Chapter 3 - Part 1: Choropleth Maps
This chapter covers various types of maps and their features. Many types of maps exist so that cartographers can visualize spatial phenomenon in the most advantageous way. It is important that you be aware of all the different map types available to you so that you can visualize your data in the format that will be most appropriate for the content of your map and your map user. This chapter covers three map types: choropleth maps, proportional symbol maps, and dot-density maps. The first part of the chapter covers choropleth maps.
3.1: Choropleth Maps
A choropleth map is a map where colored or shaded areas represent the magnitude of an attribute. For example, this map shows the population density in the year 2007 for the United States of America. For each state, the number of persons per square mile has been calculated. The states with a lower population density are shaded with a lighter gray color. The states of a higher population density are shaded with a dark gray color. The states with population densities between the two extremes are shaded on a continuum from the lightest gray to the darkest gray. Based on the five different shades of gray, the map visually represents in an intuitive manner where the most densely populated states are in the United States.
Why Create a Choropleth Map? There are many reasons why you would want to create a choropleth map. Choropleth maps are relatively easy to create and easy to interpret by map readers. A choropleth map excels at displaying variables overall geographic pattern. As each color or shade is assigned a value or range of values the map reader can ascertain the values displayed on the map easily. Choropleth maps are also excellent for comparing multiple choropleth maps with one another to see how the spatial distribution of variable changes. We refer to multiple small maps on the same page as small multiples. A small multiple is when you have multiple small choropleth maps made with similar structure and context.
It is important that the only data represented on the choropleth map is data that can be linked to an enumeration area. An enumeration area is an area where data is collected and combined. Common enumeration areas in the United States are states, counties, and regions. It is important that the data be normalized against enumeration areas or a total population true blue area or size bias. What is meant by area or size bias is that larger areas might tend to have more population simply because there is more area for people to live? Smaller areas will naturally have less population because there is less area to live in.
By normalizing data against the enumeration area and taking size out of the equation you can do apple to apples comparisons between areas that have different sizes. Examples of normalized data are population per square mile which gives us population density, and percent unemployed which is the number of unemployed people divided by the total number of people eligible for work. In a choropleth map, the boundaries of areas do not have a related value. In other words, the outline of an area is in no way related to the value of the area. Only the different color or shaded the area is related to the value.
Inappropriate Data: There are several types of inappropriate data for choropleth mapping. Continuous data is not appropriate for inclusion on the choropleth map as continuous data is not controlled by an enumeration unit. For example, air temperature is not confined to County outlines nor is it logical to assume so. Another type of inappropriate data is any map ratio not involving an area. That means if a value can be assigned to a very specific point and it is not logical to aggregate that data value to a larger area, it should not be used for the choropleth map. Total values should not be used on a choropleth map. Variables, where the values vary too much, should also not be used for choropleth maps. If you still wish to use variables with a large value range you may want to consider choosing a smaller enumeration unit so that there is a smaller variation in each enumeration unit.
Data Classification: In order to display the vast majority of data on a choropleth map, you must employ some data classification schemes. We classify to simplify generalized the data for display on the map. In general, four to seven classes are preferred. If you need to exceed seven classes you need to keep in mind that humans cannot effectively use more than eleven classes at once. Now consider the five types of classification methods to determine when to use each one on a choropleth map. The equal-area data classification method is useful for layouts including multiple maps. The equal frequency data classification method is useful if you are performing a statistical test between classes. The arithmetic and geometric data classification methods are useful for the data that shows a normal distribution. The nested means data classification method is useful for non-normal distributions. And finally, the natural breaks data classification method maximizes homogeneity within classes and is typically going to be your best general choice for displaying data on a choropleth map.
Projection: In most cases for choropleth maps the equivalent, or equal area, projection are the most appropriate. The reason why the equal area map projections are the most appropriate is that since we typically normalize data based on area, relative size is important to maintain when comparing the values of the underlying enumeration.
Symbolization: Consider the symbolization choices for choropleth maps. If you are producing your choropleth map using black and white colors only, then the black color should represent larger values in the light grey color should represent smaller values. The color white should be generally reserved for the background of the map or outlines. Additionally, you should use caution with pure black and white fills as they may obscure boundary lines. You may use pattern, dot, line, or Hatcher patterns, instead of shades of gray, but this is considered to be the “old style” of choropleth mapping.
Black and White: Here is the population density choropleth map of the United States for the year 2007. In this black and white map the color white is used for the background and state outlines. The light gray color represents lower values and the dark gray color represents higher values.
Color: If you are producing your choropleth mapping color you should consider these points. Darker or more saturated colors represent larger values. Lighter or less saturated colors represent smaller values. Make sure you can easily differentiate between colors of different classes, that is, make sure that no two adjacent colors are too similar.
