expertise the writer expects, and let that evidence decide. Register, the level of formality, is a companion clue: dense, formal, qualification-heavy prose signals a professional or academic reader, while plain, explanatory prose signals a wider one. The trap on these items is to choose an audience based on the topic rather than the treatment; a passage about a technical subject can still be written for newcomers if it explains its terms. Judge the audience by how the material is handled, not merely by what the material concerns, and the evidence will point to one reader.

What text-structure patterns should I recognize?

A small, recurring set covers nearly every short excerpt. Cause and effect traces how one event or condition produces another. Compare and contrast sets two things side by side to mark their likeness or difference. Problem and solution names a difficulty and then a response to it, with a clear hinge between the two halves. Chronological order marches through events in the sequence they occurred. Claim, evidence, and counterargument is the argumentative spine: a position stated, supported with reasons or examples, and then tested against an objection or qualification. Memorize these five as a working menu, because identifying a pattern is far faster than narrating one from scratch. When you read a structure prompt, hold the menu in mind, find the pivot line where the text turns, and ask how the parts relate across that pivot. Harder items occasionally blend patterns; when they do, pick the choice that names the dominant movement from first sentence to last rather than the local pattern of one stretch.

How do I find the function of a single sentence?

Use the function menu and let the connectives steer your first guess. A sentence that opens with “for example” or “for instance” is almost certainly illustrating the prior statement; one that opens with “however,” “yet,” or “nevertheless” is usually raising a counterargument, an exception, or a qualification; one that opens with “therefore” or “thus” is concluding; one that opens with “although” or “while” is qualifying. Read the target line, note its opening connective, and form a candidate job before looking at the options. Then confirm the job against the sentences on either side, because function is relational: a line supports, qualifies, or transitions in relation to its neighbors, and those neighbors are your evidence. Predict the job in plain words first so the content-flavored distractors cannot pull you toward a paraphrase. Finally, eliminate by mismatch: a wrong choice typically describes the job of a different sentence, restates the content instead of the function, names a move the text never makes, or over- or understates the move’s strength.

What is the difference between purpose and main idea?

A main-idea item asks what the passage is chiefly about, its content summarized into one statement; a purpose item asks what the writer is trying to accomplish, the aim behind that content. The two can diverge sharply. A text whose main idea is “coral reefs are fragile ecosystems” may have the purpose of arguing that current protections fall short. The main idea names the subject; the purpose names the intent. On a main-idea prompt you compress the content; on a purpose prompt you identify the governing intent verb, argue, inform, compare, refute, propose, analyze, describe, or narrate. The classic trap on a purpose item is a choice that summarizes the content accurately but attaches the wrong intent verb, betting you will accept a true statement about what the text says as an answer to what the text is for. Defeat it by checking the verb against the actual arc of the excerpt before you check the topic phrase. Match the intent to the movement of the piece, and the distinction stays clear.

How do I recognize a counterargument’s function?

Watch for a contrast cue at the head of the sentence and a shift toward a position that runs against what came before. A line that opens with “however,” “yet,” “critics contend,” “some argue,” or “others maintain” and then presents a competing claim or a case against the established pattern is performing the counterargument move, which sits beside the exception and qualification roles on the function menu. The job of such a line is to introduce an objection the writer will engage, complicate a prior statement, or present a case that resists the expected conclusion. The fine discrimination these items reward is between a true counterargument, which presents an opposing position, and a mere qualification, which only limits the scope of a claim without opposing it. Read whether the line genuinely sets up a competing view or simply attaches a condition. A sentence that says benefits “depend on consistency” qualifies; a sentence that says “critics reject this account entirely” introduces a counterargument. Match the strength of the move to the strength of the language the text actually uses.

How do transitions function within a passage?

A transition carries the logical relationship between two ideas, and a transition item asks you to choose the connective that names that relationship correctly. Decide the relationship from the surrounding sentences before you look at the choices: addition and continuation, contrast and concession, cause and effect, example and illustration, sequence and time, or emphasis and clarification. Then match the connective to the relationship you named. The recurring trap is a smooth-sounding word that does not fit the logic, such as “furthermore” where the ideas actually contrast, or “therefore” where the second idea undercuts rather than follows from the first. A transition item is a function item in miniature, asking what job the linking word must perform, so the same predict-first discipline applies: name the logical joint, then find the word that marks it, and reject any option whose meaning, however elegant, points the wrong way. Training your eye to read connectives as logical signals on the first pass makes both transition items and sentence-function items faster, because both turn on the same small set of relationships.

How is craft and structure different from information and ideas?

Information and ideas items test what a passage says: its central claim, the detail that supports a conclusion, the inference its evidence allows. Craft and structure items test how the passage is built and why: the function of its parts, its organizing pattern, the effect of its diction, the reader it addresses. One family reads for content; the other reads for construction, one level above. The same excerpt can host either kind of prompt, so the deciding factor is what the question asks, not what the text contains. The mechanical reason the distinction matters is that the two families reward opposite instincts. On an information item, the content of a sentence is the answer; on a craft item, the content is the distractor and the function is the answer. A reader who has drilled inference and main-idea work without ever flipping that switch will keep importing a content answer into a function prompt. Reading the prompt carefully to classify which family you face, then adopting the matching stance, prevents the most common cross-family error.

How do I answer a word-choice effect question?

Read the prompt’s verb first to confirm it wants an effect rather than a meaning, then ask what flatter alternative the writer passed over and why. An effect prompt says “the effect of the word” or “the use of the word,” while a meaning prompt says “most nearly means,” and the two want different things. For an effect item, treat the highlighted word as a deliberate choice over a neutral synonym, and name the coloring it adds: a charged verb can make a description admiring or scornful, an unexpected adjective can sharpen emphasis or undercut a claim. Confirm your reading against the lines around the word, because the surrounding prose usually reinforces the connotation the writer intends. The trap is to answer with the dictionary definition when the prompt wants the charge, or to pick a choice that names a tone the surrounding text does not support. Tie the connotation to the writer’s stance, choose the option that names the effect rather than restating the meaning, and you will avoid the meaning-versus-effect confusion that costs careless readers this point.

What is the most common craft and structure mistake?

Answering with content when the prompt asks for function. Faced with “the third paragraph primarily serves to,” a student locates the section, paraphrases what it says, finds the choice whose words most resemble that paraphrase, and selects a faithful summary that is the wrong answer, because the prompt asked what the section does, not what it contains. The reflex is stubborn because almost all ordinary reading rewards reading for content, so the function stance must be switched on deliberately under time pressure. The fix is mechanical: before reading the choices, state the line’s job in plain words using a function verb, and check that your sentence describes a move rather than a meaning. If it describes what the line reports, rewrite it until it describes what the line does. A close companion error is believing these items are subjective; they are not, because each has one defensible answer anchored in the text, and the wrong choices fail for nameable reasons. Switch from content to function and the apparent subjectivity disappears.