Hierarchy- charts ideal for displaying organisational structure & classification

A frequent topic that comes up when talking to a lot of our clients is around identifying the right types of data visualisation to support data that has been collected, parsed , filtered and prepared to be shared. We thought we start to answer them through a series of blog posts about displaying data- starting here with displaying hierarchy.

Hierarchy is the arrangement of items or information in a way which each of them are represented at a different ranks (or levels) to one another.  This is a useful technique that is used in a wide range of areas:

• Determining how companies and organisations are structured, in regards to its members and the designation of job roles.
• Tracking the biological descendants of family members or of animals through evolution.
• As a classification tool in science, information technology and other fields.
• In linguistics for phrase structure or for tracking the history of a language.
• In visual hierarchy for graphic design: where the importance of visual elements is displayed through their arrangement or presentation.

In-fact hierarchy can be used in any situation where you need to communicate any ranked relationship. Displaying hierarchy visually gives form to these arrangements.

Although hierarchy is typically depicted as a divided pyramid, there are more complex, alternative ways of showing this structure, with a variety of purposes.

Tree Diagram

Also known as organisational chart or linkage tree.  The tree diagram is oldest form of displaying hierarchical structures visually.  Taking inspiration from trees and their branching structure, tree diagrams display hierarchy through dividing points known as “nodes” that extend out at each rank.   There are three types of “nodes” in a tree diagram:

• Root node, a member that has no rank above it.
• Nodes, any member that represents the connections between members.
• Leaf nodes, are the end nodes or members who have no ranks below themselves.

Tree diagrams are typically used for family trees and the organisational charts of companies.  Below is an infographic that has used tree diagrams for displaying data of the family trees of various Game of Thrones characters.

Circular Tree Diagram

This is an alternative way of drawing a tree diagram by using a polar grid (concentric circles). Circular (or radial) tree maps start with the root node in the centre and expand outwards radially. This creates a much more visually striking effect then showing the structure on a straight diagram.Dendrogram

A variation of a tree diagram used for hierarchical clustering analysis by representing the relationships of similarity amongst a group of entities.  In Dendrograms, the branches are instead known as “clades” and the arrangement of them tells us which entities are most similar to each other.  The height of each clade indicates how similar or different they are from each other: so the greater the height, the greater the difference.

Treemap

Originally developed by Ben Shneiderman, treemaps served as a way of visualising large file directories on computer, while being as space-efficient as possible.

Treemaps can also display quantities assigned to each node via its square area size, which makes them useful for comparing the proportions between categories in the hierarchy.  Instead of having each rank on a row, treemap nest nodes inside of each other.

Circle Packing

Also known as a circular treemap as this type of chart uses the same concept as a treemap, but uses circles instead of squares.

While more beautiful to look at, circular treemaps are not as space-efficient as a treemap because of all the empty space within the circles.

Voronoi Treemap

Another variation of a treemap, which uses Voronoi tessellation or Voronoi decomposition for it’s tiling algorithm.  This generates an interesting pattern that tends to resemble a stained glass window.

Icicle Chart

Similar to the structure of a tree diagram, but instead of a node-link construct, icicle charts use as series of juxtaposed rectangles to show rank.  Icicle charts can flow from top-to-bottom (vertically) or side-to-side (horizontally).  The area of each cell can also be made proportional to additional variable, much in the same way a treemap can.  However, unlike a treemap, icicle charts do not nest each rank within their rectangles, making them less space-efficient.

If you are considering visualising hierarchy and are still considering the best visualisation to apply , please feel to get in contact with us and well be happy to talk you through the options.