Lying with data
As data visualizations are commonly used to convey meanings – and sometimes even persuations – people often improperly manipulate the demonstration of data for their own agendas. Here are some real life examples of misleading data visualizations.
Misleading through design
Sources: Correlation is not causation
Many data storytellers, especially academic researchers, are often excited when they observe a correlation relationship between two variables.
However, correlation does not imply causation. There may be other factors that lead to the outcome.
Source: New England Journal of Medicine
For example, in displaying of relationship between chocolate consumption and number of Nobel laureates, the reseacher is neglecting other factors that may have an impact (e.g. socioeconomic status, racial bias) on the outcome.
Source: Causal Flows
This is like saying “I did bad on the exam today because my neighbor’s cat was sleeping.”
Two things may occur coincidentally, but does not mean they have a causal relationship.
There are more interesting (and entertaining) examples on this website called Spurious Correlations. This, again, proves that correlation does not allow us to claim a causal relationship, and that it is irresponsible to visualize it for a publicly-facing publication.
Be a responsible data storyteller!
We hope this section was not only interesting but also empowering, especially realizing that you can do so much better than the misleading-graph-makers. If you aspire to be a data storyteller, please spread the gospel of responsible visualizations and human-centric designs.
"There is no 'Technology'. There is no 'Design'. There is only a vision of how humanity could be, and the relentless resolve to make it so." — Bret Victor, Data Storyteller & UI Designer (worrydream.com)