Principles of Good Visualization

The following are the principles of good visualization by Andy Kirk:

  1. Good data visualization is trustworthy: Truthfulness and accuracy should be an obligation. Trustworthiness is about being transparent giving readers all the information they need in order to feel confident about what they are reading and what interpretations are legitimate.
  2. Good data visualization is accessible: helps to inform judgements about how best to facilitate viewers through the process of understanding. It facilitates accessibility by removing design-related obstacle faced by the viewers. Views should experience minimum friction between the act of understanding ( effort) and achieving the understanding ( reward) has to be framed by context. Some visualisations need to be quick and simple while others offer a prolonged task to make sense of a complex subject or display. The reader should be provided with the most efficient pathway tp the things the designer want to show.
  3. Good data visualization is elegant: Elegance is to achieve a visual quality that attracts the audience and sustain the sentiment throughout the experience far beyond the initial moments of engagement. Visual look and feel is the first thing the viewer encounters before experiencing the consequence of other principle led thinking. Design features should feel seamless, uninterrupted and avoiding unnecessary obstacles.

Information visualization has two components:

  • Representation: is concerned with the mapping from data to representation and how it is rendered on the display
  • Interaction: involves the dialogue between the user and the system as the user explorers the data set to uncover insights.

The two are not mutually exclusive and contribute to end-users experience. Based on the notion of the user’s intent, the following are the principles toward a deeper understanding of the role of interaction in information visualization by Yi Js :

  1. Select: mark something as special to keep track of it. By making items of interest visually distinctive, users can easily keep track of them even in a large data set and/or with changes in representation e.g., selecting a placement in Google Maps.
  2. Explore: show something else helps users to examine a different subset of data cases. User’s see only a limited number of data items at a time to gain understanding and insight as they move on to view some other data e.g. Panning in Google Earth.
  3. REconfigure: show a different arrangement by changing the arrangement of representations to reveal hidden characteristics of data and relationships between them e.g., changing the attributes of the scatter plot.
  4. Encode: show a different representation by altering fundamental visual representation including appearance e.g., colour, the shape of each data element. Visual elements serve important orle for pre-attentive cognition and how users understand relationships and distributions of data items e.g. changing colour encoding or size
  5. Abstract/ Elaborate: show more or less detail to provide users with the ability to adjust the level of abstraction of data representation. This allows users to alter the representation from an overview down to the details of individual data cases and many levels in between e.g., drill down in the tree map
  6. Filter: show something conditionally. Users specify a range of conditions so that data items meeting the criteria are represented e.g., Dynamic query
  7. Connect: show related items to highlight association and relationships between data that are already represented or show hidden items that are relevant to specified item e.g. highlighting directly connected nodes in Sankey Diagram

Other techniques are for

  • Undo/redo to allow users to go backwards or forward to pre-existing system state
  • Change configuration allows users to change various configuration and settings of the system

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.