Geography: Reading Space and Arguing with Documents
Geography in the European Baccalaureate tests your ability to read space as an argument: you must interpret maps, graphs, photographs and statistics, connect them to geographical concepts, and write a structured explanation at several scales. The paper rewards precise use of cartographic and statistical language and penalises description that stops at ‘what’ without reaching ‘why there and why so’.
Core rules and concepts
- Space, scale and territory: every document has a scale. Identify whether it is local, regional, national, EU-wide or global, and do not mix scales in one explanation.
- Cartographic grammar: read title, legend, projection, classification method (quantiles, equal intervals, natural breaks), source and date before interpreting the pattern. A choropleth shows rates, not absolute totals, unless the data are normalised.
- Spatial differentiation: use core–periphery, urban–rural, accessibility, demographic and economic gradients to describe inequality rather than simply listing regions.
- Actors and systems: link patterns to EU regional policy, globalisation, migration, transport and climate policy. Name the actor and the territory it affects.
- Sustainable development: treat environmental, economic and social dimensions as interacting systems, not as three separate lists.
Worked example
A choropleth of EU unemployment by NUTS 2 shows a north–south and east–west gradient. Step 1: check the legend - is the value a percentage of the labour force? Step 2: identify the classification; quantiles make every class equally common, so a dark region may not be an extreme outlier. Step 3: compare the map with a second document, such as GDP per head or access to services. Step 4: explain the pattern through deindustrialisation, low activity rates, commuting flows and cohesion funding. Step 5: conclude with scale - national averages can hide regional pockets of high unemployment.
Traps that cost marks
- Describing the map instead of explaining the process behind it.
- Treating a choropleth of totals as if it were a rate.
- Ignoring source and date, then comparing incompatible years.
- Confusing correlation with causation.
- Using outdated labels such as ‘LEDC/MEDC’ when the dossier uses EU, NUTS or another framework.