Project description:

Swing States project was born as part of the Trumpland media special reports covering the USA elections. The goal was to show the swing states from a different angle and bring in the Covid-19 element. More and more people were showing interest in voting early either via email, in person or online. Because of that, most of the states had to adapt their regulations. The final result was a clean dataset and visualisations that were providing different insights to the users for them to reach a final conclusion or identify patterns.

Impact reached:

By analysing, cleaning and comparing data from several sources, we identify a pattern during the last three presidential elections. By comparing each state’s law, the number of early voters during past elections and the final results, we were able to identify that a lot of Americans had already voted two weeks before the Election date and who were more likely to vote early (democrats or republicans). 

The impact of the project was to provide, in real time, an update on the number of people who already voted, to which party and identify the number of eligible voters who could already participate in changing the overall picture. It was the first time where data regarding early voters could potentially be used to determine whether a state would turn blue or red. A week before the Election’s date, some swing states were already showing the americans’ preference and providing some first insights on how the voting process had to rapidly be adapted to a new situation instead of postponing the elections. 

Techniques/technologies used:

Data visualisations were designed both in Illustrator and Photoshop. The web editor used was Atom and in order to control the scroll states the chosen tool was Scroll States. For other visualisations, we also used DataWrapper to bring this element of interactivity. The last one was specifically because we had tons of data and we didn’t want to make any conclusion for the reader. Because of that, we cleaned the dataset from several sources and we presented it as it was, in a visual manner, for the reader to filer by party, elections, type of vote, etc and reach a specific conclusion or insight by themselves. 

What was the hardest part of this project?

Data from early voters was available in several sources, as well as data from eligible american voters (meaning the americans who said that wanted to vote before a deadline). Once this data was cleaned, it had to be adjusted to the total population, all this for the last three elections. On top of that, we also had to merge this with the final results by states to identify if there was an interesting pattern to highlight. All this work was done through Excel tables and forms until a clear picture was reached before moving to Illustrator or Datawrapper.

What can others learn from this project?

On a reportering note, the ability to relate two newsworthy topics (such as the american elections and the Covid pandemic) without having to go to the most obvious element (such as the number of covid cases). 

 

On a creativity note, the ability to combine several data visualisation tools depending on the elements of interactions and data gathered. While static infographics are less manejable, these also highlight on a quick overview what needs to be transmitted. SImilar to short calls to actions. For interactive graphics, it makes more sense when there’s tons of data and, instead of providing a final statement, you let the user play and decide. The audience has enough knowledge to learn and understand before making an assumption. 

Project links:

trumplandmedia.com/especiales//analisis-swing-states/index.html