What if a map stopped giving land area the final word about how large a place should look?
On an ordinary map, a vast sparsely populated territory can dominate the page while a compact city containing millions of people appears tiny. For some questions, that physical-area emphasis is useful. For others, it can hide the quantity we actually care about.
A cartogram deliberately resizes geographic areas according to a chosen variable such as population, votes, economic output, disease burden or emissions.
A cartogram asks geography to surrender physical area so another quantity can become the map’s geometry.
Quick Read: The Cartogram Mechanism
GEOGRAPHIC AREAS + NUMERICAL VARIABLE → RESIZING RULE → DISTORTED GEOMETRY → AREA REPRESENTS DATA MAGNITUDE
1. Ordinary Maps Give Area to Land
On a conventional equal-area map, the size of a region reflects its physical land area. That is appropriate when land extent matters. It can be misleading when the reader’s real question concerns people, votes or economic activity.
2. Cartograms Give Area to a Variable
If population drives the cartogram, a region with twice the population should occupy roughly twice the visual area under an ideal area-proportional design. Geographic size becomes quantitative symbolism.
3. Contiguous Cartograms Preserve Neighbour Relationships
Contiguous cartograms distort polygon shapes while trying to keep neighbouring areas touching. Familiar geography bends and stretches, but broad topology survives.
4. Noncontiguous Cartograms Preserve Shape Better
Noncontiguous cartograms can shrink or enlarge regions independently, leaving gaps between them. Shape may remain more recognisable while adjacency is visually weakened.
5. Dorling Cartograms Replace Areas With Symbols
Dorling cartograms commonly replace regions with circles sized by the chosen variable and rearranged to reduce overlap while retaining approximate geographic neighbourhood. They sacrifice original polygon shape almost entirely.
6. Cartograms Are Not Map Projections
Map Projections own distortion created when curved Earth is transformed to a flat plane. Cartograms deliberately distort geographic geometry according to a data variable after or within a mapped coordinate representation.
7. Cartograms Are Not Proportional Symbol Maps
Proportional Symbol Mapping keeps the underlying geography intact and places variable-sized symbols on top. Cartograms change the geography itself—or replace regions with resized forms.
8. Cartograms Are Not Choropleths
Choropleth Classification keeps polygon area fixed and changes colour according to value. Cartograms encode magnitude through area.
9. Topology Can Be More Important Than Shape
A contiguous cartogram may distort a country severely yet remain readable if readers can still recognise which regions neighbour one another. This connects directly to Spatial Topology.
10. Recognition Is a Design Constraint
If distortion becomes too severe, readers may no longer recognise places. A technically accurate area transformation can therefore fail as communication if geographic identity disappears.
11. Small Places Can Become Visible
A physically small region with a large population or economic role can expand dramatically on a cartogram. This is one of the method’s strengths: quantities hidden by land area can become visually dominant.
12. Large Empty Areas Can Shrink
Sparsely populated deserts, forests or tundra can contract on population cartograms. The map stops asking how much land exists and begins asking how many people occupy each place.
13. One Variable Controls the Geometry
A population cartogram and GDP cartogram of the same world can look very different. The geometry is conditional on the selected variable, so the title and legend must make that variable unmistakable.
14. Colour Can Carry a Second Variable
A cartogram can resize areas by population while colouring them by income, vote share or disease rate. This creates a powerful bivariate display but also increases cognitive load and the risk of confusing area with colour.
15. Primary Geography: Make the Classroom Map by Number of Pupils
Imagine four groups occupying equal floor space but containing 2, 4, 8 and 16 pupils. Ask children to redraw each group’s area according to the number of pupils. They discover that maps can size space by something other than metres.
16. Secondary Geography: Compare Land Area and Population
Students can compare a conventional world map with a population cartogram. Which countries expand? Which contract? The contrast reveals how strongly ordinary maps privilege physical land area.
17. Advanced Geography: Area Error and Shape Preservation
Cartogram algorithms balance competing goals: make area proportional to data, preserve recognisable shape, maintain adjacency and avoid severe deformation. No solution optimises every objective perfectly.
18. Singapore Example: Population
A cartogram of Singapore planning areas resized by resident population would expand heavily populated residential areas and shrink low-population areas such as large industrial, water or protected-land zones. The result would answer a different question from a conventional island map: where does the population weight of Singapore actually sit?
19. Singapore Example: Employment
Resize areas by employment instead and commercial and industrial centres could expand relative to primarily residential zones. Comparing the two cartograms would expose the geography between where people live and where economic activity concentrates.
20. Election Example
Election maps coloured by winner can visually exaggerate large low-population territories. Resizing areas by electorate or votes can make the map better reflect the number of people represented by each region.
21. Public Health Example
A cartogram resized by population can help readers interpret disease counts more sensibly than a conventional land-area map, while colour can represent disease rate. The two variables must remain clearly separated.
22. Economic Geography Example
GDP cartograms make major economic centres expand and low-output territories shrink. They can reveal economic weight dramatically, but GDP area should not be confused with population, welfare or physical territory.
23. Hostile Test: “That Country Is Huge on the Map”
Huge in what? Land area, population, GDP, votes or emissions? On a cartogram, visual area is a data variable. Reading the map without identifying that variable reverses the entire logic of the representation.
24. Where Cartogram Reasoning Breaks
- Area-variable blindness: forgetting what quantity controls polygon size.
- Shape literalism: treating distorted outlines as physical geography.
- Recognition failure: distorting places until readers cannot identify them.
- Projection confusion: treating data-driven distortion as ordinary map-projection distortion.
- Area-colour collapse: confusing the variable encoded by size with the variable encoded by colour.
- Single-variable absolutism: treating one cartogram as the definitive geography rather than one analytical lens.
25. Ten Questions for Cartograms
- What variable controls area?
- Is area approximately proportional to that variable?
- What cartogram type is used?
- Are neighbouring relationships preserved?
- How much shape distortion occurs?
- Can readers still recognise places?
- Is colour encoding a second variable?
- What does the conventional map hide that the cartogram reveals?
- What geographic information is lost through resizing?
- Would a proportional-symbol map communicate the same quantity more clearly?
26. Where This Fits
Map Projections own curved-to-flat transformation. Proportional Symbol Mapping owns magnitude through symbol size while retaining base geography. Choropleth Classification owns colour classes over fixed areas. This article owns deliberate resizing of geographic areas according to a numerical variable.
The Idea to Keep
A cartogram changes the size of places so we can see the size of something else.