Method

Transparency is a valuable tool in ensuring that we distribute not only accurate information but also information that enables others to make informed decisions about difficult topics. We share every step of the process for how we acquired the data, how we organized it, what the graphs are made of, and what the goals are behind this project.

Step 1: Data Collection

The first step in this project was to find and collect data on the most popular news sources from various political spectrums. In order to decide what news sources were representing what political spectrum, we used existing data from AllSides on the degrees of bias in popular media.

The next step was to create a spreadsheet and fill out a table with information on every single media source for how many YouTube subscribers they have, what their total view count is on a YouTube channel, as well as how many followers they have on Twitter. This represents a diverse online presence, however, we understand the limitations it has by failing to represent the large popularity of cable news channels, subscription-based news channels, and other forms of media.

Once all of that data was present, I created an equation to represent each factor in a way that gives credit for popularity in some sources while also accurately displaying popularity in others. The equation for popularity gives 45% weight to YouTube subscribers, 15% weight to YouTube views, and 40% weight to Twitter followers. This disproportionately prefers YouTube as a platform, but that is due to the popularity and reliability of YouTube over Twitter, and the larger active user base on YouTube.

Step 2: Data Visualization

The second step of the project was to code a graph or chart to show this data in a way that accurately and fairly represented the reality of the situation. I utilized the d3.js library for this and through experiences with HTML, JavaScript, and CSS as well as the use of AI tools, I created two different graphs.

The first graph represents each media source based on its position on the graph, while the second is only based on its x-position and the scale of the bubble. All of the code for these charts is publicly available on GitHub and is linked for anyone to review.

(P.S. if you find anything I'm doing incorrectly please feel free to reach out with suggestions on improvements)

Step 3: Web Design

The final step in this project was to design a website to show the graphs, methods, and implications of this research. We host the website on Hostinger and used their website builder to design the aesthetics. The actual graphs were completely custom coded however and were embedded within the web pages.

We wrote the method, implications, and home page descriptions in an unbiased way and we hope it accurately represents the data we collected. If there are any flaws with this method, feel free to reach out and share ways that this project can be improved to remain an unbiased and open-source collection of data.

Thanks for taking the time to read through this!