| Publications | Analyses & Studies | Insults on social media: How affected are users on social media and how do they defend themselves?
bidt Studie Beleidigungen auf Social Media Cover

Insults on social media: How affected are users on social media and how do they defend themselves?

Steliyana Doseva bidt
Fay Carathanassis bidt
Prof. Dr. Hannah Schmid-Petri University of Passau
Prof. Dr. Dirk Heckmann Technical University of Munich

Insults on social media platforms – a new phenomenon or an old social problem in a new guise? A recent bidt study shows the extent to which social media users are affected by insults offline and online and how they react to online attacks. The results show that those affected continue to experience insults more frequently in the offline world than in the online world. When it comes to dealing with insults online, it is clear that almost one in two of those affected have never reported an offensive online post. Lack of interest, legal uncertainty and hopelessness are just some of the reasons why users refrain from reporting offences.

Against the backdrop of the planned law against digital violence, the research team in the project “Challenges of regulating digital communication platforms” at the Bavarian Research Institute for Digital Transformation (bidt) is conducting a current survey to find out whether social media users perceive and experience insults and whether they are aware of and use existing complaints mechanisms.

In June and July 2024, the market research institute Ipsos was commissioned by bidt to survey 5,000 people aged 18 and over living in Germany with internet access who use social media (Instagram, Facebook, TikTok, Snapchat, X and/or LinkedIn) at least once a week.

A special feature of the study – compared to existing surveys – is that it focuses on insults that are clearly not permissible in the legal sense (so-called violations of personal rights). This was specifically defined in the questionnaire1 and enables a much more precise assessment of the phenomenon. Comparable studies often do not differentiate between the various forms of offence and therefore cannot make any statements about the legal assessment and permissibility of the perceived statements. This specific enquiry makes it possible to identify cases that need to be deleted quickly and effectively without the need for lengthy deliberations.

Insults are not just a problem of the online world

Just under 60 percent of respondents (n = 4,289) reported that they had seen someone else being insulted on social media in the last six months.

Klicken Sie auf den unteren Button, um den Inhalt von Datawrapper zu laden.

Datawrapper laden

A good 20 percent (n = 4,434) stated that they themselves had been affected - this is around 30 percent less than the proportion of people affected in the study “Lauter Hass – leiser Rückzug (in German)” from February 2024. One possible explanation for the difference in figures is how respondents understand the term “insult”. In the current survey, respondents were explicitly asked whether they had been insulted in the last six months in a text post on Instagram, Facebook, TikTok, Snapchat, X (formerly Twitter) or LinkedIn by patronising posts in which swear words were used or on the basis of certain characteristics, such as their gender, nationality, ethnicity, religion, sexual orientation or disability.

However, the proportion of insults in the offline world is higher, with over 35 percent of the users surveyed having been insulted themselves.

Klicken Sie auf den unteren Button, um den Inhalt von Datawrapper zu laden.

Datawrapper laden

People in the public eye in particular experience insults on social media. To this end, the survey asked people to indicate the main role they play on social media. The information on this was analysed together with information on whether the persons had already been insulted at least once on social media. The results show that over 80 percent of the politicians surveyed (n = 76) have been subjected to insults online. The figure is over 65 percent for activists (n = 90), just under 50 percent for influencers (n = 148) and just over 60 percent for journalists (n = 48). Volunteers are not spared online insults either: within this group, over 50 percent stated that they had been affected by such content themselves (n = 109). Of the respondents who stated that they appeared on social media as a private individual (n = 3,372), a relatively small group had been subjected to insults: Just over 14 percent here.

The same groups of people were also asked to what extent they had also experienced insults in the offline world. According to their own statements, the respective proportion is even higher here: For politicians (n = 75), the figure is 84 percent, for activists (n = 94) 70 percent, for influencers even almost 70 percent, for journalists (n = 49) 63 percent, for those performing voluntary work (n = 103) 55 percent and finally for private individuals (n = 3,363) almost 43 percent.

Half of offensive posts are reported

Just over half (54 percent) of the respondents directly affected by online insults (n = 890) stated that they had reported the patronising posts directly to the platform or to a government agency.

Klicken Sie auf den unteren Button, um den Inhalt von Datawrapper zu laden.

Datawrapper laden

Of these, most reports (66 percent, n = 465) were only made on the platform, while 24 percent of those affected (n = 465) only contacted government agencies such as the police, public prosecutor's office, court or other authorities.

Klicken Sie auf den unteren Button, um den Inhalt von Datawrapper zu laden.

Datawrapper laden

Reasons for not reporting mainly due to lack of prospects of success

Of the respondents directly affected, almost 46 percent (n = 890) stated that they had not reported the posts to a public or private body (e.g. directly on the platform). Accordingly, only just under 35 percent (n = 484) stated that they had not reported the content in question because they were unsure whether the content was really illegal.

There is also a lack of knowledge regarding existing reporting channels. Almost 44 percent of respondents (n = 455) simply did not find a reporting channel on the platform. In addition, 66 percent (n = 456) stated that a lack of interest on their part played a role in why they did not report what they considered to be an offensive post. Furthermore, a majority of respondents do not see sufficient chances of success because the identity of the offending person is not known (57 percent, n = 460). 57 percent of respondents (n = 476) do not believe that the platform will deal with their report. Over 60 percent (n = 651) also do not expect to receive any help from the state authorities. The latter could also be related to the general trust in state institutions. Respondents were asked to indicate the extent to which they trust state institutions on a scale of 5. Only the police are trusted by more than half of respondents (54 percent, n = 4,721) to a great or very great extent. Trust in other state actors is lower (28 percent, n = 4,699, trust the federal government, 30 percent, n = 4,673, trust the public prosecutor's office, 48 percent, n = 4,659 and 46 percent of respondents, n = 4,629, trust the national courts).

Klicken Sie auf den unteren Button, um den Inhalt von Datawrapper zu laden.

Datawrapper laden

Another reason for not reporting is legal uncertainty for over a third of respondents. They were unsure whether the post actually violated the law or the platform's general terms and conditions. With this in mind, it can be seen that the respondents' level of knowledge on the topic of platform regulation is relatively low. Not even a quarter of respondents (n = 4,878) had heard of the European Digital Services Act (DSA), which provides for reporting channels for illegal content on platforms across Europe, among other things. This figure is 37 percent (n = 4,881) for the law against digital violence, which is still planned at national level.

Klicken Sie auf den unteren Button, um den Inhalt von Datawrapper zu laden.

Datawrapper laden

Conclusion

Ultimately, legislators have to deal with familiar problems that existed before the social media age. Insults are not a problem specific to the online world. As expected, people who are in the public eye are particularly affected. These people need to be better protected from attacks - both online and offline. Otherwise, increasing digital violence could also lead to a decline in the commitment of voluntary and politically active people in society.

However, the special feature of such online attacks is the digital environment. It offers both risks and opportunities when it comes to upholding and enforcing the law. According to the legal obligations in the DSA, effective reporting channels must be available directly on the platforms. However, there is a lack of interest and, not least, trust in these instruments on the part of platform users. In order to build trust in the reporting channels and promote law enforcement, simple, transparent reporting channels should be publicised, general knowledge about them should be communicated and the means of legal prosecution should be improved. In addition, trust in state institutions should be strengthened in general.

Footnote

1 In this survey, insults are understood as condescending statements that consist in particular of insults or serious degradation (e.g. comparison of a person with an animal/an object, reference to sexual activities (with animals) or threats of physical and/or sexual violence up to and including murder), formal insults or swear words (e.g. “bitch” or “asshole”) or attacks on human dignity (e.g. “He likes to f*ck goats best”).