How to Read this Dashboard
This project attempts to give an overview of the online political landscape until the 2020 US Presidential Elections. It is a live dashboard, which collects and analyzes political content from Facebook, Tik Tok, and 45 online news media sources. For a detailed discussion on the data collection process, on the creation of the plots, and the limitations of the dashboard’s methodology, read here. Note: The results presented on this dashboard might not replicate the full online interactions.
This page provides a detailed analysis of political news from 45 online news media sources and the number of interactions they generate on Facebook. We mostly exclude non US political news. We follow the classification from allsides.com to assign each outlet to a political orientation. If you have any inquries regarding the data or seek further insights contact us here.
Trending Topics
From all the articles, this plot shows the top seven topics media wrote about in the last 24 hours. A topic is a set of content-significant words, as generated by a machine learning algorithm. Each bubble represents a topic. The size of a bubble corresponds to the proportion of the articles referring to a topic. By clicking on a bubble, you can see the eight most significant words for a topic and the relevance of each word per topic. The word in bold for each topic is the most relevant one. For more information about the online media sources we considered and the creation of the plot see here.
Candidate Attention
This plot shows how many times the collected news articles mentioned the main political candidates in the last 24 hours.
Top Articles on Facebook
The table shows the 50 news articles on Facebook with the most user interactions. Due to API limitations, we only track news published in the last five days. The table shows how many interactions the news article has received. According to Crowdtangle, this includes both public and private interactions.
Topics Comparison between Media Sources
Similarly to the trending topics plot, this spider plot shows the top 7 topics in the last 24 hours (Only the top word per topic is presented). It further shows the proportion of articles that media sources create on these topics. We categorize news sources in the database according to their political orientation: left, left-center, center, center-right and right. Each color-shape on the plot corresponds to media sources of a given political orientation and it intersects the lines corresponding to the topics. For one topic, the dots closer to the outer circle tell us that the media sources of that political orientation are proportionally sharing articles related to the topic more often than the media sources of other political orientation. If all dots for one topic are close to the center, all news sources have articles related to it in similar quantities. Note: The topic proportions are not normalized, as there are further topics media talked about.
Total and Weighted Impressions by Source
The following plots show a comparison between news sources on the number of interactions on Facebook. We include only political news storiespublished in the last five days. Some news outlets publish more political news than others and may have a higher “total interaction” number than others. For this reason, we also include “weighted interactions”, which is simply the total number of interactions divided by the number of published news by the news outlet. We also include plots for aggregated interactions given the five considered political orientations.
