GEO381/550 Lecture note of October 28th 2004 Choropleth Map

What is choropleth mapping?

 

Choros+pleth = area + value

Mapping techniques which symbolize values applied to area

Often called enumeration mapping b/c values are shaded in enumeration units

The most common way to map areal aggregated data

In most cases, the data is classified

Dates back to the early 1800s

Continuous value within enumeration units; separated by abrupt boundaries

 

Choropleth map defined: A method of cartographic representation which employs distinctive shading applied to areas other than those bounded by isolines. These are usually statistical or administrative areas. (Dent)

 

 

Data model concept in choropleth mapping

 

Best way of understanding the data model concept in choropleth mapping is to visualize a prism map.

 

  à  

Prism map                               Choropleth map)

or Figure 7.2                           or Figure 7.1

 

Each enumeration unit is a prism raised vertically in proportion to the value it represents

The higher prisms are normally represented by darker shade

The lower prisms are normally represented by lighter shade

 

Q. Suppose you have to make a map using the following data. Decide whether to use the choropleth map.

 

(1) Temperature records at stations

(2) Infant death rate by county

(3) Total population by census tract

(4) Average persons per household unit by census tract

(5) Sex ratio by county subdivision

(6) Number of hospital beds by county

(7) Average SAT scores by school district

(8) Net sales growth rate of shopping malls

 

When to use choropleth maps

 

I. Phenomenon

 

II. Data

 

In the real world, data is not available in a manner that is consistent with the phenomenon

 

  • Crime occurs at an individual level, but the aggregated value of crime is reported in the unit of jurisdictions
  • Income occurs at an individual level, but the aggregated value of income (e.g. average income) is reported in a certain level of the census geography (also for the reason of confidentiality)

 

Guidelines for deciding whether to use a choropleth map with regard to data:

o        Population

o        Population density

o        Median home value

o        Share of public transportation use

o        Per capita income

o        Crop yield per acre

 

Ø      Do not use choropleth map for data which has nothing to do with enumeration units:

Choropleth maps are used when data occur or can be attributed to definite enumeration units (e.g. The distribution of temperature is not controlled by administrative boundaries)

 

Ø      Do not use choropleth map for total values:

In most cases, the enumeration units on a map are unequal in area

Unequal size of enumeration units can alter the impression of the distribution

 

Mapping total values on different sizes of units obscures fact that the densities are the same

                

             Total Hispanic Population                          Hispanic per square miles

Watch out for the effect of large size of Texas

 

Figure 7.4 (b)

Uniform distributions will be obscured when totals alone are mapped

 

Ø      Do not use maps if the interest is to show exact values:

This applies to all other map types; Lists of value will do better for this.

Remember that a map is “abstracted image” (cartographic abstraction): simplification, and classification masks the details at the expense of structuring the message

 

Limitations of choropleth maps

 

  1. In most cases, enumeration units have nothing to do with the distribution of data. That is, enumeration units create artificial division of values

 

e.g. population density map: does the map represent the reality?

Is it desirable to make the boundary of enumeration units pronounced?

 

  1. Very strong assumption is made in choropleth mapping: the value in the enumeration unit is spread uniformly throughout the unit.

 

Figure 7.5

If the variable is changing within the enumeration unit, the change cannot be detected on the choropleth map

 

Any alternative?

Dasymetric mapping

 

 à

Choropleth map                                            Dasymetric map

 

Source: Mennis, J, 2003, Generating surface models of population using dasymetric mapping, The Professional Geographer 55(1): 31-42

 

Dasymetric maps are a variation of choropleth maps, also mapping standardized, classified data using areal symbols. The difference is that the mapped zones do not correspond to enumeration units, but ancillary variables are used to change the boundaries of zones to better represent the actual distribution of the data within zones.

 

Preliminary considerations in choropleth mapping

 

Ø      Geographic phenomenon

o        Optimal measure of the map subject? (e.g. retail?)

o        Value is normalized by which attribute? (e.g. traffic crash aggregated by enumeration units)

Ø      Number and kinds of areal units are determined by:

o        Map purpose

o        Map scale

o        Symbolization (in particular when you use pattern instead of tint)

 

Data classification

 

  1. Qualitative classification

Classifies nominal data

  1. Quantitative classification
    1. Exogenous schemes: set by external criteria
    2. Arbitrary schemes: set by arbitrary schemes
    3. Ideographic schemes: defined by data distribution
    4. Serial schemes: statistically defined

 

 

Qualitative classification                            Quantitative classification

 

Unclassed choropleth map

 

Not all of choropleth maps are classed:

 

 

Which one do you think present the proportion of married couples better?

 

Unclassed map convey values more accurately than classed maps

Unclassed map convey values less effectively than classed maps

 

Demo in ArcView:

Set the number of values to the number of classes

 

Designing choropleth maps

 

  1. Data classification

What is the best way of representing the distribution of data while serving the map purpose?

§         Rectangular distribution

§         Normal distribution

§         Well-defined break

 

  1. Legend design

Continuous legend and classing designs
 

Noncontinuous legend and classing designs
 

       When the class break is obvious, it is not desirable to assign continuous values in legend designs

       In ArcView, you can type in value in the legend editor

    But it’s tedious; alternatively, include a histogram of the data array (Figure 7.15) or report descriptive statistics (mean and standard deviation)

 

  1. Areal symbolization

 

Recall visual variables used for areal symbols

There are two types of areal symbols which

(1) Use pattern (2) Do not use pattern

 

(1)   Pattern

a.      Arrangement (dots, lines, hachures)

b.      Texture (spacing between elements)

c.      Orientation (vertical, horizontal, oblique)

 

(2)   Color

a.      Hue: used for different variables

b.      Value (gray tones): used for a single variables

c.      Saturation (gray tones)

In ArcView

For pattern, use Fill Palette from the legend editor

For color, sue Color Palette from the legend editor

 

 

  1. Base-map design

Often neglected element of map design in choropleth mapping

Providing relevant background information can facilitate map readings

Figure 7.18: see how choropleth symbolization has been altered due to base-map material

 

  1. Several other considerations
    1. Avoid the black and dark hues

                                                               i.      Can mask base-map information as shown in Figure 7.18

                                                             ii.      Black obscures black boundary symbols

                                                            iii.      White may be interpreted as no data

    1. Always use equal-area maps (map projection property)
    2. Accommodate overprinting for printout

                                                               i.      It may come out darker than the actual tone

 

Census Geography of the US

 

p. 82 (Figure 4.8)

TIGER/Line website

 

Next class: proportional symbol map (READ CHAPTER 9)