Another Way to Show Multiple Variables on the Map
Example: Patterns of Residency by Ethnicity
Pie charts are not the only graphic form which can display a lot of information in a compact form on the map. Figure 2.57 shows an example of the often complex pattern observed in American cities in the relationship between residential living patterns by ethnicity and the availability of economic resources.
FIGURE 2.57 Ethnicity and median rent, Santa Ana, California, 2000
This map shows the relationships among several variables of interest to those who want to understand issues of residential settlement patterns, segregation by ethnicity, and the impact of economic resources like income and subsequently the ability to own or the need to rent housing on these relationships. The stacked bar graphs in each census tract in Santa Ana, a city in Southern California of about 340,000 residents located south of Los Angeles, represent the percent of residents in the census tract that either rented (in light gray (yellow) on the bottom of the bar) or owned (in darker gray (green) at the top of the bar) the residence they live in. The thematic background gives the percent Latino in each tract; the city as a whole was 76.1 percent Latino in 2000.
A number of patterns emerge from an examination of Figure 2.57. First, notice that the areas with the highest concentration of Latinos are all in the center of the city—from what we all have experienced about city centers in America, you realize that these areas are likely to be poorer, with less desirable housing, more multi-unit apartments and duplexes, and fewer single-family home-style neighborhoods. Likewise, the outskirts of the city are characterized by a medium tone in the thematic map, with fewer Latinos—remember that Santa Ana is 76 percent Latino. So the distribution of residential settlement is closely linked to ethnicity in this city, as it is in many American communities.
The bar graphs tell us something about the type of housing in each tract—the places in the center of the city are more likely to have the bottom (yellow) portion of the bar greater than the top (green) portion, indicating a majority of renters in the tract. This is especially true in the upper portion of the map and to the right of the map in the black (red) thematic areas where there are fewer Latinos and more Anglos. This is also the case in some of the light gray (green) areas to the right of the center of the city and the black (red) areas at the top right of the city. These are areas with fewer Latinos than Anglos, but they still show a majority are renters rather than owners.
The opposite is the case on the left side of the center of Santa Ana. In the light and medium, and even some of the dark areas, where Latinos are in the majority, home ownership is higher than renting—but the map overall shows that in a city with a large Latino majority, most areas are dominated by renters versus owners. Obviously income is part of the issue, and many more Latinos are poor than are Anglos in this city as in most places in the United States. The map in Figure 2.57 suggests that economic resources are not the full explanation, however, as the patterns displayed there seem to suggest that some kind of segregation between these two groups had operated to produce the link between ethnicity and home ownership. Census data show that overall in Santa Ana in 2000, about 18 percent of Anglos were poor, whereas about 23 percent of Latinos were poor, according to the U.S. Government definition of an income below the poverty line for a family of four, about $18,500. Although this is an important difference, it is not so great as to fully explain the differences in Figure 2.57. If you were concerned about the fact that fewer Latinos own their homes in Santa Ana, you might start looking at lending practices (Do banks tend to lend money for apartment complexes in some neighborhoods and single family housing in others, and does this correspond to the dominant ethnicity in the area?), or you might want to start and/or strengthen existing programs to assist first-time home buyers, especially those who are Latino and/or Spanish-speaking. You can see how maps can give you ideas for policies and practices that you would not have known were needed until you see the information the maps convey.
To produce the map in Figure 2.57, start with a map of the city of Santa Ana, as shown in Figure 2.58; these maps can be downloaded from the US Census Bureau’s website (www.census.gov) in a format ready to open in ArcMap.
FIGURE 2.58 The city of Santa Ana, California, 2000
Step 1 Add the map and layer for California census tracts from the ArcGIS database; this map is a detailed map of the entire state of California (Figure 2.59) and the accompanying attribute table has the basic census data needed to build the thematic map and the stacked bar graphs in Figure 2.57.
FIGURE 2.59 California Census Tracts, 2000
The city of Santa Ana is lost in the density of a large number of census tracts in Southern California around Los Angeles. Use the Santa Ana city map to cut out of the state tract map a selection that has all the tracts that are either in part or completely inside the city limits.
Step 2 Open the Selection tool from the main tool bar, and click on Select by Location.
Step 3 Make sure the first box reads "select features from," and check the California Census Tracts, 2000 layer as the location you want to select from.
Step 4 In the box under "that:" select the method for this selection,"intersect"; this will give you all the tracts within the city and those that are partially within the city that intersect the boundary.
Step 5 Under "the features in this layer" make sure that the Santa Ana city layer is visible; your submenu should look like that in Figure 2.60.
Figure 2.60 Select by Location submenu
Step 6 Once you click on apply, the selection will be made and highlighted on the census tract map of the state; left-click on the census tract map layer in the display contents, and click on Selection, and click on Zoom to: Selection; your map should like Figure 2.61.
FIGURE 2.61 Selected tracts for Santa Ana from California
Step 7 Left-click on the original layer again, click on Selection, and click on "Create layer from selected features," and the new layer will be added to the display contents under the same name as the original layer with the word, "selection" added to the end of the title, as in Figure 2.62.
FIGURE 2.62 Census tracts for Santa Ana, California, 2000
Use the procedures described previously to produce the thematic map for the percent Latino population. To produce the stacked bar graphs, use the following steps.
Step 1 Copy the thematic layer and paste into the display contents as a new layer; change the name to prevent confusion. Double-click on the new layer and select the Symbology tab.
Step 2 Select charts from the Show box, and click on stacked from the submenu.
Step 3 Select the two fields to display in the bar graph and click them into the box on the right.
Step 4 Click on each symbol to select the color scheme you want to use to represent each field in the bar.
Step 5 Unclick the box labeled "Prevent chart overlap" and click on the Properties button; turn off the 3D option and unclick the box for "show leader lines."
Step 6 Click on OK to get back to the main Layer Properties menu, and your screen should look like that in Figure 2.63; click on Apply and OK.
FIGURE 2.63 Building stacked bar graphs to display over a thematic map
In these last two examples, you have learned how to add graphic elements to your map to represent additional information, thus expanding the way in which multiple sources of data can be mapped and more complex relationships made clear, especially the way these relationships have a geographic aspect. This is only one way of displaying more complex information; the next examples show how underlying models of geospatial data can be mapped and displayed in very powerful and imaginative ways using some of the more advanced features of ArcGIS.
