162 sYMBOLizaTiOn
DaTa COnsiDeRaTiOns
Unless the enumeration areas are all the same size, densities, ratios, percentages, or
other derived values should be used, rather than absolute values, for the map to be
meaningful. Figure 8.21 shows two areas, each of which contains 5,000 people. If
the two areas are shaded according to the total population, both will be shaded alike.
It is obvious that the density of population is greater in the smaller area and that the
map is misleading when the population characteristics appear equal. If the two areas
are shaded according to population per square mile, area B will show 200 people
per square mile and area A only 50 people per square mile. This becomes especially
No Population
Low Population
High Population
fIgURe 8.20.
Choropleth maps cannot show variation within the enumeration area.
100
square miles
25 square miles
5000 PEOPLE IN
EACH AREA
50 people/sq. mi.
200 people/sq. mi
A
A
County
B
County
B
County
County
fIgURe 8.21.
Values used on choropleth maps should be “derived” (e.g., ratios, percentages,
or densities), or the map is misleading.
Designing and Choosing symbols 163
important when mapping county or state data for the United States where such areas
vary widely; Rhode Island has 1,545 square miles, while Alaska has 663,267 square
miles. San Bernardino County in California contains 20,106.4 square miles, which
is larger than many states, while San Francisco County contains only 231.9 square
miles. The map in Figure 8.22 uses raw data for population, which is not as meaning-
ful as a map of population density.
Dasymetric
Choropleth Technique
The
dasymetric choropleth technique,
although not widely used,
was devised to
overcome some of the limitations of the simple choropleth. Specifically, it takes into
account variations within enumeration areas and therefore permits a somewhat bet-
ter graphic view of distributions. Like the simple choropleth, dasymetric areas are
shaded uniformly, and categories, which are usually derived values, are used. Unlike
the simple version, however, areas of similarity and zones of rapid change are deter-
mined. Figure 8.23 illustrates the procedure. First, areas are determined and densities
or ratios computed. The lines that separate areas have no value; they mark enumera-
tion boundaries and zones of rapid change between categories.
This technique has some of the limitations of the simple choropleth. Exact values
cannot be determined within the area from the map, and like the simple choropleth
it is not designed to represent smooth, continuous surfaces. However, since the areas
chosen have inherent homogeneity, dasymetric maps can come closer to reality than
simple choropleths.
(AWAII
