Designing and Choosing symbols 177
tABle 8.4. Using Visual Variables
Type
Visual
variables
Data
Nature of
phenomena
Notes
Qualitative point Shape, hue
Qualitative
Point
Qualitative line
Shape, hue, pattern Qualitative
Linear
Qualitative area Pattern, hue,
orientation
Qualitative
Areal
Dot
Location, hue,
shape
Quantitative
totals
Point
Additional location
information
required
Dot density
Location
Quantitative
totals
Point
Best if enumeration areas
are small
Proportional
circle
Size
Interval
or ratio,
totals
Point, area
Range-graded
figure
Size
Totals, ordinal,
interval, or ratio
Point, area
Choropleth
Lightness, texture
Interval or ratio,
derived
Volume
Dasymetric
choropleth
Lightness, texture
Interval or ratio,
derived,
Volume
Additional
information
needed
Unclassed
choropleth
Lightness, texture
Interval, ratio,
derived
Volume
Isometric
Hue, lightness
pattern with
texture (marginal)
True or actual
points,
actual or
derived data
Volume
Symbol itself is a line,
but shading can be added
between
the lines
Isopleth
Color value,
pattern with
texture (marginal)
Conceptual points,
derived data
Volume
Shading added between
the lines
178
chapter 9
Multivariate Mapping
. . . for all the interesting worlds (physical, biological, imaginary,
human) that we seek to understand are inevitably and happily
multivariate in nature.
—E
dward
R. T
ufte
,
Envisioning
Information (1990)
Often we need to show more than one variable. If the variables are unrelated, then
two separate maps might be needed, but if the variables are related in some way and
the goal is to show that relationship, then two or more variables can be combined on
one map.
One of the simplest ways of representing two variables is to combine two con-
ventional symbols on one map. The data from Figure 9.1a and 9.1b are combined in
Figure 9.2 as choropleth and graduated circle symbols. Some symbols are specifically
designed to represent multiple variables.
Some multivariate maps represent the variables with graphs or point symbols and
others utilize colors in much the same way as conventional choropleths and qualita-
tive area maps.
