Data visualization is the graphical representation of data. These visual displays of information can communicate large quantities of data and aid in, identifying and interpreting trends, outliers, patterns, and changes over time (IBM, n.d.). Data visualization is a powerful communication tool when dealing with big data, for it makes information accessible and easy-to understand for a range of audiences. Some common examples of data visualization are charts, graphs, infographics, and even animations.
So.. what makes a data visualization good?
Obviously, a good data visualization must effectively achieve its purpose of aiding in communication, investigation, and understanding. There are a few key principles all good data visualizations use:
1. Hierarchy: the elements of the visualization must be arranged in a hierarchical fashion to emphasize certain points. The positioning of titles, labels, and the size of various elements can help readers interpret data better by establishing a logical flow throughout the graph. A hierarchical structure can also ensure that the correct message is conveyed by highlighting certain aspects of the data.
2. Audience: the designer must take in account the audience’s graph literacy (e.g. in a STEM context). The text used in the title and labels must be appropriately specific for the target audience; abbreviations or the usage of specific terms should consider audience expertise. For example, the x-axis labels of the graph below assumes the reader’s knowledge of “SEM” and “SEO.”
3. Placement: data visualizations are only tools to aid a powerful presentation. Therefore, they should allow viewers to reach conclusions about the data naturally as the presentation progresses. The data visualization must not be “in-your-face” or draw unnecessary attention. Only use it if it is coherent in the argument you are making.
4. Presenting data: before worrying about colors and font sizes, designers must ensure that the graph can show data clearly and accurately. Data points plotted must be distinguishable, and are not covered up or obscured by other elements. The scale is also an important feature to consider; use a linear scale whenever possible and consider transposing a figure to find the easiest way to display the scale. To clearly display data, gridlines can be used to assist with accurate estimation. For example, in the image below, the lack of gridlines can affect the audience’s understanding of accurate data; it is hard to decipher the values of data points without gridlines.
In conclusion, data visualization can be a powerful tool to communicate large amounts of data. However, only by considering key principles in design, will data visualization be effective and useful.
Reference list
DeBois, P. (2020). What Makes a Good Data Visualization? [online] CMSWire.com. Available at: https://www.cmswire.com/digital-marketing/what-makes-a-good-data-visualization/.
Gomes, M.M. (n.d.). Data Visualization: Best Practices and Foundations | Toptal®. [online] Toptal Design Blog. Available at: https://www.toptal.com/designers/data-visualization/data-visualization-best-practices#:~:text=A%20great%20data%20visualization%20should.
Maillardet, D. (2021). Five principles of good graphs. [online] Statistical Consulting Centre. Available at: https://scc.ms.unimelb.edu.au/resources/data-visualisation-and-exploration/data-visualisation.
Comments