How to Read this Dashboard This project attempts to give an overview of the online political landscape in Germany. It is a live dashboard, which collects and analyzes political content from Twitter, Facebook and 40 online news media sources. The dashboard tries to: 1. Locate and present the trending political topics discussed in Germany online. 2. Provide insights on the attitudes and political interests of online partisan users. Each page of the dashboard is platform specific and contains illustrative plots and graphs. For a detailed discussion on the data collection process, on the creation of the plots, and on the limitations of the dashboard’s methodology, read here. Note: The results presented on this dashboard might not replicate the full online interactions.
Facebook interactions in the last seven days
Overview The following plots present the posts, likes and shares generated on the pages of the seven major German parties and their regional pages in the last seven days.
Reactions To construct this plot, we use the information from the five different user reactions on the pages’ posts. We count the number of times each reaction has been used by the users on each party’s pages in the last seven days. We normalize by the number of reactions collected for each party. We are interested in showing percentages and not absolute numbers. This allows us to compare the importance of a reaction for discussions of each party’s pages. Each color-shape on the plot corresponds to a German party and it intersects the lines corresponding to the hashtags. For one reaction, the dots closer to the outer circle tell us that the users of these party’s pages are proportionally using the reaction more often than the users of the other pages. If all the dots for one reaction are close to the center, then users of all pages are using the reaction in similar quantities.
Mood This bar plot shows the sentiment in the text extracted from the posts that were created by the political parties on Facebook. We use the SentiWS dictionary to evaluate the words and assign them with a sentiment score. We then average the word scores to assign a sentiment to each party. Positive mood corresponds to burgundy and negative mood to light blue.
Word Cloud We present the top nouns used in the posts from the Facebook pages. The size of the word is related to how often it is used across the collected political pages. By browsing over the word, it is possible to see the contribution of each political party to the word size. The colors are the same used in the reactions plot above.
Active Ads on Facebook We present an overview of all German political advertisements on Facebook. We obtained them by using the public Facebook Ads API. More information on the collection and processing of the data can be found here.

Top Advertisers This table presents the funding entities that have the most active advertisement campaigns on Facebook right now.

Top Impressions This table displays the effectiveness of the advertisements placed by the funding political entities as measured by impressions. An impression is counted as the number of times an instance of an ad is on screen for the first time. (Example: If an ad is on screen and someone scrolls down, and then scrolls back up to the same ad, that counts as 1 impression. If an ad is on screen for someone 2 different times in a day, that counts as 2 impressions.) Facebook provides an approximate lower and upper bound of impressions generated because of an ad. The table includes also the spending of the parties for generating the ads.

Targeting Map This map visualizes geographically the parties’ advertisement campaigns on Facebook. It shows the percentage of ads targeting each federal state by party. The intensity of the colour on the map translates to a higher percentage of placed advertisements.
AfD   CDU/CSU  FDP  Gruenen  Linke  SPD