Coding and AI classes in the East of England

The learner reads each station's raw file, keeps the rain column, skips months marked missing, and averages each calendar month over 1991 to 2020. The year's total divided by twelve is what every month would get if rain fell evenly. Each month's gap from that even share is measured without its sign, the gaps are added, and the sum is divided by the year's total. The result, a seasonality index of the kind introduced by the geographers Walsh and Lawler in 1981, is 0 for perfectly even rain and about 1.83 if a whole year fell in one month.

The East is not just dry; its rain is also fairly even. Both East stations score under 0.16, while Camborne, far wetter, has a much stronger winter season, with November more than twice as wet as May. Within the East there is still a pattern: Cambridge's February to April gets under 19 per cent of the year's rain against an even 25, and at Lowestoft the last three months of the year bring about 30 per cent. Spring is the dry season in both places, which matters for crops that are growing fast just when rain is scarcest.

The telling slip is forgetting to ignore the sign. Months above the even share and months below it always cancel exactly, so without absolute values every station scores zero, and the code seems to prove that all rain is perfectly even. The learner writes a test with an invented year of all rain in one month, which must score 1.83. Our figures are averages from the station files, Lowestoft's record spans a small site move in 2007 that the file itself notes, and three Lowestoft months are missing, so the index is a careful estimate rather than an official statistic.

Checking a program against an answer you already know is a habit an East of England teenager builds by coding in 2026. The longer argument is in why teenagers should still learn to code in 2026.