Arguments, assumptions and evidence

Identify what needs support

An argument offers a reason for accepting a claim. The claim being defended is its conclusion; the supporting statements are its premises. A passage can contain several facts without making an argument at all. “The office opens at nine and closes at five” simply reports a schedule. “The office should open earlier because many visitors arrive before nine” uses an observation to support a recommendation. Before judging either statement, ask what the writer wants you to accept on the basis of something else.

Words such as “therefore”, “so” and “because” can reveal the relationship, but do not let their position replace reading. A conclusion can come first, last or between its reasons. Restate it in a short sentence without changing its strength. “The service may reduce waiting” is a possibility claim; “the service will eliminate waiting” is a much stronger prediction. Evidence adequate for one may be inadequate for the other. Keep recommendations distinct from observations too: what happened and what should be done require different kinds of support.

Find the missing bridge

Suppose a shop owner says that extending opening hours will increase profit because more people will be able to visit. The observation concerns possible visitors; the conclusion concerns money remaining after costs. The reasoning needs a bridge between them. Extra visits might produce too little revenue to cover extra staffing and electricity. An assumption is a proposition taken for granted in making that move. It need not be stated, and it need not be something the author consciously considered.

For a necessary-assumption question, deny the proposed assumption and ask whether the argument can still work on its own terms. If the plan depends on extra sales covering its extra costs, the denial removes that financial bridge. By contrast, “all visitors will buy something” is stronger than necessary: some visitors could buy nothing while enough others buy to make the plan profitable. Necessary does not mean helpful, and it does not mean the most ambitious version of an unstated claim. Test how much the reasoning actually needs.

Strengthen a link rather than a topic

New evidence strengthens an argument when it makes the conclusion better supported by its premises. It need not prove the conclusion. If a centre predicts that an earlier opening time will bring more visitors, a survey showing many nearby workers can attend only before their shifts supports the proposed link. A fact about the centre's attractive paint colour concerns the same place but may do nothing for that inference. Topic overlap is weaker than logical relevance.

Likewise, weakening evidence makes the inferential step less secure without needing to establish the opposite conclusion. If the survey counted expressions of interest but most respondents cannot reach the centre before work, the prediction loses support. That does not prove attendance will fall. Read the requested task carefully: “weakens” asks for a direction of change in support, whereas “disproves” would demand more. Avoid rejecting relevant evidence merely because it leaves some uncertainty.

Check comparisons and causal claims

A common argument observes that one group has a better outcome and attributes the difference to one feature. Suppose stores using a booking system have shorter queues. Perhaps the system helps. But perhaps those stores also have more staff or attract fewer customers. A useful comparison keeps competing explanations as similar as possible. Evidence that staffing, customer volume and measurement conditions were comparable strengthens the claim about the booking system. Evidence that one group had far fewer visitors weakens that particular causal inference.

Time order alone is also insufficient. A fall in complaints after a new policy is compatible with the policy helping, but another change might have occurred at the same time. Ask what would have happened without the policy. A comparison with otherwise similar sites that did not adopt it can be useful. Do not insist that every short argument must settle every imaginable alternative; focus on the evidence actually offered and the specific gap named in the answer choices.

Evaluate before you choose

An evaluation question asks which missing information would help assess an argument. Try giving the proposed question opposite answers. If one answer would strengthen the reasoning and the other would weaken it, the issue is diagnostically useful. For a claim that a cheaper supplier will reduce total costs, ask whether delivery and defect costs remain comparable. A low purchase price may be outweighed by those other costs. Asking whether the supplier has a long name is unlikely to change the inference in either direction.

Finally, explain your selection using a complete chain: the new fact changes this premise or missing link, which changes support for this conclusion. If you can say only that an option is “relevant”, return to the gap. A precise account helps distinguish a targeted answer from a broadly plausible statement. It also keeps your own views about the policy, product or institution from substituting for the reasoning presented.