How Maps Mislead | Scale, Projections and Data

Did you know? A map can put every country in the right general location and still give a misleading impression of how large those countries are. A different map can show a perfectly accurate number of events while making one neighbourhood appear more dangerous simply because it has more residents.

That is how maps mislead. Maps are selective models of the world. They simplify information so that people can navigate, compare, plan or understand patterns. Every simplification involves choices about scale, projection, boundaries, labels, symbols, data and what remains unseen.

Quick answer: a map becomes misleading when readers mistake its symbols or projection for a literal picture of reality, interpret raw counts as comparable rates, use the wrong scale for a decision or overlook the date and limitations of its data. The repair is to ask what the map was designed to show and what it cannot show reliably.

The first misunderstanding: every flat map distorts something

Earth is roughly spherical; a map on paper or a screen is flat. Flattening a curved surface changes some geometric properties.

The United States Geological Survey’s guide to map projections explains that no flat world map can perfectly preserve all directions, distances, areas and shapes. Projections make different trade-offs to serve different purposes.

This does not mean all maps are false. A projection can be useful and accurate for a defined purpose while unsuitable for another.

For instance, a map developed for a particular form of navigation may prioritise directions. A map comparing countries by land area should prioritise area relationships. A city street map needs sufficient local detail and usable routes.

Why the Mercator world map changes our visual intuition

The Mercator projection is historically important in navigation. It preserves local angles and supports constant-bearing navigation lines, but its distortion of area becomes very large at high latitudes.

Land masses near the poles therefore appear much larger relative to equatorial regions than their actual areas would suggest. A viewer who treats the printed shapes as a direct comparison of country size may reach an incorrect conclusion.

This is a consequence of projection geometry, not proof that a particular mapmaker acted deceptively. The choice becomes misleading when the purpose and limitation are not explained.

The USGS comparison of map projections illustrates why an equal-area projection, Mercator projection and other projection families serve different purposes.

Map scale changes what a map is capable of answering

Imagine looking at a world map and trying to determine which side of a street has an accessible entrance to a clinic. A world map cannot show that detail. It was not designed to.

Now imagine using a detailed neighbourhood plan to compare the relative area of continents. The local map’s purpose and geographic coverage make that impossible too.

A scale such as 1:25,000 means that one unit on the map represents 25,000 of the same units on the ground in the relevant map context. On a printed map at that nominal scale, one centimetre represents 250 metres before considering the practical conditions and limitations of the representation.

Digital screens introduce a further complication: zoom levels, device display sizes and exported images can change the apparent scale. Use a current scale bar or the mapping service’s actual measurement tools instead of assuming every screenshot preserves the printed distance ratio.

The UK’s Ordnance Survey map-skills resources explain how map grids support the location of features. Grid references and scale are tools for finding and comparing locations, not substitutes for checking the route on the ground.

A nearby destination may be far away in practice

Suppose a school and bus stop are only 300 metres apart in a straight line. A fenced parcel, drainage channel or major road may require a longer walk along an approved route.

The map may correctly show where both are located, yet a person using that straight-line distance to estimate a journey could be badly misled.

For a wheelchair user, steps or inaccessible crossings can make a seemingly short route unusable. For a student travelling in heavy rain, shade and shelter can affect the practical journey.

This connects map literacy with How Urban Planning Stalls. The difference between geographical proximity and actual accessibility is a real planning problem, not merely a drawing error.

A shaded area can hide the size of its population

A choropleth map uses colours or shading to represent data for geographic areas such as districts or countries. It can make patterns easy to see. It can also mislead when raw totals are compared across areas with very different populations.

Consider two fictional towns. Town A has 200 reported service complaints among 20,000 residents. Town B has 100 reported complaints among 2,000 residents.

A map of raw totals might make Town A darker because 200 is greater than 100. But the rates are different: Town A has 10 complaints per 1,000 residents, while Town B has 50 per 1,000 residents.

Neither measure automatically tells us which town has worse services. Reporting behaviour, population definitions and the issue being counted also matter. The lesson is that a raw count and a population-adjusted rate answer different questions.

Esri’s 2025 guidance on choropleth normalization recommends making appropriate comparisons using standardised values such as rates and density rather than simply filling geographic areas by raw totals.

The denominator is as important as the headline number

A statistic might report incidents per 1,000 residents, per 100,000 residents, per square kilometre or per hour of service use. These have different meanings.

For example, comparing the number of bus passengers in two districts by resident population may be unsuitable if a large share of riders are commuters or visitors from elsewhere. The relevant denominator depends on the actual question.

A percentage also needs a defined base. Ten percent of what? Adults, households, survey respondents or residents who were eligible for a service?

A good map provides the units, data source and basis of comparison. Readers should resist treating a dramatic colour as a complete explanation when the underlying statistic is unclear.

Colour choices can change what the eye notices

Imagine a map using pale yellow for lower values and deep red for higher values. A reader may interpret red as immediate danger even if the underlying measure simply represents more library visits.

Colour schemes can help, but the choice should fit the type of data. Ordered numerical values often need a readable progression; categories need sufficiently distinguishable colours; meaningful thresholds should not be invented merely to exaggerate contrast.

The legend must say what every colour represents. If bins are unequal, a map can make two areas look very different even when their values are similar, or group distinct values into the same shade.

A strong map makes its classification visible and avoids relying on colour alone where that would undermine accessibility.

The missing-data trap: blank does not always mean zero

A district with no reported figures may appear white on a choropleth. Does white mean no events occurred, the data were not collected or the area lies outside the study? Those are different statements.

