The Handbill: Three Arguments Before Friday

Argument 1 - undistributed middle (with a touch of affirming the consequent). The term 'users of the old question banks' is never shown to apply only to toppers. Toppers are a subset of question-bank users, and Anita may sit in the much larger remainder. The conclusion only follows if all question-bank users become toppers, which the premise does not say.

Argument 2 - post hoc ergo propter hoc, with a suppressed confounder. Two events in sequence are treated as cause and effect. The register shows heavy rain on Tuesday, which independently explains a fall from 96 to 61; the price list shows the rise was known three weeks earlier, so Tuesday's announcement cannot be the trigger. One before-after pair with no control group cannot isolate a cause.

Argument 3 - false dilemma (false dichotomy). 'Cancel the contractor' and 'results keep falling' are presented as the only options, ignoring a third possibility such as renegotiating rates, and the two are not even causally linked.

A good answer notes the common thread: arguments 1 and 3 fail formally (bad form or a missing disjunct), argument 2 fails materially (unexamined evidence and an ignored rival cause).

Standard form

1. All toppers in the last three years used the old question banks.
2. Anita uses the old question banks.
3. Therefore Anita will top this year.

Form: All T are Q; a is Q; therefore a is T. Invalid - the middle term Q is not distributed in either premise.

Counterexample: All toppers wore spectacles; Anita wears spectacles; therefore Anita will top. The premises are true, the conclusion plainly does not follow, so the form is bad.

Repair: Add 'All users of the old question banks become toppers' (All Q are T). The argument then becomes valid by the first figure: All Q are T; a is Q; so a is T. But the added premise is false on the handbill's own evidence - it claims only that toppers used the banks, not that users become toppers. A weaker repair, 'Most Q are T', yields only a probabilistic conclusion, which the handbill's confident 'will top' overstates.

Lesson: validity can be bought with a premise, but not truth. The class should ask which premise is doing the work and whether the college's own records support it.

The present evidence is a single before-after comparison with no control. To support causation the class would need:

1. A comparison group. Attendance in lectures where the canteen plays no role, or attendance at the same lectures on rainy days in earlier weeks, to see whether 96→61 is unusual.
2. Control for confounders. Rain, examinations, a festival, a changed timetable - each must be ruled out or held constant.
3. Dose-response or repetition. Do further price rises produce further falls, and does a price cut restore attendance?
4. A mechanism. A stated reason why higher tea prices would keep students away from lectures (e.g. students skipping the 8 a.m. slot to buy tea later), not merely a plausible story.

Correlation is insufficient because a third variable can drive both, and because a single pair of days gives no way to separate the candidate cause from weather, season or coincidence. The council's phrase 'the data proves it' confuses a coincidence with evidence. A defensible claim would be: 'Attendance fell on a rainy day after a price rise was announced; the cause is undetermined.' That is weaker rhetoric and better reasoning - which is exactly the distinction the Logic paper tests.