Data graphs encode the data into a visual design object.
Data graphs are not objective facts like tabulate numbers – they are subjective design objects. As design objects, data graphs require aesthetic attention.
But design is not data, and that is why design is ordered as the last quality of data graphs in order of priority. Design must be exercised with restraint and never compromise any other quality of higher priority. That is to say, design can only be met without compromising accuracy, completeness, relevance, consistency, efficiency, and simplifying complexity.
Good design
Good design in data graphs must be solely focus on highlighting data. Remember that any decorative element violate the quality of relevance and therefore do not qualify.
Good design must be focused in improving the following aspects:
- Reduce the visual prominence of non-data elements, and maximise the data-ink ratio. That is, bring into focus the data and reduce the focus of any non-data information.
- Provide complete and sufficient identification and think about fonts.
- Selects colour hues and patterns with excessive caution given the potential of creating visual artefacts and reducing accuracy in decoding.
- Aim for representational coherence and ease in interpretability, in accord to the quality of consistency.
- Take into consideration the decoding effect on visual perception, and take advantage of our pre-attentive visual strengths.
White space
White space is the best friend of the data graph designer.
Great design adorns the data through the application of ample white space, because it is through white space that the encoded data becomes visibly divisible information.
If a graph is packed with encoding element and identification then it would be very hard to perceive any information. It is the white space that helps the user decode discernible messages.
Consider the following example of a chart that fails to consider the effect of white space. There is very little white space. But then again, this chart was probably designed specifically to confuse rather than inform:
Regardless the context (and I do not want to take sides), this graph uses design and the lack of white space to highlight data complexity in terms of network connections. By doing so, it is in direct violation of the quality of simplifying complexity, because the design does not help the user understand the complexity in the data. All we can say is “wow this data is so complex”, which seems to be aim of the designer. Other than that we learn nothing.
Back to Complexity ⟵ ⟶ Continue to Data management
Graph workflow 