Take two or three hands-up answers, not open call-outs. Give five seconds of quiet think-time before any hands go up. Pupils will say the bar doesn't fit, the chart is too small, we need a bigger grid. Revoice toward the lesson question: so the squares aren't worth enough — what could each square be worth instead of 1? Do not name yStep yet; keep this short.

Walk each example aloud, one at a time. The class watches; nobody is at the board yet.

• Same data, yStep 1 — point at the Walk bar climbing all the way to 12; say 'the chart works, but it is tall and cluttered for what we need'.
• Same data, yStep 2 — point at the same Walk bar now reaching square 6 instead of 12; say 'same number, less paper'.
• Same data, yStep 5 — pause on the Cycle (5) and Car (3) bars; ask 'can you tell those two apart?'. The class will say no — that is the point. yStep 5 was too coarse for this data.
• Big counts, yStep 5 — now the 30 fits and the small bars (4, 6, 8) are still distinguishable. This is the lesson's punchline: the dominant bar decides the scale.

Do not move on until the class can say in their own words 'we pick the yStep that fits the biggest bar without flattening the small ones'.

This round is for talking it through together — pupils take turns at the board and the class agrees or corrects out loud.

Start at the bar-chart-builder with the data already loaded (Walk 12, Cycle 5, Bus 8, Car 3). Run through the yStep settings in turn — 1, 2, 5, 10. For each, ask the class first 'predict — does this scale work for this data?' before a pupil at the board makes the switch. Listen for pupils naming yStep 2 as the sweet spot for these numbers, and yStep 10 as obviously too coarse (every bar lands on the same square). Revoice 'so the smallest difference we want to see decides how small our step has to be'.

If time allows, swap to a second data set (Soccer 20, GAA 15, Tennis 10, Other 5) and ask 'now what changes? Does yStep 2 still work?' — the class should reach yStep 5 by reasoning aloud.

This round is the practice bank — pupils take turns at the board, check each answer, and the class confirms before moving on. Keep the board work brisk rather than over-explaining.

Pupils rotate at the IWB; the class predicts the readable yStep BEFORE the pupil at the board drags bars. The fourth core problem (Apples 30, Pears 4, Bananas 6, Grapes 8) is the dominant-bar case — pause there and ask 'why does this one feel different from the others?'.