Bullying is a major public health issue. It is estimated that 32% of children globally were victims of bullying at least once a month, and 7% will have experienced persistent bullying over six or more days in the same time period (UNESCO, 2019).
Through the data insights gleaned from our Shout text service, we have a unique opportunity to understand bullying in young people. To date, we have held almost 20,000 conversations where bullying was an issue.
We have now conducted an analysis of 10,000 anonymised conversations about bullying using thematic analysis and natural language processing computational approaches.
Analysis of data from Shout conversations presents us with a unique opportunity to examine young people’s perspectives and experiences with bullying. We found that young people’s views tend to reflect the impact and consequences bullying has had on them, in addition to the feelings and emotions they’ve experiences, rather than talking about bullying in the same way as the definitions used in research.
What is perceived as bullying?
Bullying is often defined as when a person is repeatedly exposed to aggressive or negative actions by another person whose intention is to cause physical harm, psychological distress, or humiliation (Olweus, 1993). Types of bullying range from physical, relational, emotional, verbal, and cyber. The field struggles to agree on a single definition of bullying, but most researchers agree that defining features include power differential, intended harm, and repetition (UNESCO, 2020). Nonetheless, inconsistencies in the definition of bullying can significantly impact our ability to measure its prevalence, with some studies reporting prevalence estimates of face-to-face bullying that range from 10% to 90% (Modecki et al., 2014, Zych, Baldry & Farrington, 2017).
Surprisingly little is known about young people’s perceptions of what constitutes bullying, particularly amongst those aged 13 or under. Importantly, some studies suggest that young people’s perceptions are inconsistent with widely accepted research definitions of bullying (Byrne et al., 2016). For example, young people may consider a single incident as bullying, which is at odds with the commonly held defining feature of repetitiveness (Hellstrom, Persson & Hagquist, 2015). It seems important that young people’s perceptions of bullying be prioritised to ensure that researchers and clinicians alike be better informed about its impact on wellbeing and what support is required to best safeguard young people.
What we’ve learned about bullying from the Shout text service
- Shout has held almost 20,000 conversations with texters about bullying
- Almost three-quarters of conversations about bullying are with texters aged under 18
- The majority of these conversations refer to bullying at school
- Although online bullying can be a factor, texters are predominantly concerned about in-person bullying
We found that in the majority of these conversations, texters were being bullied by schoolmates, and they were predominantly concerned with in-person bullying. Despite the growing concerns in society about online bullying, it was rarer for texters to mention online bullying, although this is clearly a factor in some cases.
This is, in fact, consistent with a survey conducted by the Department for Education in 2018 which found that bullying still predominantly occurs within a school context, and a meta-analysis of 80 research studies showing that traditional forms of bullying remain twice as common as online bullying. This may explain why we saw a significant decrease in texters experiencing bullying during the pandemic school closures.
Some people are at higher risk of being bullied due to their physical appearance, gender identity, sexual orientation, disability, race and religion. People who identified as gay, lesbian or bisexual are more likely to be bullied than those who identified as straight (Berlan et al., 2010), and students with disabilities were more likely to be bullied than those without disabilities (Rose et al., 2015). A large meta-analysis showed that in a sample of over 55,000 young people, those who were overweight were more likely to be bullied. These findings highlight the importance of targeted interventions to support vulnerable populations.
Consistent with these findings, in our analyses, texters describe bullying as vocal offences, teasing, taunts, threats, insults and harassment and some of these texters tell us that they are bullied about their sexual orientation or about physical attributes, such as ethnicity, disability, weight, height and facial features. In the context of name calling, texters most often reference sexual, racial and gender themes, and that the term “taunt” most often echoes some of these gender and sexual offences. We also found that the word “tease” was most often associated with weight and body image-related contexts. To make matters worse, vocal intimidations are often accompanied by physical threats to be beaten up or, even, killed.
In rarer cases does bullying actually become physical for our texters. Topic modelling results revealed that in only around 5% of conversations were threats and police involvement the dominant topic. Texters often describe physical bullying as bodily harm towards themselves or as destruction to their property. And analysis conducted by natural language processing of physical bullying-related terms suggests that physical violence often takes place in a school environment (i.e. corridor, toilet, stairs). Nonetheless, regardless of whether bullying actually becomes physical, the threat of physical violence can be just as damaging to our mental health as the act of physical violence itself.
The links between adolescent bullying and mental health
Being bullied in childhood and adolescence has been consistently linked to mental health struggles with depression and post-traumatic stress disorder, as well as poor physical health, social and educational outcomes (Carney, 2000; Hawker and Boulton, 2000; Nansel et al., 2001). In fact, bullying appears to be one of the most tractable risk factors for depression, with research showing that over a quarter of cases with people aged 18 who suffer from depression might have been attributable to being bullied in early adolescence (Bowes et al., 2015). In more severe cases, bullying victimisation in adolescence has also been associated with an increased risk for suicidal ideation and suicide attempts (Geoffrey et al., 2016).
The texters who most often talked to us about bullying were aged 13 or under (32% of all texters), followed by those aged 14 to 17 years old (41%). Furthermore, the likelihood of these conversations also involving struggles with anxiety and stress (32%), relationships (32%), isolation and loneliness (23%) and self-harm (20%) was often higher when compared to other conversations.
How are our texters feeling?
Most texters describe a great sense of isolation as a result of being bullied and often identify bullying as the catalyst for their low self-esteem and negative body image. Where bullying is one of a few presenting issues, it is most often the main presenting issue. Our findings suggest that the two most dominant topics were associated with mental health and needing help (89%), and that most frequently used phrases included references to feeling low, anxiety, self-harm and suicide. This further highlights the potential for bullying to have a severe impact on mental health. Effective support for those who experience bullying is essential to reduce and prevent long-lasting impacts.
Seeking support can be challenging because bullying often seems to involve texters’ social circles and friendships. Some texters were bullied by exclusion from friendship groups and group activities, or had endured hazing (a test or task involving harassment, abuse or humiliation which is used as a way to initiate a person into a group) in order to maintain their friendships. Others had rumours spread about them, which served to sabotage their reputation and friendships. The social damage and distrust caused by bullying appeared to result in a significant loss of texters’ primary support networks.
These findings bridge an important gap in the field between definitions and experiences, highlighting the importance of incorporating young people’s views into practice to improve the quality of support delivered.