Throughout my summer internship at TeraThink I was fortunate enough to work on a variety of different projects. They exposed me to many different areas within which TeraThink delivers solutions. The final assignment I was tasked with was on a data analytics project. I was lucky enough to work with the Director of TeraThink’s Data Analytics Community of Practice, Beth Bauer.
We focused on creating modernized, interactive visuals to complement the pre-existing ones on the Veterans Affairs (VA) website. The efforts were inspired by the Veterans Affairs Open Data Portal. This portal allows anyone to figure out ways to improve and modernize the VA’s current data visualizations. Tableau was the perfect tool for us to accomplish this.
Going from 0-60 with Tableau
Before beginning this assignment, I had no prior experience with Tableau. However, I luckily was given some time to take courses offered through TeraThink’s partnership with Udemy for Business as well as free training videos on the Tableau website. The Tableau website offers 2-10-minute videos on different topics within Tableau. Therefore, I easily navigated to any topic I wanted to learn more about. For each video there is a Tableau workbook available for download. This way the user can follow along in their own version of Tableau.
I found this interactive approach helpful since I tend to learn best from experiential/hands-on learning. Also, I obtained a free one-year license of Tableau Desktop since I’m currently enrolled at an accredited university (RPI). This was thanks to Tableau’s Academic program.
Beth also connected me with Jose Robles, a TeraThink expert on Tableau. Jose proved to be a vital resource in helping create interactive yet simple visuals while also answering any questions I had about Tableau.
The Current VA.gov Data Set
We chose to modernize the data visualizations for the Top 50 Diagnosed Conditions Among Veteran VHA Patients aged 18-44, 45-64, and 65+. The current format on the VA.GOV site has 3 different visuals and tables for each age group – each contained the percentage of veterans diagnosed per ethnicity. The only variable between graphs was the age range.
Wouldn’t it be more helpful for consumers of this data to have the option to visualize the entire population in one data visualization?
We believed there would be value in creating a single visual which displayed the graphs and supporting data for all three age groups at a glance – especially if a user could interact with it and further segment if desired. We also wanted to see what additional information – via graphs and charts – we could uncover in the data when using a modern visualization tool.
The three data tables we worked with are available for download as Excel files. After this step, we performed everything within Tableau.
First, in Tableau we unionized the original three Excel data sets in order to join the data into one table. Then, we added filters so the user could view the data by the age group, diagnosis, or ethnicity of their choosing. We created a single dashboard with the graph, a conditions table, and these filters. This way we could view all the data on a single page. Lastly, on a separate worksheet, we created an interactive graph with the Top 5 Diagnosis by Ethnicity. This way the user could select any ethnicity and see what the top 5 most diagnosed conditions are for that specific ethnicity.
The ease of which Tableau simplified this data was super impressive. Our hope in submitting our modernized data visuals is that this data can help to more clearly identify trends in veterans’ conditions.
Jumping into a project like this significantly accelerated my learning of Tableau and data visualization in general. I would recommend Tableau to anyone looking to display interactive data in a visually appealing and simple way.
A Powerful Tool
I believe Tableau can be a very effective tool to modernize visuals. Those visuals create a more efficient way for people to share and interact with data. Personally, I plan to continue to use Tableau in the future whether it be as a student or any other facet of life to better understand the data I am working with.