Taxonomy of Data Visualization Methods - Data Visualization: A Successful Design Process - page 115

Weighing up the following factors helps us to narrow down the variety of options
within each method classification to find the most suitable choice of specific chart
type or graphical method:
• Does it accommodate the physical properties of your data?
• Does it facilitate the desired degree of accuracy?
• Is it potentially capable of conveying a certain metaphorical and design
consistency with our subject matter?
Choosing the appropriate chart type
Attempting to organize chart types based on their primary portrayal method
is not new (see http://queue.acm.org/detail.cfm?id=1805128 and
http://www.visualizing.org/stories/taxonomy-data-visualization ).
The classifications presented in this chapter reveal a personal view—informed by
knowledge, experience, and instinct—of a logical way to organize thinking about
the relationship between data variables, visual variables, and chart frameworks.
The examples presented are intended to cover the most typical and contemporary
approaches being used today. It is an arbitrary statement, but you should find that
on 95 percent of occasions one or several of the chart types shown will cover your
requirements. The remaining 5 percent will probably require custom-built solutions
for very specific data shapes and contexts.
Note that many of the chart types presented hold numerous presentational
characteristics and could belong to more than just one classification of method. For
example, an area chart shows changes over time and enables comparison between
categories. As the chart types have been organized based on their primary method,
the prominence of the time-series nature of this example would lend itself more
towards the "showing changes over time" method category.
As you go through the chapter, you will see the following information:
• The popular and alternative names used to describe each chart.
• The type and quantity of typical data variables you would normally use
with each type of chart. On most occasions any categorical or quantitative
variable is suitable though more specific variable types (nominal, ordinal,
ratio-, interval-scale) are proposed where applicable.
• The visual variables that have been used to represent data (optional variables
are italicized) in each chart.