In this interview Alex Dixon shares his experience with Data for a Cause. Alex is the winner of the 11th Data For A Cause challenge for the Institute for Economics and Peace. The goal of the challenge was to analyze countries that are making progress towards peace.
Alex is a data analyst for University of Kentucky Research. He is not a traditional analyst. He has a background in graphic design - print design and visual identity - and web development that he picked up through a series of job roles in a wide range of industries.
Alex loves design and the conversations and connections it can create for people, especially when applied to data visualization. He often tells people that he started as a data analyst doing ground surveillance in the military because it required him to have a heightened focus on small details.
How did you built your winning Data for a Cause visualization?
To begin, I reviewed the Global Peace Index (GPI) 2018 report and highlights PDFs to get an idea of what data the Institute for Economics and Peace (IEP) analyzes. The data visualizations that they include in their PDFs are top notch, and almost deterred me from even contributing because I thought, “What can I do that would come close to topping this?” But I decided to move forward anyway because something about it was inspiring.
Next, I made notes about key design staples in the GPI reports: fonts, font weights, colors used for different types of charts, and the types of visualizations already being used for particular data. This helped me ensure that what I was going to create would be in line with their existing design standards and be more appealing to use for their marketing.
After that, I thought about what I could add that wouldn’t be just a copy of what currently exist—images. I sketched out a few dashboard ideas on paper, and once I decided on one I went hunting for the images. I searched through Unsplash to find photos that not only stimulated an emotional response, but also included location information. I ended up downloading about 40 images that I felt like I could use and converted them to grayscale in Photoshop to allow the text and visualizations to “pop” a bit more.
Once I had my final dashboard design sketched and picked out, I created the tables, charts, graphs, and maps that I would need using Tableau. After those were completed, I created what I call the “skeleton” of the dashboard in Photoshop. This consisted of me determining which images to use, their order, reducing the size of the images and adding in necessary text and legends. Tableau doesn’t currently support the embedding of non-standard fonts, and since I wanted the final dashboard to look consistent to every user, Photoshop was the route I had to take. Once the skeleton was complete, I inserted it as an image in Tableau, added my sheets as floating objects, and made any necessary adjustments. That’s how I did it (in a nutshell).
What learning resources would you recommend to those who are just getting started?
Everyone has a different learning style. My advice for new users is to be patient and find your style and process, and don’t be afraid to let that style and process evolve as you learn. Also, don’t be afraid to ask questions of those whose work or advice you value. Personally, I have 3-4 people I run visualizations by for feedback before they ever see Tableau Public.
I remember as a new user I often downloaded visualizations that I admired on Tableau Public and reverse engineered them. That is, I looked at how different visualizations were being created and attempted to recreate those in real-world projects I was working on.
Finally, I would also recommend the following books for new users, as they gave me a lot of great insights on data visualization in general: The Functional Art by Alberto Cairo and Data Visualisation by Andy Kirk.
What are your favorite data visualization tools?
Tableau has a firm grip among the business intelligence platforms, and it’s where I spend most of my time when developing data visualizations. However, I believe that Google Data Studio is poised to make a strong entrance in the market when it comes out of beta. From a design and end user usability perspective, I prefer what Google Data Studio offers, but right now I feel like Tableau is the best well-rounded product.
Alex's data visualization (click on the image to view the interactive version):