Geographical Skills: Maps, Graphs and Statistical Techniques
What this domain tests
Geographical Skills tests practical competence: map scale and grid references, cross-sections and gradients, selecting and reading graphs, calculating descriptive statistics, and applying a rank-correlation test with a significance decision.
Core rules and concepts
- Scale: 1:25,000 means 1 cm represents 250 m; 1:50,000 means 1 cm represents 500 m. Convert units before measuring or comparing.
- Grid references: eastings (vertical grid lines) first, read left to right; then northings, read bottom to top. Six-figure references estimate tenths of a square.
- Gradient = vertical interval ÷ horizontal distance, in the same units; give it as a ratio, percentage or angle.
- Choose graphs by data type: pie for parts of a whole, bar for categories, line for continuous change, scatter for two variables, logarithmic for very wide ranges.
- Descriptive statistics: the mean is distorted by outliers; the median suits skewed data; standard deviation and interquartile range describe spread.
- Spearman's rank: rs = 1 − 6Σd²/(n³ − n). Interpret the sign and strength, then compare with the critical value for that n and significance level.
Worked example
Eight sites are ranked for distance from a road and for noise level. Σd² = 42, n = 8. Then 6Σd² = 252 and n³ − n = 512 − 8 = 504, so rs = 1 − (252 ÷ 504) = 1 − 0.5 = +0.5. This is a moderate positive relationship, but the 5% critical value at n = 8 is around 0.64–0.74, so the result is not statistically significant and the null hypothesis is retained.
Traps that cost marks
Reversing eastings and northings; mixing metres and kilometres in gradient calculations; squaring ranks rather than rank differences; claiming causation from correlation; reporting a correlation without checking significance.