In this article, I will tell stories using data with population visualization
One of the first things I wanted to do after receiving my COVID-19 vaccination was to visit a museum (with masks and social distancing, of course), and my first such visit was to The Museum of Fine Arts in Houston, Texas.
While milling about the galleries, I was immediately struck by one exhibit of Norwood Viviano’s work titled Cities: Departure and Deviation. The exhibit is, at its core, a visualization of time series data representing long-term changes in the populations of 24 cities. I’ve been focused on time series data at work for the last two years, so was intrigued that this artist had managed to visualize this data type so well that it literally belonged in a museum.
Each city is represented by its own glass-blown pendulum, the length of which represents time, the width of which represents population, and with changes in color that represent significant events. Population trends for each city are therefore captured in a unique 3D visualization reminiscent of a violin plot. Most pendulums contain 3 data points, resulting in fairly long, smooth shapes.
The exhibit effectively draws viewers in with its clean lines and modern greyscale before it even becomes apparent that you’re looking at a representation of real data. It’s an aspect of this visualization that I think is incredibly powerful — the audience of visitors to fine arts museums is broader than the audience that typically seeks out data visualizations, and museum-goers can be effectively drawn in and reached by this exhibit. Before you know it, you’re trying to think through these massive shifts in population trends and considering how the local industries must have changed to create the patterns you’re seeing. The visualization is minimalist, but its impact makes you want to seek out more information.
As a data person, this exhibit also got me thinking about what it left out — what decisions Viviano made and what their implications are. The average person does not spend their days thinking through the nuances of data, and has been shown to be overly quick to place trust in data and in the outputs of data-driven models. I’m inclined to believe that this is especially true in the case of a visualization such as Cities — the clean, greyscale appearance invites you to consider its representation as an absolute truth.
But as we data people know (or at least should know), data do not represent a single inherent truth. They may be cherrypicked and highlighted to tell particular stories, and even the best efforts at transparency can leave out important details or alternative angles. In the case of the Cities: Departure and Deviation exhibit, I couldn’t help but wonder what trends and events might be hidden by these clean glass lines.
I decided to look at the population trends in 10 of these cities, but to instead include all population data that I could find. I was curious how the patterns might differ from Viviano’s visualizations with this more granular level of detail, and what stories might have been left out in the one he chose to tell. To be explicitly clear, I in no way undertook this project to criticize Viviano’s work. To my knowledge, all data he represented is accurate and represented truthfully. Further, seeing as he blew each of these pieces in glass it’s hardly reasonable to have expected him to represent changes at 20 different time points. And even if he could have, I personally think that a more nuanced shape would have taken away from the striking nature of his pendulums. He is an artist, after all, and free to take such liberties with his work.
I do, however, think that looking under the hood of these data a bit is a useful exercise in reminding ourselves that data are not some all-powerful force of truth. Different conclusions can be reached and stories told from exactly the same data, depending on who is analyzing them and what they wish to say.
I made a reasonable effort to find the same data used by Viviano as well as any other reliable population estimates. The data that I used comes overwhelmingly from the US Census, which is conducted every 10 years, although full data citations can be found at the end of this article. The ten cities I analyzed were chosen based upon my own general interest in their histories and not due to any expected findings in the data.
While at first glance Viviano’s pendulums do look like violin plots, actual violin plots were not useable here as they represent distribution (like a boxplot or histogram) and are not designed in their traditional 2D form to represent changes in scale over time. I therefore captured these trends with simple line graphs, although shaded the area underneath the line for easier visual comparison with Viviano’s pendulums. If you were to bisect a pendulum lengthwise and place it down on the cut side, you should see the trend approximated similarly to these line graphs. So, to compare these line graphs to the pendulums, it may be easiest to picture the mirror image of the shaded area as well as the graph itself.
In Conversation With Dr Suman Sanyal, NIIT University,he shares his insights on how universities can contribute to this highly promising sector and what aspirants can do to build a successful data science career.
Avail The Data Science Courses in Bangalore and Kick Start Your Career as a Successful Data Scientist in Bangalore within 4 months. Classroom/Online Data Science Course in Bangalore with Placements or Money Back.
Online Data Science Training in Noida at CETPA, best institute in India for Data Science Online Course and Certification. Call now at 9911417779 to avail 50% discount.
In the digital era that we live in, data has become the biggest and most valuable asset for most organisations. Data is rapidly transforming the way we live and communicate, and it is by collecting, sorting and studying this data, that organisations across the world are looking for ways to impact their bottom lines. In this post, we'll learn Data Science vs Big Data: Difference Between Data Science & Big Data.
Data Science and Analytics market evolves to adapt to the constantly changing economic and business environments. Our latest survey report suggests that as the overall Data Science and Analytics market evolves to adapt to the constantly changing economic and business environments, data scientists and AI practitioners should be aware of the skills and tools that the broader community is working on. A good grip in these skills will further help data science enthusiasts to get the best jobs that various industries in their data science functions are offering.