Statistical maps are a popular way to present various types of quantitative data. They display the distribution of the variable related to location.
Cartographers use many different methods to visualize nominal information. Their goal is to use the best design that communicates the statistics without misleading the reader.
It is important to understand the different types of statistical maps. This can enable users to interpret the data quickly and correctly.
This article breaks down some of the most common visualization techniques. It also looks at the process of statistical mapping.
Choropleth maps display data variations by using color progression (shading). This information is usually in relation to a particular geographic area.
The design allows a comparison of the variable patterns across a selected location. For example, it can display the vote distribution by political party per county in the United States.
Or it could show the unemployment rate ratios across different city areas. This data could be later compared with the crime rates across the defined locations.
There are three keys to the correct use of choropleth maps:
An isopleth map (a contour map) uses lines (isopleths) to connect areas of a region where similar phenomena or values occur. The isopleths can also separate higher measurements from lower measurements.
The spacing between the lines displays the rate of how much the quantity changes in relation to distance. This rate is the gradient.
This visualization method is used in maps that show relief changes over a certain area. The lines connect points with the same elevation.
Another typical example of isopleth maps is weather maps.
Meteorologists use contours to describe weather patterns. These include pressure (isobars), wind speed (isotachs), or temperature (isotherms).
Keys to the correct use of isopleths:
Another type of statistical map is a dot map. It uses dots to represent the density of variables.
Each dot can represent a single data point (one-to-one dot-density maps). Or they can represent a number of points or values (one-to-many dot-density maps).
Dot maps are an efficient way to visualize the geographic distribution of a given value. They can display population density, racial distribution across an area, etc.
Dot maps can work for both raw data /simple counts (e.g., number of farms) or rates and ratios (e.g., number of farms per sq kilometer). They are easy to read and highlight densities and clusters of the variable.
But it is difficult to retrieve specific numbers from the design. It would be time-consuming to count the individual dots to find out the exact number of the value.
Two key parameters for dot map design:
The bigger the dot, the smaller the space between dots. At some level of density, the individual dots will start to overlap each other.
Once this happens, it is impossible to discern any higher densities.
It is important to calculate the balance between dot size and dot value. The best approach is to make the dots big enough to see them individually.
At the same time, the dot value should be small enough that even the area with the lowest values has more than one dot.
A proportional circle map uses a symbol, usually a circle, to represent the data value in a particular geographical area. The bigger the symbol, the bigger the measurement in that location.
One approach is to scale the size of the symbol in proportion to the number it represents. For example, a number twice as big will be represented by a circle that is twice as large.
Another method is to create a graduated symbol map. It may use a small number of symbol sizes corresponding to a small number of categories of value size.
The categories will then stand for a range of measurements. For exaple, city size categories could cover: <500,000, 500,000-2 million, 2-5 million, and >5 million people.
This type of statistical map can be used to show the location and magnitude of earthquakes across a geographical area.
When using a pie chart in relation to a geographic area on a map, designers create a pie chart map.
On a pie chart, the individual slices represent a percentage or numerical proportion of the data. They are useful for visualizing a smaller number of categories, around 8 or less.
When used on a map, it is possible to link the statistics with their corresponding locations. They can be also combined with a Bubble Chart (Proportional Symbol Map).
This form of design will add an extra dimension for visualization.
Pie charts are useful to express a part-to-whole relationship in the data. The whole pie represents 100%, while the individual parts make up a portion of the data.
This kind of map is often used for business statistics.
For example, if a business offers many services, a pie chart can provide valuable insights. It can reveal which service is most popular and how the data varies at different locations.
Heat maps use different colors or color shades to represent data variations. They display the level of concentration of the value in relation to a location.
Heat maps are not the same as choropleth maps. The data isn’t categorized by regions or geographical limits.
The focus is the concentration or the intensity of the value. This means that geopolitical boundaries are not relevant for these types of statistics.
Heat maps are often used to display weather and natural phenomena. By visualizing high-intensity measurements, they can communicate the most important data.
They are also used by companies to describe business data, such as sales.
Working with geographical and statistical data requires a systematic process. It includes the following important steps:
Geographic data needs to be enumerated. This process involves “a complete, ordered quantitative listing of all spatial data items in a data collection.” (University Consortium for Geographic Information Science)
In this step, convert the measured value into another form of a variable that is easier to work with, e.g. a ratio. This process eliminates the possibility of misinterpretation.
Group similar measurements together and create data categories. This process breaks down the information in a way that enables a clear interpretation.
The examples of the different types of statistical maps show that cartographers need to understand the purpose of the data visualization.
Some variables need to be grouped in relation to geographical regions.
Certain maps show clusters or high concentrations of the value. Other maps create connections between the same measurements.
So cartographers must identify what kind of information they are working with. Then they will be able to find the best way to display it.
If you enjoyed reading this article about statistical maps, you should read these as well: