Geography Reference
In-Depth Information
used for multivariate epidemiological data as described by several authors
(Battersby et al.
2011
; Stewart et al.
2011
). Small multiples use areal symbols for
depiction of spatial or temporal variables and can serve as a comparable represen-
tation with ring maps. Other multivariate visualization methods use different
graphical variables and are not easily comparable with ring maps. Moreover,
experimental methods such as trivariate dot maps are not traditionally used and
their use by Mar
ˇ´
k(
1998
) caused an extensive discussion about the usability of
such maps among the health community. The main goal was to validate the
suitability of a relatively new visualization method (ring maps) for multivariate
data analysis and to compare it with the conventionally used method (small
multiples). We adopted the hypothesis, that ring maps are a more useful tool for a
complex analytical task rather than for simple map reading.
The Use of Ring Maps
The pursuit toward reaching the aforementioned goal was initiated in several
studies of multivariate visualization using ring maps (see below) and introducing
this method as: “
positive and encouraging in terms of continued production of
the maps for (epidemiological) data analysis” (Battersby et al.
2011
). Huang
et al. (
2008
) and Zhao et al. (
2008
) constructed and described the ring map as a
depiction of multiple attribute data sets as separate rings of information surrounding
a base map of a particular geographic region of interest. Battersby et al. (
2011
)
utilized ring maps for visualizing spatial and temporal relationships in multivariate
epidemiological datasets and demonstrated a spectrum of possible uses. Stewart
et al. (
2011
) deployed the ring map method for non temporal (spatial attribute) data
and also discussed the disadvantages of ring maps, citing the complete loss of
topology in the first place followed by the limited ability to display a large number
of enumeration units.
The other authors proposed two future research directions for studies evaluating
the advantages and disadvantages of ring maps. The internal direction is focused on
the method itself and revelation of possible limits—number of aggregation units,
data types and predetermination of data classes, number of rings and their optimal
shape, and the effect of graphical variables used for symbolization (hue, saturation,
point symbols and texture). The external research is directed towards a comparison
of the ring map visualization method with other multivariate visualization methods
and also with method usability for a different map use (exploration, analysis,
synthesis, and presentation). Both research directions can be further divided into
the static and dynamic visualization branches. Table
1
compares several existing
studies from the internal research point of view.
Explanation of terms is given on Fig.
1
.
Chan et al. (
2013
) developed a Python tool enabling the creation of a simple ring
map within the ArcGIS environment and thus opening the possibility of more
extensive use of this visualization method. This tool allows its users to design
...
