Project advisors: Dr. Deborah Littlejohn & Dr. Matthew Peterson
2017
Centuries before “Big Data”, information visualization was mostly a tool for inquiry and documentation, which magnify the power of statistics. However, as physical forms meant to represent abstract concepts, visualizations lie. Perhaps, Mark Twain put it best – “There are three kinds of lies: lies, damned lies, and statistics”.
The purpose of the Data Stories project is to understand the relationship between forms and abstract information (i.e., data) in the context of dynamic Data Viz & Big Data. The visualization in this project will highlight the argument for the false correlation between two data sets. And the ultimate goal is to reveal the lie in the data by telling a compelling visual story.
+ DATASET 01 – New building permits in Raleigh (2000-2016)
+ DATASET 02 – North Carolina severe weather damage data (1995-2016)
I picked up two unrelated data sets and followed a systematic process of cleaning the data. Each data set was dissected into possible variables. I filtered down the datasets to only include data between 2000 – 2016.
I became aware that each step was actually a lens of representing the data. From data to information to knowledge, a “story” could be delivered through the designer’s lens. Likewise, it can be interpreted divergently by different readers through their lens.
Moving to a more elaborate visualization, I shaped the information generated from the 1st dataset in a mountain-like layout with a Layered Area Chart which makes use of the perspective to compare the new building permits development over time (by months and years). It is pretty clear as you can see the number of permits decreased dramatically since 2007.
Digging further, I was wondering if there was any “story” between the proposed use of new buildings and the severe weather cost. So I made the Bubble Timelines to display how different new building usages changed over time. However, there seems no more correlational story to tell.
Taking a big picture, this visualization compares the amount of Raleigh new building permits each year vs. N.C severe weather cost from 2000 to 2016. I could easily see that when the number of permits reached the peaks, severe weather costs are low. While severe weather costs are reported high in 2011 and 2016, the amounts of building permits in Raleigh are relatively low. The lie was born – the more building permits, the less cost from the severe weather damages…
I continually played with round shapes and put the two datasets together. Although the visualization looks pretty, the structure makes it difficult to tell any correlation between them.
Overall, from this project, I confirmed once again that a designer is not simply and objectively representing truths (and designers even define truths in some situations). Visual communicators have the advantage of showing people what they want people to see. It is easy to manipulate data as a beam of light passes through a triangular prism. They said, “there are a thousand Hamlets in a thousand people’s eyes”. Actually, the “Hamlet” you had read was chosen from infinitive possibilities on purpose.