Cornell University defines digital literacy as ‘the ability to find, evaluate, share, and create content using IT and the Internet’. Data Visualisation is a concept we explored with James O’Sullivan through text analysis and the creative outputs of programs such as ‘R’. For a group presentation, we chose to focus on Data Visualisation, and examine some of the benefits and pitfalls of using such a powerful tool.
The history of Data Visualisation can be traced ‘back to 1160 B.C. with the Turin Papyrus Map which accurately illustrates the distribution of geological resources and provides information about quarrying of those resources’. Today, the propagation of Data Visualisation techniques has been propelled forward by the increased availability of open source software and open data which Tim Berners Lee advocated for in his Ted Talk in 2009.
The process of producing a Data Visualisation, comes with challenges. There is an entire Twitter account (accidental_art) dedicated to collecting the art created by faulty data sets, or mistakes that occur within the processing of that data.
Mistakes are of course harmless, as long as they are actually mistakes and are rectified to ensure the integrity of the data being represented.
Darrell Huff’s 1954 book ‘How to Lie with Statistics’ is still relevant today — even more so perhaps than in 1954, as we see an increase in the utilisation of Data Visualisations to convey information. Some intentionally misleading examples of Data Visualisations, echoing those outlined by Huff, are presented by the Huffington Post in an article entitled ‘How to Lie with Data Visualization’.
- Gun deaths in Florida appear to be declining, however, the Y axis is upside down. Zero is in the top left corner.
- This Y axis can be truncated and drawn to represent which ever truth you find convenient today!
- This pie chart adds up to more than 100%
In 2015, Pandey, Rall, Satterthwaite, Nov, and Bertini conducted a study titled ‘How Deceptive are Deceptive Visualizations?; An Empirical Analysis of Common Distortion Techniques’. They found that participants were indeed effected by misleading visualisations, even when presented with the most common tactics or obvious deceptions.
Out of the 38 selected participants who saw the deceptive visualization, 30 responded incorrectly, 7 correctly and one chose the uncertain response. For the 40 participants who saw the control condition, 39 responded correctly, 1 responded incorrectly and no participant reported uncertainty.
Consider the prevalence of data visualisations you encounter daily. What selective data is your favourite news channel pushing? If you switch to another channel, will you see the data displayed differently, suited to an opposing agenda? We increasingly rely on data visualisations to give us the information we need. As open source, and the semantic web grows, I put it to you that we need to be more critical and analytical of these visualisations. We need to be more literate in all areas of our lives, in reading the words that aren’t even there. If you find yourself in the position of being a Digital Producer, be responsible in presenting your data and be ‘sure that every piece of data and every visualization has been scrutinized before it goes public’.