Home > News > Germany’s far-right party AfD won the Facebook battle. By a lot.
181 views 8 min 0 Comment

Germany’s far-right party AfD won the Facebook battle. By a lot.

- September 26, 2017
Supporters of the hard-right Alternative for Germany (AfD) party shout slogans during a September campaign rally for German Chancellor Angela Merkel in Torgau, Germany. (Reuters)

For the first time since the early days of the Federal Republic of Germany, a party on the far right, the AfD (Alternative for Germany), has entered the Bundestag. AfD’s strong social media presence appears to have helped it win 12.6 percent of seats in Parliament seats, as some observers have noted.

To understand whether that impression might be true, we undertook an empirical analysis of AfD’s social media use. We began by examining how all the German political parties used Facebook in 2017. We learned that the AfD publishes more content, gets more likes and engages more people than any other parties. Similarly, its posts are shared more widely than all the others combined.

[interstitial_link url=”https://www.washingtonpost.com/news/monkey-cage/wp/2017/09/25/the-anti-muslim-afd-just-scored-big-in-germanys-election-what-does-this-mean-for-german-muslims/”]The anti-Muslim AfD just scored big in Germany’s election. What does that mean for German Muslims?[/interstitial_link]

AfD gets more “likes” than other parties

You can see this activity measured in the figure below, which shows the per post average number of likes users made on posts by the AfD and its three largest rivals. Since January, AfD’s Facebook posts averaged around 3,500 likes per post, almost double the nearest rival the CSU, which received 1,900 on average. In addition, AfD consistently got more likes per post than any other party.

Weekly average number of likes on Facebook posts by each party (Facebook Graph API)

AfD posts a high volume of sensational material that gets strong emotional reactions

How did this work? The AfD built up a large following on both Facebook and Twitter  by sharing a high volume of highly sensationalist tweets and posts, which other social media users reacted to emotionally. For example, shortly after the terrorist attacks in Barcelona last month, the AfD posted a picture of bloody tire marks with the headline:

“Frau Merkel, die Opfer Ihrer politischen Amok-Fahrt sind nicht vergessen! Doch wie viele müssen noch sterben, bevor Sie verstehen?” (Translation: “Mrs. Merkel, the victims of your political rampage are not forgotten! But how many have to die before you understand?”)

It published two similar posts with different headlines, both implying that the terrorist attacks are a direct consequence of Chancellor Angela Merkel’s refugee policies. These posts received 6,880 likes from Facebook users; the response by the center-left Social Democrats (SPD) received 973.

[interstitial_link url=”https://www.washingtonpost.com/news/monkey-cage/wp/2017/09/25/building-a-new-german-coalition-government-wont-be-easy-heres-why”]Building a new German coalition government won’t be easy. Here’s why.[/interstitial_link]

Other indicators point in a similar direction. Before the polls closed at 6 p.m., more than 20,000 users shared an AfD public post indicating they had voted for the AfD; only around a 10th of the number did the same with the SPD’s post.

AfD has a larger number of social media followers who interact more often

Could all these likes come from a small number of users who are liking all of the AfD’s content? We used each individual’s unique identifier to examine the cumulative number of unique users who liked posts on each party, to see the overall number of people each party engaged over the period.

It turns out that both are true: AfD’s users are more active and also more numerous than those of other parties. In the figure below, we can see that posts on the AfD page were liked by almost a quarter-million unique Facebook users since the beginning of 2017, more than any other party. The party engaged around one-fifth more users than its nearest rival, the CSU. AfD’s users were three times as active as its rivals. We cannot guarantee that all of these likes have come from the AfD’s supporters. When we count likes, we count all responses, including “angry” emoji, which may well have come from opponents angry at the party. However, these results show that the party has been more successful at engaging the public on social media than its rivals.

Cumulative number of unique users who liked each party’s posts. (Facebook graph API)

AfD’s posts are shared more often than its rivals

When individual users share a party’s post in their timelines, they make it potentially visible to their friends and followers who have not themselves followed that party. That happened with AfD’s content much more frequently than with other parties, particularly in the months preceding the election. Because AfD’s posts were more widely shared, its content had the potential to spread more widely through the Facebook network than that of the mainstream parties.

[interstitial_link url=”https://www.washingtonpost.com/news/monkey-cage/wp/2017/09/24/behind-the-boring-german-election-are-four-deeply-disturbing-trends”]Behind the ‘boring’ German election are these 4 deeply disturbing trends[/interstitial_link]

In the figure below, we see that AfD’s posts were shared over 800,000 times during the year, more than the posts of all the other major parties combined.

Cumulative number of shares of each party’s posts. (Facebook graph API)

AfD has used social media very successfully

Overall, we find that the AfD has been extremely successful in using social media to spread its message broadly. Of course we cannot directly or precisely assess how much this contributed to its electoral success. But we can see that the AfD has used social media to build a large virtual grass-roots base of support and has captured many German social media users’ imaginations.

More generally, we see how important social media can be for challenger parties that want to mobilize supporters but lack the established groups’ more conventional resources.

Thomas Davidson is a graduate student in the department of sociology at Cornell University. He uses computational methods including social network analysis, natural language processing, and machine learning to study social movements and political parties. Find him on Twitter @thomasrdavidson.  

Julius Lagodny is a graduate student in the department of government at Cornell University. He works on political behavior and public opinion of immigrants in Germany and the United States using a mix of computational methods and survey experiments. Find him on Twitter @JuliusLagodny.