Displaying 2 attributes at the same geography
We have a base map at the state level with 4 categories.
We want, also at the state level, to show the percent of poverty (4 categories).
I have tried lines/crosshatches, transparencies levels, symbols, graphics, and other things.
I know this is a ridiculous request, but I get this all the time.
Mapping Center Answer:
I am interpreting your question as asking how you can show for one type of enumeration unit (in your cases, states) four themes (which would be related to four different attributes in your data) simultaneously. I assume that your attributes are all quantitative. I also assume that you want to use a choropleth mapping method to show at least one of the attributes, and that you are hoping to use some symbology other than a solid color fill to show the other three attributes on top of the attributes that is shown on the choropleth map.
If I am right about this interpretation of your question, then I suggest that the BEST solution for you might be the use of small multiples which would NOT require the four themes to be overlaid on the same map. Instead, with small multiples, you would create four separate maps - one for each theme. Small multiples are defined as "a series of displays with the same graphic design structure used to depict changes in variables from multiple to multiple (i.e., map to map)". The consistency of design assures that the map reader's attention is directed toward variations in the data and relationships between the geographic distributions of combinations of themes.
An alternative, though in most cases not as good as small multiples, is to display multiple variables in a single symbol, which is sometimes referred to as a "glyph". In this case, parts of the glyph are used to represent different data attributes or themes. An example is a Chernoff face, in which the eyes, nose, mouth, eyebrows, shape of the head, etc..., each represent one theme. Another example is a bar chart map in which a chart is shown in each enumeration unit and the bars on the chart relate to four attributes. A map made this way may be good for inspecting individual attributes, but it may be difficult to estimate and compare proportions, especially if there are many different segments in the glyphs. Also, the visual "field effect" of nearby symbols can alter perception of any one symbol, and it may be difficult to compare parts of symbols widely separated on the map, especially if there are many intervening symbols.
For two or three attributes, you could use bi-variate or tri-variate mapping, respectively, in which you simultaneously depict the magnitude of attributes within the enumeration unit using color. However, this won't work for four attributes.
Note that if I am interpreting your question incorrectly, then there may be a different answer, although small multiples will also work with qualitative data!
If you would like to post a comment, please login.