Assessing Claims and Evidence

By this stage, you already know that a strong argument cites evidence rather than assertion alone. The harder and more consequential skill, especially on timed reading assessments, is deciding whether the evidence offered actually does the work a claim demands of it. A fact can be true, relevant, and still fail to prove the point it is attached to. This lesson sharpens that judgment: distinguishing evidence that merely supports a claim from evidence that sufficiently establishes it, and recognizing the reasoning traps test writers build into plausible-sounding wrong answers.

Every piece of evidence you evaluate must first pass a relevance test: does it actually bear on the claim being made? But relevance alone proves nothing. A single anecdote is relevant to a claim about a broad trend, yet one example cannot establish that the trend is real, common, or significant. Standardized passages exploit exactly this gap. A question might ask which piece of evidence "best supports" a claim, and several answer choices will be technically relevant while only one is strong enough in scope, source, and logic to actually carry the argument's weight.

Treat every claim-evidence pair as a mini-argument you must test, not accept. Ask what the claim would require to be true, then ask whether the cited evidence, taken alone, delivers that much.

When you assess whether evidence adequately proves a claim, apply four criteria simultaneously rather than checking them in isolation.

Relevance asks whether the evidence directly addresses the claim's specific terms, not a nearby or loosely related idea. Reliability asks whether the source is credible, current, and free of obvious bias or conflict of interest. Representativeness asks whether the evidence reflects a typical case or an outlier, and whether a sample size or study design allows for generalization. Sufficiency asks the cumulative question: even if the evidence is relevant, reliable, and representative, is there enough of it, and is it strong enough, to justify the claim as stated, including any qualifiers such as "always," "most," or "the primary cause"?

A claim's own wording often determines the sufficiency bar. A modest claim ("this factor can contribute to the outcome") requires far less evidence than a sweeping one ("this factor is the main cause of the outcome"). Reading the claim's precise language before judging the evidence is one of the most reliable habits for this kind of analysis, and it connects directly to the work you have already done in Evaluating Evidence And Reasoning.

Consider a passage claim: "The new study time policy improved student performance across the district." A test question offers this evidence: "At one school, average test scores rose four points after the policy began."

Apply the four criteria. The evidence is relevant, since it concerns performance after the policy. Its reliability is uncertain without knowing how the study was designed or who reported the scores. Its representativeness is weak, since one school cannot stand in for an entire district. Most decisively, it is insufficient: the claim asserts a district-wide effect, and the evidence describes a single, unreplicated, local change that could result from many other causes, including a stronger incoming class or a different test. A four-point rise at one site does not establish a causal, district-wide improvement.

Contrast that with evidence stating that scores rose across every school in the district over three consecutive years following the policy, with comparable results in a similar neighboring district that adopted the same policy later. This version satisfies representativeness (multiple schools), corroboration (an independent district shows the same pattern), and a stronger, though still not conclusive, case for sufficiency. Practicing this side-by-side comparison is the core method behind Analyzing Evidence For Claims.

Standardized reading sections repeatedly test the same handful of sufficiency failures, so learning to name them speeds up your analysis considerably. Hasty generalization draws a broad conclusion from too small or too narrow a sample. Cherry-picking presents only the evidence that favors a claim while omitting contradictory data that a careful reader would need to weigh. Correlation mistaken for causation treats two events that occurred together as though one necessarily caused the other, ignoring alternative explanations. Appeal to a biased or unqualified source cites someone whose expertise, incentives, or perspective undermine the evidence's reliability even though the statement itself sounds authoritative.

When you encounter an answer choice built on one of these traps, you can eliminate it even if the evidence it cites is superficially relevant to the claim, because relevance without sufficiency or reliability cannot prove the point at issue.

On a timed reading section, budget your attention deliberately: identify the claim's exact scope first, scan the offered evidence choices, and eliminate any that fail relevance outright before applying reliability, representativeness, and sufficiency to the remaining options. The correct answer is usually the choice that satisfies all four criteria at once, not merely the one that sounds most impressive or cites the largest number. This disciplined, elimination-based approach is the same strategy developed in Supporting Arguments With Evidence, applied here specifically to the pressure and pacing of standardized assessment.

Sufficiency: Whether the quantity and strength of evidence are enough to justify a claim as precisely worded, including its qualifiers.

Representativeness: The degree to which evidence reflects a typical case rather than an unusual or isolated outlier.

Corroboration: Independent confirmation of a finding through separate sources, studies, or contexts.

Hasty Generalization: Drawing a broad conclusion from a sample too small or narrow to support it.

Cherry-Picking: Selecting only evidence favorable to a claim while omitting contradictory information.

Correlation Versus Causation: The error of treating two co-occurring events as proof that one caused the other.

Work through short test-style passages where a claim is paired with two or three evidence options, and rank the options by sufficiency rather than by relevance alone. Practice explaining, in one sentence, exactly why the weaker option falls short, whether due to sample size, source bias, or a scope mismatch with the claim's wording.

Related lessons

• Evidence That Proves Claims