Non-Contiguous Cartogram
A map where each region keeps its original shape but is scaled by a data variable and detached from its neighbours — making size comparisons easy without distorting familiar geography.
// 01 — The chart
What it looks like
A non-contiguous cartogram of European GDP. Each country keeps its recognisable shape but is scaled and separated to reflect economic output.
// 02 — Definition
What is a non-contiguous cartogram?
A non-contiguous cartogram is a thematic map in which each geographic region is resized according to a data variable — such as population or GDP — and then detached from its neighbours. Unlike a contiguous cartogram, regions no longer share borders; they float in approximate geographic positions with gaps between them.
This approach solves the biggest problem with contiguous cartograms: distortion. Because regions are allowed to separate, each one can retain its original, recognisable outline while still encoding the data variable through its size. Readers can identify countries or states by shape and simultaneously compare their scaled areas.
The trade-off is that spatial relationships — which regions border which — are lost. But when the primary goal is magnitude comparison rather than adjacency analysis, this is a worthwhile exchange.
Key insight: Non-contiguous cartograms were popularised by Judy Olson in 1976 as a way to preserve recognisable shapes while still encoding quantitative data through area, offering a middle ground between standard choropleth maps and fully distorted contiguous cartograms.
// 03 — Anatomy
Parts of a non-contiguous cartogram
// 04 — Usage
When to use it — and when not to
- You want to compare magnitudes across regions while preserving recognisable shapes
- Your audience is geographically literate and can identify regions by outline alone
- Shape distortion of contiguous cartograms is confusing your viewers
- The data variable is strictly positive (population, GDP, area) and best encoded as area
- You need a visually striking editorial or infographic map
- Adjacency between regions is not important to the story
- Spatial adjacency or border relationships are central to your analysis
- You have too many tiny regions that become unreadable when shrunk
- Your data includes zero or negative values that cannot be encoded as area
- Precise quantitative comparison is needed — area perception is imprecise
- A simple choropleth or bar chart would communicate the same insight more clearly
- Your audience is unfamiliar with the geographic region and cannot recognise shapes
// 05 — Reading guide
How to read a non-contiguous cartogram
Follow these steps to interpret any non-contiguous cartogram you encounter.
Identify the variable encoded by size
Read the title, legend, or annotation to understand what quantity drives the scaling. Is it population, GDP, electoral votes, or something else? Without knowing the variable, sizes are meaningless.
Orient yourself geographically
Find a few familiar regions — large countries, your home state, or well-known shapes. This anchors your reading and lets you navigate the map even though regions are detached.
Compare relative sizes
Look for which regions are visually largest and smallest. Because shapes are preserved, your brain can quickly rank them. Remember that area perception is less precise than length, so focus on rough ordering rather than exact ratios.
Note the gaps
The white space between regions is a feature, not a bug. It prevents overlapping and makes each shape independently readable. If two regions that are normally adjacent are now far apart, it means their values differ significantly.
Look for colour encoding
Some non-contiguous cartograms combine scaling with colour to encode a second variable (a bivariate approach). Check if colour represents a rate, category, or change over time.
// 06 — Common mistakes
Mistakes to watch out for
Overlapping regions
If scaling is not handled carefully, enlarged regions can overlap smaller neighbours. Non-contiguous cartograms must guarantee separation; overlapping shapes make it impossible to read individual regions and defeat the purpose of the technique.
No legend or scale reference
Without a reference shape showing what a specific data value looks like, readers cannot translate visual area into numbers. Always provide a legend or annotate key regions with their values.
Using area for rates or percentages
Rates (like per-capita income) do not have an inherent area interpretation. Scaling a region by a rate conflates the rate with the region’s geographic size. Use area only for absolute quantities or provide very clear labelling.
Drifting too far from geographic positions
If regions are moved significantly from their true positions to avoid overlaps, the map loses its geographic context. Readers can no longer use spatial memory to locate regions, which removes the primary advantage over abstract charts.
Too many small regions
When a map has dozens of tiny regions (e.g. small island nations), scaling them down further makes them invisible. Consider grouping small regions or switching to a Dorling or Demers cartogram where each region gets a minimum-size symbol.
// 07 — Real-world examples
Where you’ll see non-contiguous cartograms used
Journalism: Population vs. electoral power
A news outlet scales U.S. states by population to reveal how some small-area but high-population states (like New Jersey) visually dominate, while large-area but low-population states (like Montana) shrink. The detached shapes keep each state recognisable.
Data JournalismPublic health: COVID-19 cases by country
An epidemiological dashboard scales European countries by total confirmed cases. The non-contiguous layout prevents France from visually absorbing Monaco and Luxembourg, giving each country its own readable space.
Public HealthEconomics: Export value by nation
A trade report scales countries by merchandise export value. Seeing Germany dwarf its neighbours in physical size immediately communicates the scale of its export economy, while shapes remain familiar.
Economics// 08 — At a glance
Quick reference
// 09 — Variations
Cartogram variations
Non-contiguous cartograms belong to a broader family of cartogram techniques, each with different trade-offs.
Contiguous cartogram
Regions are distorted to encode data but maintain shared borders. Preserves topology at the cost of shape recognition.
Dorling cartogram
Replaces shapes with proportional circles. Sacrifices geographic shape entirely but makes comparison extremely easy.
Demers cartogram
Uses proportional squares instead of circles. Squares pack more tightly and feel more “map-like” than circles.
Tile grid map
Equal-sized tiles in a grid layout. Every region gets the same visual weight, ideal for showing rates or categories.
// 10 — FAQs
Frequently asked questions
What is a non-contiguous cartogram?+
A non-contiguous cartogram is a thematic map in which each geographic region is resized according to a data variable — such as population or GDP — and then detached from its neighbours. Unlike a contiguous cartogram, regions no longer share borders; they float in approximate geographic positions with gaps between them.
When should you use a non-contiguous cartogram?+
Use a non-contiguous cartogram when you want to compare magnitudes across regions while preserving recognisable shapes. It also works well when your audience is geographically literate and can identify regions by outline alone, and when shape distortion of contiguous cartograms is confusing your viewers.
When should you avoid a non-contiguous cartogram?+
Avoid a non-contiguous cartogram when spatial adjacency or border relationships are central to your analysis. It is also a poor fit when you have too many tiny regions that become unreadable when shrunk, or when your data includes zero or negative values that cannot be encoded as area.
Is a non-contiguous cartogram suitable for dashboards?+
Yes — a non-contiguous cartogram can work well in dashboards as long as the panel is large enough for readers to perceive the encoded values, has a clear title, and includes the legend or axis labels needed to interpret it.
What category of chart is a non-contiguous cartogram?+
Non-Contiguous Cartogram belongs to the Geospatial family of charts. Charts in that family are designed to answer the same kind of question, so they often work as alternatives when one doesn't quite fit your data.
How do you read a non-contiguous cartogram?+
Start with the axis labels and legend, then look at the overall shape before zooming into individual marks. Compare prominent features against the rest of the data, and verify any conclusion against the underlying numbers when precision matters.