Beautiful Clean Coal
WPI
I developed a data-driven narrative about the United State's energy production over the past 60 years.
For our data-visualization course at Worcester PolyTechnic Institute, my friends and I determined that we wanted to go beyond making a fancy dashboard and present a unique project narrating learnings revealed by our dataset. We decided we would make our project relevant to current events, ensure it would have beautiful interactive charts, and finally inform users to make an unbiased decision about a topic. We drew many of our inspirations from www.pudding.cool.
Timeline
1.5 Week Build
My Role
Front End Engineer
Tools
Team
Derek Feehrer
Michael Moukarzel
Themes
Data-Visualization | Data Driven Narrative
The Problem
My team and I were given an assignment to create a set of data visualizations as the final project for our data visualization course. We were tasked with choosing a dataset and then using any framework/library of our choice to create 3+ visualizations.
The Solution
My team and I decided to draw upon storytelling to create a data driven narrative about energy production in the United States over the past 60 years. We centralizing our insights around the "War on Coal" and asked the question of whether or not coal can trully be clean and how it's history has compared to other forms of energy production in the United States.
Goals
Pick a relevant topic to tell a story about
Choose the right tools for the job
Provide the readers with data to develop their own opinions
My Role
My primary role on the 3 person team was creating the interactive visualizations for the user to explore the topic. We had 1 other frontend engineer focusing on the user experience and another engineer working on data munging[1].
1. Munging - "Data wrangling, sometimes referred to as data munging, is the process of transforming and mapping data from one 'raw' data form into another format with the intent of making it more appropriate and valuable for a variety of downstream purposes such as analytics."
Development Process
01. - Choosing a Topic and Dataset
For this project we knew that it would be sufficent enough to make a dashboard with some fancy interactive elements, however we wanted to change it up a little and use this assignment as an opportunity to educate readers on a topic. Therefore, we agreed to focus our final project around a current event and brainstormed several possible topics we could write about. One of the most important criteria for the topic was the amount of available data for it. After 2 days of deliberation, knowing we only had about a week and a half to complete the project, we determined that our project would be focused around coal's role in energy production in the United States. One of the deciding factors was the ample amount resources about the US's energy production located at the U.S. Energy Information Administration (EIA).
02. - What the Data Told Us?
There was lots of data to go through, so we decided that we had to narrow our scope to specific details about energy production. We determined that we would use 2 datasets, one highlighting how coal was used compared to other sources of energy and the other to determine where each energy producing factory was located and how much energy they produced. This would give us a historical view of how coal was used and the trends production for energy generating facilities. After our initial analysis, it was clear that the United States saw an early boom in coal energy production during the early 20th century. However, as the 20th century progressed there was a larger and larger shift towards renewable energy and eventually to natural gas.
03. - Design It & Determine the Data Visualizations
To narrate the history of energy production we brainstormed several potential visualization ideas. These are the designs we decide to move forward with:
Use a stacked area chart to display the 60 years of data production available to us and annotate important key events that changed how energy was produced moving forward.
A map of the United States that marks the locations of factories while coloring and scaling those markers depending on on the type of factory producing energy and how much energy it produces for a specific year. In addition, this map would be connected to a line graph showing energy production totals as a sum of the energy produced by factories accross a date range.
A line graph displaying coal miner employemnt rates across the past 60 years.
The different ways in which you can burn coal in a factory to create electricity and how that compares to pollution from other energy sources.
04. - Develop It
With our data and designs ready there was only one thing left to do, develop the webpage. To do this we had to break down the expreience into 3 main components the data, the user experience for the page, and the more specfic user expeerience for each data visualization. First, the data was in an unruly format and we needed to isolate it and put it into a format better suited for our graphs. Next, we needed to create the user experience each of the supporting visualizations. To create the each graph we used Semiotic, a data-visualization library created by Elijah Meeks. Finally, we needed to combine everything into a seemless experience for readers.
05. - Present It
Within in couple days we had the project completed and launched it on Github pages! Feel free to check it out!
Check it Out!