Which Inference Can Be Best Made About How Steve Feels About Goodman: A Close-Reading Framework

The best inference is b: Steve feels thankful for Goodman. In most passages built around this question, Goodman acts as a benefactor—offering help, resources, or a second chance—while Steve's responses signal relief rather than fear, anger, or a protective stance. The correct option depends on the exact text, but gratitude is the inference the evidence most consistently supports. Below is a full method for verifying that claim, weighing competing options, and avoiding the traps that make thin explainers unreliable.

How inference works in reading-comprehension items

An inference is a conclusion built from textual evidence plus background knowledge, not a guess and not a restatement. Test designers construct items so that one option is supported by multiple clues, one or two are plausible distractors, and one is contradicted. The mechanism matters: the correct answer usually requires combining at least two data points—an action and a reaction, a cost borne and a benefit received.

For "which inference can be best made about how Steve feels about Goodman," the decisive evidence is almost always behavioral. Steve accepts help he could not obtain alone. He returns, thanks, or defers to Goodman. He defends Goodman when others criticize him. Each behavior is a low-cost signal; two or three together form a pattern that outranks a single ambiguous line.

Reading scientists describe this as coherence-building: readers connect clauses into a causal chain rather than storing isolated facts. A grateful inference emerges when the chain runs "Goodman gave something costly; Steve received it; Steve's later behavior toward Goodman is positive." The competing options each require a different chain, and those chains usually lack a second link.

The four options, ranked by evidentiary demand

Each answer choice makes a distinct claim about Steve's internal state, and each claim requires different proof. Protective feeling requires Steve to shield Goodman from harm or criticism. Thankfulness requires Steve to acknowledge a benefit. Anger requires friction, grievance, or retaliation. Fear requires avoidance, flinching, or compliance under threat. Only one of these is typically demonstrated by action rather than asserted by narration.

The table exposes the logic. Protective and afraid both require Steve to act against his own interest for Goodman-related reasons, and passages rarely supply that. Angry requires negative affect that contradicts visible cooperation. Thankful requires only that a benefit occurred and that Steve's behavior reflects it—the lowest evidentiary bar among the four, which is why it wins when the text is otherwise sparse.

Why gratitude beats fear and anger in most passages

Fear produces avoidance. A frightened Steve would decline Goodman's offer, leave the room, or minimize contact. When a passage instead shows Steve seeking Goodman out, staying, and following through, fear is contradicted, not merely unsupported. This is the counterintuitive detail many thin explainers miss: the absence of avoidance is affirmative evidence against option d, not neutral silence.

Anger produces friction. An angry Steve would push back, withhold cooperation, or speak against Goodman. If the passage shows him thanking Goodman, defending him, or crediting him later, anger loses its footing. The distractor survives only when the text contains a single sharp exchange—one sarcastic line, one moment of frustration—with no follow-up. Isolated affect is weak evidence; sustained behavior is strong.

Gratitude produces reciprocity. Steve's thanks, return favors, or public acknowledgment all point the same direction. The inference is "best" not because it is certain but because it requires the fewest assumptions. Gratitude also fits the common narrative arc in which a benefactor's cost is emphasized precisely so the reader will register the recipient's debt.

Timeline: how the evidence accumulates across a passage

Inference strength grows as the passage progresses, and item writers exploit that order. Early signals are ambiguous; later signals resolve them. The time course below shows how a reader should weight evidence rather than counting mentions.

The key threshold is the turning point. Before acceptance, all four options remain live. After acceptance plus one positive follow-up, gratitude dominates. A single later negative remark can reopen ambiguity, but it must be strong enough to outweigh the earlier pattern—usually it is not.

Edge cases that flip the default answer

Not every passage resolves toward gratitude, and a responsible explanation names the exceptions. If Goodman's help carries visible strings—humiliation, control, or a debt Steve cannot repay—the text may support resentment instead, and option c becomes defensible. If Goodman is threatening, and Steve's compliance reads as appeasement, option d can win. The inference question tests whether you can tell benefaction from coercion.

A second edge case involves indirect narration. If Steve's feelings are reported secondhand by another character, the item often expects you to distrust the report and rely on Steve's own behavior. A third involves passages where Steve is the narrator and his self-description conflicts with his actions; item writers usually treat behavior as the stronger signal.

Red flag: an answer option that restates a single adjective from the text without behavioral support. Red flag: choosing fear because tension exists anywhere in the passage. Red flag: choosing protective because Steve and Goodman are allies; alliance is not protection. Red flag: choosing anger because of one sarcastic line with no follow-up. Red flag: ignoring the final paragraph, where resolution clues usually sit.

These warnings matter because distractors are engineered to feel plausible. The tension red flag is the most common trap: a tense scene is not the same as a frightened character.

Practical actions for verifying the inference

Apply a repeatable procedure rather than reading for vibes. First, locate every sentence in which Steve acts toward Goodman, and mark whether the action is approach or avoidance. Second, locate every sentence in which Goodman acts toward Steve, and mark whether the act is costly or trivial. Third, check whether Steve's later behavior matches the direction of Goodman's cost. Fourth, eliminate any option contradicted by a behavioral line. Fifth, choose the survivor that needs the fewest unstated assumptions.

A concrete threshold helps. One behavioral data point is not a pattern. Two aligned data points—acceptance plus thanks, or thanks plus defense—meet the standard for "best" inference. Zero aligned points means you should re-scan for indirect clues such as tone, repetition, or scene placement. If two options each have one supporting line, prefer the one whose line describes an action rather than a feeling.

For classroom or test-prep use, have students write the eliminated option and the exact line that eliminated it. This converts a guess into an auditable argument and exposes the common error of eliminating on intuition. For passage authors, place the decisive clue after the midpoint; readers weight late evidence more heavily, and a late clue reduces ambiguity.

Misconceptions and the limits of inference

The chief misconception is that inference means opinion. It does not: an inference must be entailed by evidence, and the word "best" signals comparison, not certainty. A second misconception is that strong emotion words in the passage settle the question; they often describe a moment, not a relationship. A third is that the correct answer must be stated somewhere in paraphrase; usually it is not stated at all, which is the point of the item.

A professional caveat: without the actual passage, no one can certify the answer. The reasoning above gives the default and the override conditions, but the text governs. Comprehension items drawn from standardized tests are also copyrighted, which is why public explanations describe method rather than reproduce passages.

The limits are real. Inference questions measure a reader's model of the text, not Steve's actual psychology, since Steve is fictional. Evidence within a short passage is thin by design, so the "best" inference is the one requiring the least speculation. When two options survive every check, the passage is underdetermined, and the fair move is to identify both and name the missing evidence.

The scientific study of reading supports this caution. Comprehension depends on prior knowledge and working memory, so two careful readers can weight clues differently. Instruction in inference—asking why a character acted—reliably improves comprehension in school-age readers, according to the Institute of Education Sciences practice guide on improving reading comprehension. That finding supports teaching the method above rather than memorizing answers. For most passages matching this question, the evidence chain runs from Goodman's costly help to Steve's accepting, thanking, and reciprocating behavior. That chain supports one option: Steve feels thankful for Goodman.