You should avoid qualitative color schemes on a quantitative choropleth map. That means, if you only have one variable on your map, such as population, you should choose a single color, or hue to represent that you are only showing different values of a single kind of thing. To represent the different quantities of that single thing, vary the saturation or value of the chosen color. On a color map, white suggests ‘light gray’ can effectively represent “no data”. Black or white is an effective boundary color on a color choropleth map.
On this color choropleth map that deals with a single variable of population, a single hue of green was chosen. The green hue is varied in saturation so that the lighter color green represents lower population density in the darker color green represents a higher population density. The color black was chosen as the state outlines as it provides great contrast against the green hue.
Color Schemes: In general, there are three color schemes that should be used on a color choropleth map based on the type of data being displayed. If your data is considered unipolar data, which means that there is no natural dividing point, you should use a sequential color scheme. An example of unipolar data is population density.
If you have bipolar data, which means that has a natural dividing point such as 0, or mean, then you should use diverging color scheme. Examples of bipolar data are population gains and losses. If you have balanced data, this means you have two complementary phenomena. In this case a diverging color scheme is appropriate. An example of this would be the ratio of males to females.
Reference Features: Thematic maps should be simple by design by focusing on the featured variable. When creating them you should avoid placing reference features on the map and less they are important in explaining the pattern of the variable being mapped.
Legend Design: To wrap up our discussion of choropleth maps this section will focus on legend design.
Legend Boxes: Choropleth maps typically use legend boxes. The legend boxes are typically square or rectangular and are large enough to provide a visual anchor but not too large to distract the eye from the main map body. The symbols in the legend should be identical to the symbols on the map in both color and line weight.
If the enumeration units on the map are reasonably small then the size of the symbol on the legend should be about the average size of the enumeration unit on the map. If the enumeration units on the map are very large then you may consider making the boxes ½ to 1/3 the average size of the enumeration unit on the map.
Legend Layout: The legend should be laid out into orientations: horizontally or vertically. With the horizontal orientation, the lowest value should be on the left and the highest value should be on the right. The value numbers should be located below the boxes. In a vertical layout you can either have the lowest or highest value on the top and the lowest or highest value on the bottom as neither layout, is considered standard. The numbers representing the value should be to the right of the boxes.
Continuous Classes: In the case where the values on your map represent continuous data, which means that the maximum value in one class is slightly less than the minimum value of the next class, you should follow these guidelines. Your legend should emphasize the degradation of values and show that the values on the map exhaust the data. Additionally, the legend should reinforce the fact that there is no data that falls between the cracks. By having a continuous legend it allows the same legend to be applied to multiple maps. In our example, to reinforce the fact that the data is exhaustive and there are no cracks, the boxes of but each other and look like a single continuum.
Non-Continuous Classes: If you have non-continuous classes on your map then your legend should show the actual extreme values in each class. You would want to do this to narrow the reader’s estimate of the actual values in each class. A non-continuous legend is best with a single map displaying non-continuous data. For example, our legend shows each box separated from the other boxes to reinforce the fact the classes are not exhaustive and on a single continuum.
Formatting Conventions: Use either “to” or “-“ between class ranges. If there are more than four numbers a value, use a comma after every third number left of the decimal value. If you choose to have a legend title, the legend title should match the topic of the map and should not use abbreviations including symbols.
You should place any ancillary text below the legend of the information will not fit into a concise legend title. If you are creating an animated map that shows the change of the variable throughout time, you should use a single continuous legend that encompasses the global maximum and global minimum over the entire series of the maps included in the animation.
Storymap about choropleth maps
This part of the chapter covered various types of maps and their features. This first part of the chapter focused on choropleth maps. You learned about appropriate and inappropriate data as well as data classifications and symbolization for each map type. Map legends and the elements that should be considered when using this feature were also covered.
This work by the National Information Security and Geospatial Technologies Consortium (NISGTC), and except where otherwise noted, is licensed under the Creative Commons Attribution 3.0 Unported License.
Authoring Organization: Del Mar College
Written by: Richard Smith
Copyright: © National Information Security, Geospatial Technologies Consortium (NISGTC)
Development was funded by the Department of Labor (DOL) Trade Adjustment Assistance Community College and Career Training (TAACCCT) Grant No. TC-22525-11-60-A-48; The National Information Security, Geospatial Technologies Consortium (NISGTC) is an entity of Collin College of Texas, Bellevue College of Washington, Bunker Hill Community College of Massachusetts, Del Mar College of Texas, Moraine Valley Community College of Illinois, Rio Salado College of Arizona, and Salt Lake Community College of Utah.
This workforce solution was funded by a grant awarded by the U.S. Department of Labor's Employment and Training Administration. The solution was created by the grantee and does not necessarily reflect the official position of the U.S. Department of Labor. The Department of Labor makes no guarantees, warranties or assurances of any kind, express or implied, with respect to such information, including any information on linked sites, and including, but not limited to accuracy of the information or its completeness, timeliness, usefulness, adequacy, continued availability or ownership.