The Average Isn't Always in the Middle Connecting the Dots...

Guessing whether someone is more likely to be male or female purely from their height (5 feet versus 6 feet) is the kind of question that can only be settled by actually gathering data, not by fixed fact alone -- this is exactly what separates a statistical question from an ordinary one. A statistical statement makes a numerical or proportional claim about the world (cricket form, daily screen time, population change), and answering it properly always requires collecting real data first.

Two groups of friends collecting guavas, one group of 5 collecting 3, 8, 10, 5, and 4 guavas (30 total), the other group of 6 collecting 5, 4, 6, 3, 4, and 8 (also 30 total), reveal the real meaning of the arithmetic mean: it's simply what everyone would get if the total were redistributed perfectly fairly -- 6 guavas each for the first group, 5 each for the second. Ancient Indian mathematicians (Brahmagupta, Mahaviracharya, Sripati, Bhaskaracharya, Ganesha) each had their own Sanskrit term for this exact 'fair share' idea, showing the concept has a genuinely long mathematical history.

Comparing two families' heights reveals something the mean alone can hide: Yaangba's 6-member family has a mean of 164.3 and a median of 164.5, nearly matching, with no unusual member -- but Poovizhi's 5-member family, which includes one much younger, much shorter child, has a mean of only 160.2 while the median sits noticeably higher at 170, because that one low outlier drags the mean down without moving the median at all (since the median only cares about the MIDDLE value's position, not how far away the extreme values sit). The general rule: mean equals median when data is balanced; mean is lower than median when a low outlier is present; mean is higher than median when a high outlier is present.

Analysing a Grade 5 class's heights (17 boys, 11 girls) using dot plots reveals a genuinely counter-intuitive fact: the single tallest student and the single shortest student in the whole class are BOTH boys, and yet the girls' average height (146.9cm) comes out higher than the boys' average (142.94cm) -- because averages summarise an entire group, and a group's extremes (its single tallest or shortest member) don't have to match which group has the higher typical value overall.

A double bar graph places two related categories' bars side by side for direct comparison -- worked through real onion-price data across two towns, real worldwide rocket-launch counts by organisation, and real daylight-hours-per-month data contrasting Helsinki (near-24-hour summer daylight) against Wellington in the opposite hemisphere (troughing in the same month Helsinki peaks). A genuinely striking real dataset closes the chapter: India's average height at every age from 5 to 19, tracked across four census years (1989, 1999, 2009, 2019), showing measurable real growth in the population's average height over three decades -- while explicitly warning against over-generalising from any single school's data to the whole country.