Online Toxicity Around the 2025 German Federal Elections
The 2025 German federal elections on February 23 sparked widespread discussions across social media, with political debates playing out in real-time across major platforms. To analyse the online hate landscape, we monitored the four largest political parties—CDU/CSU, AfD, SPD, and Die Grünen—as well as their leaders—Friedrich Merz, Alice Weidel, Olaf Scholz, and Robert Habeck.
Data for this report was based on 108,000 posts in German, across 8 social media platforms: Reddit, X, 4chan, Gab, YouTube, TikTok, 9Gag and Facebook.
Content warning: Presented data may contain disturbing language related to online hate speech.
A key question driving this analysis was whether online hate was disproportionately directed at certain parties or political figures. To investigate this, we tracked toxicity levels, violent rhetoric, and common toxic keywords, comparing how hostility was distributed among different political actors.
However, it is important to acknowledge the limitations of our methodology. We do not have access to the entire social media landscape, nor do we have the capacity to analyse every post in existence. Instead we use a representative sample of social media discussions. This means that while our findings offer valuable insights into online hate trends, they do not reflect every single instance of political discourse online.
Despite these constraints, the data presents a compelling picture: toxicity levels were remarkably consistent across all parties and leaders, suggesting that online hate in the German election discourse is a widespread issue rather than one that disproportionately affects the parties or leaders of the four largest parties.
Comparing toxicity across political parties
Between February 19 and February 27, 2025, we collected over 76,000 posts discussing the four largest parties. Here’s what the data revealed:
AfD had the highest number of mentions (32,358), followed by CDU/CSU (20,543), while SPD (11,794) and Die Grünen (11,451) had similar discussion volumes.
Twitter (X) was the primary platform for all parties, followed by Facebook. TikTok had a small but notable share, particularly for AfD (8%).
Toxicity levels targeting the main parties were nearly identical with CDU/CSU, SPD, and Die Grünen averaging 0.15, and AfD slightly higher at 0.16.
Posts with extreme toxicity (0.8 or higher) were highest for AfD (166 posts) but still comparable to CDU/CSU (110), SPD (63), and Die Grünen (56).
High-toxicity messages were more frequently found on Reddit for all four major parties.
Mentions of violence were present in 9–12% of messages across all parties, showing no drastic variation.
| CDU/CSU | AfD | SPD | Grünen |
|---|---|---|---|
| 20,543 posts | 32,358 posts | 11,794 posts | 11,451 posts |
| Most data was collected from Twitter (57%), Facebook (36%) and/or Tiktok (4%) | Most data was collected from Twitter (53%), Facebook (37%) and/or Tiktok (8%) | Most data was collected from Twitter (50%), Facebook (41%) and/or Tiktok (5%) | Most data was collected from Twitter (55%), Facebook (41%) and/or Youtube (2%) |
| The average toxicity score is 0.15 | The average toxicity score is 0.16 | The average toxicity score is 0.15 | The average toxicity score is 0.16 |
| 110 posts (0.54%) have a toxicity score of 0.8 or higher | 166 posts (0.51%) have a toxicity score of 0.8 or higher | 63 posts (0.53%) have a toxicity score of 0.8 or higher | 56 posts (0.49%) have a toxicity score of 0.8 or higher |
| About 12% of messages involve Violence | About 9% of messages involve Violence | About 10% of messages involve Violence | About 10% of messages involve Violence |
| Some common toxic keywords include: horden, dreck, fckcdu, vergewaltiger, and fcknzs | Some common toxic keywords include: dreck, faschos, ausländer raus, fcknzs, and dreckspack | Some common toxic keywords include: horden, dreck, multikulti, heil, and neue weltordnung | Some common toxic keywords include: dreck, missbraucht, dreckspack, vergewaltiger, and pädophile |
Comparing toxicity around political leaders
In addition to party-wide discussions, we tracked over 32,000 posts mentioning Friedrich Merz (CDU/CSU), Alice Weidel (AfD), Olaf Scholz (SPD), and Robert Habeck (Die Grünen). Key findings include:
- Merz had the highest number of mentions (13,545 posts), followed by Weidel (8,467), Habeck (6,144), and Scholz (4,665).
- Twitter (X) and Facebook remained the dominant platforms across all leaders, with TikTok engagement highest for Scholz (12%).
- Toxicity scores showed little variation—Merz and Weidel both had 0.15, Scholz had 0.13, and Habeck had 0.14.
- Extremely toxic posts (0.8 or higher) were distributed similarly, with Merz (47), Weidel (28), Scholz (21), and Habeck (22)
- High-toxicity messages were more frequently found on Reddit for Merz, Scholz, and Habeck, while for Weidel, the most toxic discussions were concentrated on 4chan.
- Mentions of violence ranged from 7% (Habeck) to 12% (Merz), again showing no major disparities.
| Merz | Weidel | Scholz | Habeck |
|---|---|---|---|
| 13,545 posts | 8,467 posts. | 4,665 posts | 6,144 posts. |
| Most data was collected from Twitter (51%), Facebook (41%) and/or Tiktok (4%). | Most data was collected from Twitter (76%), Facebook (14%) and/or Tiktok (7%). | Most data was collected from Twitter (53%), Facebook (29%) and/or Tiktok (12%). | Most data was collected from Twitter (60%), Facebook (31%) and/or Youtube (4%). |
| The average toxicity score is 0.15. | The average toxicity score is 0.15. | The average toxicity score is 0.13. | The average toxicity score is 0.14. |
| 47 posts (0.35%) have a toxicity score of 0.8 or higher. | 28 posts (0.33%) have a toxicity score of 0.8 or higher. | 21 posts (0.45%) have a toxicity score of 0.8 or higher. | 22 posts (0.36%) have a toxicity score of 0.8 or higher. |
| About 12% of messages involve Violence | About 8% of messages involve Violence. | About 11% of messages involve Violence. | About 7% of messages involve Violence. |
| Some common toxic keywords include: horden, dreck, alt parteien, neue weltordnung, and fuck. | Some common toxic keywords include: dreck, fckafd, spinner, dreckspack, and killer. | Some common toxic keywords include: fckafd, dreck, sklaverei, fuck, and dussel. | Some common toxic keywords include: nichtskönner, lügenpresse, dreck, fckafd, and killer. |
Key takeaways
Our data does not indicate significant differences in the levels of online hate directed at different parties or politicians. Instead, toxicity levels were consistently distributed across all four major parties and their leaders:
No party or leader stood out as being uniquely targeted—all faced similar levels of online hate, both in frequency and severity.
Platforms like Reddit hosted the most extreme toxicity, though X and Facebook remained the primary hubs for political discourse.
Violent rhetoric was present, targeting all parties and leaders, with only minor variations in how frequently it appeared.