A missing observation should not silently become zero. Doing so can create the appearance of excellent performance in a place where the analyst has the least information.

A good legend distinguishes missing data, zeros, suppressed values and areas not covered by the dataset. If privacy or statistical reliability requires suppression, the map should explain this limitation appropriately.

This is especially important for health, education and public-service data where readers may mistake absence of recorded information for absence of need.

A map is also an argument about where a boundary matters

Statistical maps group people and events into geographic units. District boundaries are practical tools, but a pattern may look different when the same underlying information is grouped into larger or smaller regions.

For instance, a city-level average can conceal differences between neighbourhoods. Breaking it into small areas can reveal variation but may also produce unstable rates if too few observations occur in each area.

The right geographic unit depends on the problem, reliability of the evidence and whether the resulting comparison is meaningful.

A careful map does not imply that a shaded region is internally uniform or that every resident shares its average characteristic.

A date can change the entire meaning of a map

A road map prepared before new transport connections were built may still show the old network. A statistical map using data collected several years ago may no longer describe the current population or service conditions.

This does not necessarily make the historical map wrong. It makes the date essential to interpretation.

Readers should look for the date of the underlying observation, the publication date and the stated geographic coverage. A newly designed graphic may be built from older data. A current-looking basemap can contain an old statistical layer.

For fast-changing issues—such as active transport disruptions or emergencies—use official current sources rather than trusting an undated image shared in a group chat.

Mapping is not the same as explaining why a pattern occurs

Suppose a map shows that one area has a higher percentage of residents using a service. It can help identify where usage differs. It does not automatically explain the difference.

Possible explanations might include age, access, service availability, geography, hours of operation or measurement effects. The map may point to a question worth investigating; it cannot establish the cause without appropriate additional evidence.

This is the same distinction that matters in Science: an observed association is not automatically a causal relationship. A map can reveal a pattern, but the explanation has to be tested.

See How Scientific Experiments Fail for why alternative explanations need to be considered before a strong causal claim is made.

Seven warning signs of a misleading map

A world map is used to compare land area without explaining the projection.
The map displays coloured raw totals for areas with very different populations and no suitable context.
A legend omits units, intervals or the meaning of missing values.
The data source or observation date is unavailable.
A straight-line distance is used as though it were a real accessible journey.
A geographic correlation is described as proof of causation.
Colours and selected bins create an alarmist impression unsupported by the underlying values.

These signals invite closer reading. A map with limitations can still be useful if it communicates them honestly.

A map-reading routine for students and parents

Step 1 — Identify the purpose. Is this a route map, world map, land-use map, scientific distribution or statistical comparison?

Step 2 — Check the scale and projection. Does the representation preserve the property you want to compare, and is the scale suitable for the question?

Step 3 — Read the legend fully. What are the units, colour ranges, missing-data rules and symbols?

Step 4 — Inspect the source and date. Are the observations current enough and properly matched to the geographical area?

Step 5 — Ask about denominators. If areas vary in size or population, would a rate, density or other appropriate standardisation be more meaningful?

Step 6 — Separate pattern from explanation. What additional evidence would be required before claiming that one factor caused another?

Step 7 — Consider the real route or person. If the map will guide an actual decision, is accessibility, safety and local ground truth represented?

The questions are short enough to use during a Geography or Social Studies exercise, but they are relevant to journalists, planners and researchers too.

A classroom exercise: two maps from the same information

Give students a fictional table of library visits and district populations. First, create a map of total visits. Then calculate visits per 1,000 residents and make a second map with clearly labelled rates.

Ask students which areas appear busiest on the first map and which have the highest recorded usage rate on the second. Discuss why the answers differ and what the data cannot tell us about people who do not visit libraries.

This lesson shows how two accurate maps can answer different questions. It also teaches why a map’s message depends on appropriate normalisation and honest labels.

The Esri’s guide to statistical data for mapping discusses choosing suitable cartographic symbols and making comparisons with derived rather than misleading raw quantities.

Frequently asked questions

Are all flat world maps inaccurate?

All flat projections distort some geographic properties. They can still be accurate and useful for particular purposes when the chosen projection suits the task.

Why does Greenland look so large on some maps?

Projections such as Mercator greatly exaggerate the apparent area of high-latitude regions relative to tropical regions. It is a projection effect, not Greenland becoming physically larger.

What is wrong with colouring districts by total incidents?

Raw counts can reflect population size or exposure. Rates or other suitable measures may be more useful for comparing areas, depending on what the reader wants to know.

Does a map prove why something happened?

No. Maps can help display and investigate patterns. Explaining the cause usually needs additional observations, a suitable research design and consideration of alternatives.

The final idea

Maps mislead when a useful simplification is mistaken for the whole reality. Scale, projection, statistical denominators, dates, symbols and missing context all affect the story. The best map reader asks not only ‘What do I see?’ but ‘What choices made it look this way, and what question can it honestly answer?’

Sources and further reading

Educational geography and data-literacy guide. Route, weather and emergency decisions should use appropriate current, official information for the location concerned.

Explore the other How Things Fail guides

How Career Guidance Stalls | When Qualifications and Opportunities Disconnect
How Scientific Experiments Fail | When Variables and Controls Are Confused
How Public Procurement Breaks | When Tenders Do Not Deliver Public Value
How Maps Mislead | When Scale, Projections and Context Are Lost — current guide

For more systems-based thinking, see How Public Policy Fails and How Scientific Research Fails. The unchanged Secondary 1 Mathematics Tutor Clementi | Small Groups Tutorials offers an example of identifying actual needs before choosing an intervention.

How Maps Mislead | When Scale, Projections and Context Are Lost