Reading Precisely

A sentence can be almost right and still make a different claim. Before judging an answer, identify its scope: whom or what it concerns, how many, where, when, and under what conditions. 'Most seedlings in the shaded greenhouse survived the first week' concerns a particular group, setting, and interval. It does not establish what happens to seedlings generally.

Some means at least one; by itself, it does not mean 'some but not all.' Most means more than half of the relevant group. All allows no exceptions within that group. Words such as 'many' and 'several' do not supply an exact percentage. Preserve these distinctions rather than translating every positive quantity into 'usually.'

A weaker answer may still follow: if 18 of 30 participants succeeded, 'some participants succeeded' is true. But cautious wording cannot repair a change in population. 'Some people who did not participate succeeded' remains unsupported. Check the whole claim, not just its quantity word.

• Mark the group: participants, surveyed residents, or all residents?
• Mark the quantity: at least one, more than half, or every member?
• Keep restrictions attached: in this trial, during winter, or under low light.

Writers distinguish what could happen, what probably happens, and what does happen. 'The marks may indicate tool use' offers a possibility. 'The marks probably indicate tool use' judges that explanation likely. Neither establishes it as certain. Similarly, 'tends to' describes a pattern that can have exceptions; it does not mean 'always.'

Do not choose the most cautious answer automatically. It can still be irrelevant or reverse a relationship. If all five tested samples contain copper, an answer saying 'all five' is appropriate. Strong words are wrong only when the evidence does not justify their strength.

Watch the verb as well as the qualifier. 'Is consistent with,' 'supports,' and 'proves' make different commitments. Evidence can support an explanation while leaving competing explanations open. Ask: is the answer reporting a fact, suggesting a possibility, or announcing a settled conclusion?

Locate exactly what a negative word cancels. Not all means at least one exception; it does not mean none. 'Not all seedlings survived' allows some to survive or none to survive. None means zero. 'The treatment did not always work' rejects perfect consistency, not every instance of success.

Separate a statement about reality from a statement about knowledge. 'The team did not establish that the treatment worked' does not mean 'the team established that the treatment failed.' The first leaves the result unresolved. Likewise, no surviving written record of an event is not automatically proof that the event never happened.

Missing evidence can matter when evidence would be expected. If a reliable detector should register a substance above a stated concentration, a negative result supports a conclusion about that concentration. Read the method's sensitivity and coverage before interpreting silence.

Academic negatives sometimes express limited support. Cannot rule out X means X remains possible, not that X is established. Not inconsistent with X means compatible with X, not proof of X. For example, marks that are not inconsistent with tool use could still have another explanation.

A necessary condition is required for an outcome. A sufficient condition guarantees the outcome under the stated rule. 'An application is accepted only if it is complete' makes completeness necessary: acceptance guarantees completeness. It does not guarantee that every complete application is accepted.

Compare 'If an application is complete, it is accepted.' This second rule makes completeness sufficient for acceptance. 'Only if' and 'if' point in opposite directions. 'Only members may borrow equipment' means anyone allowed to borrow must be a member; it does not say every member is allowed to borrow.

For an only-if rule, translate the necessary condition into a plain sentence: 'To get the outcome, this condition must hold.' Then test the tempting reversal. Could the condition hold without the outcome? If the passage allows that, the reversed answer is unsupported. A missing required condition rules the outcome out; merely having the condition does not guarantee it.

For every comparison, identify what is measured, which groups are compared, and the reference point. 'Warmer than last year' does not mean warm in absolute terms. 'The largest increase' does not mean the highest final value. A group can improve faster and still finish behind another group.

Keep counts, percentages, and percentage points separate. A rise from 20% to 30% is 10 percentage points; relative to the original 20%, it is a 50% increase. Percentage change uses the starting value as the baseline: change divided by starting value. You need only simple arithmetic here; the reading task is recognizing which comparison the sentence makes.

An increasing number does not automatically mean an increasing share. A library can add ebooks while ebooks become a smaller percentage of a faster-growing collection. An average also does not establish every individual's result. Read labels and units before deciding what a number supports.

When two things occur together, several explanations may fit. One may cause the other; causation may run in the reverse direction; a third factor may influence both. Students who attend more tutoring sessions might earn higher grades because tutoring helps, because motivated students do both, or for several reasons together.

Do not replace careful reading with the slogan 'correlation never supports causation.' An association can be relevant evidence for a causal explanation without proving it. Random assignment and a suitable comparison group can strengthen a causal inference because they help separate a treatment's effect from preexisting differences. Still preserve what was measured and the conditions of the study.

Check three dimensions together: direction (which thing affects which), strength (possible, likely, or established), and scope (which population, period, and conditions). A result after two weeks does

not establish a permanent effect. Evidence from one species does not by itself establish the same effect in every species.

Keep these checks in mind

• Match direction, strength, and scope before deciding that an answer follows.
• Track exactly what quantity, negative, or condition applies to which group.
• Separate totals from changes, percentages from percentage points, and averages from individual results.
• Preserve uncertainty where it exists, and accept strong conclusions when the evidence actually supports them.

Begin with four short drills, then complete twelve multiple-choice questions. Difficulty increases within the set. For every multiple-choice answer, identify a decisive clue and explain why the closest alternative fails.