The Cropped Ticker

Do not adjudicate. Convert the argument into an investigation.

1. Name the disagreement, not the culprit. 'We have a claim and three opinions and no evidence yet. What would we need to know?' Take answers and write them on the board.
2. The list that emerges should be: the date, who published it, what the percentage measures (year-on-year or month-on-month; headline or food-only), and whether 'lowest in 50 years' is a claim anyone has checked.
3. Let the crop do the teaching. Hold up the screenshot and ask: what has been removed from this picture? The missing year is the whole question. A ticker with no year cannot be dated, and an undated number cannot be current.
4. Set two tasks for next session - you do the checking, pupils shadow it: one pupil finds the official release, one finds the 2015 report.
5. Close honestly: 'I don't know yet whether this is true. I know how we will find out, and we will finish this next week.'

Avoid: declaring it fake to end the argument (you would be right for the wrong reason - you had not checked either); mocking the pupil who shared it, which guarantees nobody brings you a claim again; and dropping the thread. An unresolved claim plus silence reads as endorsement.

Routine: the four-question stamp. Every claim gets the same four questions, written on the board and applied to this claim:

1. When? The 1.3% figure is real - it exists in a 2015 report. The number is genuine and ten years old. 'Real but stale' is a category pupils do not have, and it is the category this forward belongs to.
2. Who? Official statistics bureau release vs television ticker vs WhatsApp forward. Rank them by proximity to the measurement, not by how confident they sound.
3. What exactly is measured? 3.9% year-on-year CPI is the current figure. A pupil must be able to say what the percentage is a percentage of before repeating it.
4. Does the claim need a series? 'Lowest in 50 years' is a superlative: it asserts a comparison across 50 years of data. Neither the ticker nor the forward contains that series, so the superlative is unsupported even if the 1.3% were current. Teach the superlative test: any claim containing 'lowest', 'highest', 'first' or 'record' needs a series behind it.

Close with a 3-minute transfer: give two fresh forwards (one true, one stale) and ask for the four-question stamp in writing. The true one is essential - if every example is false, pupils learn 'forwards are lies' instead of 'forwards are unevidenced'.

Rotating verification duty, graded by rubric. Pairs take one forward per fortnight - 19 pairs, so each pupil is assessed about once a month - and present a two-minute verification. Grade on four criteria, one mark each: identified the claim precisely; traced a date; located a primary or near-primary source; stated a conclusion with a stated uncertainty. The fourth criterion is the discriminating one.

Keep a class log (date, claim, pair, four marks, one comment). Over a term this gives you 38 individual profiles without a single sit-down test, and the log itself becomes evidence for the head.

What a weak answer looks like:

• 'It's fake because it's on WhatsApp' - correct conclusion, no method. This is the failure mode the exercise exists to prevent; the pupil has learned to distrust a channel, not to check a claim.
• 'It's true, I saw it on TV' - same error in the opposite direction: source authority substituted for verification.
• A confident conclusion with no date traced. Mark it down even when the conclusion happens to be right, and say why in the comment: right answers with no method do not transfer to the next claim.