Cyberbullying in the UK: The Effect of Global Crises on the Victimization Rates

ABSTRACT

Previous research has shown that the utilization of electronic devices with internet access has increased rapidly over the past years. With that increase, comes the increased victimization of internet users from cyberbullies. However, we do not know to what level that increase affected the cyberbullying rates in the UK. The current study assessed whether cyber-bullying incidents had increased in the UK during the lockdown period, mainly because of the skyrocketing of the social media use and other online platforms. Overall, the results of this study indicated that the prevalence of cyber-victimization and perpetration declined during the lockdowns when compared to pre-March 2020 rates, and cyber-bullying bystander behaviors increased during the pandemic. Further research is in need to examine the latter assumption. Implications and limitations are discussed.

KEYWORDS:

• Cyber-bullying
• social media
• lockdown period
• aggression
• bystander behavior
• online harassment

Introduction

The term bullying has been defined by researchers as the repeated exposure to the negative actions, such as harassment or abuse, of another person (Sanders, 2004). Traditionally, researchers state this abuse occurs over a period of time (Gaffney et al., 2019; Sanders, 2004). More recently researchers have included single harmful attacks such as actual bodily harm or sexual assault within their definitions (Finkelhor et al., 2012; Hellström et al., 2015; Tzani-Pepelasi, 2018a). There are many different types of bullying, including physical bullying (physically attacking the victim), verbal bullying (verbally harassing the victim), and relational bullying (socially excluding and/or spreading rumors about the victim). Both physical and verbal bullying are categorized as direct confrontations since they involve overt displays of aggression toward the victim (Björkqvist et al., 1992; Wang et al., 2009). In contrast, relational bullying involves the manipulation of victims and the purposeful causation of damage to their peer relationships, such as excluding the victim from their group (Bauman & Del Rio, 2006; Björkqvist et al., 1992). For this reason, Wang et al. (2009) suggested that relational forms of bullying can be categorized as indirect confrontation.

Similarly, cyber-bullying can be defined as “the use of information technology to bully a person by sending or posting text or images of an intimidating or threatening nature” (Oxford English Dictionary, 2021), which often occurs in a repetitive manner (). As the bullying is conducted online, cyber-bullying could also be categorized as indirect confrontation. But it is distinguished from relational in-person bullying through the medium used by the bully to harm the victim. The Office for National Statistics (ONS, 2020) estimated that around one fifth of children aged 10 to 15 years of age experienced online bullying between March 2019 and March 2020. These behaviors included: spreading rumors, directly sending harmful messages, name-calling and swearing, and posting hurtful messages about the victim for others to see (ONS, 2020). According to a YouGov (2019) poll, cyber-bullying is also common among young adults (aged 18–24 years old), with harassment being the most common form of cyber-bullying experienced. The ONS (2020) also reports that victimization by cyber-bullying is more prevalent amongst those who are disabled (28%) than those who are not (18%). Like the prevalence rates for in-person bullying, rates for those whose ethnicities were defined as White (21%), Black or Black British (18%), and other ethnicities (19%) were higher than for those who defined themselves as Asian or Asian British (6%). Despite girls experiencing more in-person bullying than boys, the same cannot be said for cyber-bullying, as the ONS (2020) reported that there is no difference in the prevalence of cyber-bullying between genders.

Reasons for cyber-bullying

One of the main theories behind cyber-bullying is the online disinhibition effect, the theory that the anonymity provided by the Internet allows an individual to act in a manner which is unrestrained, and this implies a reduction in concern about one’s self-representation (Udris, 2014). According to Suler (2004), the online disinhibition effect can be divided into two main types of disinhibited behaviors: toxic and benign. Whilst benign disinhibition indicates behaviors such as oversharing of emotions, fears, and unusual showings of generosity and kindness (Cheung et al., 2016), it is toxic disinhibition which relates more closely to the behaviors portrayed in cyber-bullying. This is because toxic disinhibition refers to more harmful behaviors such as threats, rude language and anger that is displayed by perpetrators of cyber-bullying (Cheung et al., 2016).

Lowry et al. (2016) proposed that alongside online disinhibition, deindividuation can help to explain cyber-bullying perpetration. The theory of deindividuation posits that individuals will lose their self-awareness in a crowd, and this makes them more likely to act in an anti-normative way (Postmes, 2001). Lowry et al. (2016) argued that the feeling of anonymity provided by technology perpetuates disinhibition, as anonymity adds to the loss of self-awareness when in a group mentality. Such perpetuation then leads to an alteration in social learning which makes the perpetrator become more willing to engage in cyber-bullying behaviors. According to Willard (2004), there are three main reasons behind cyber-bullying perpetration: The perpetrator does not harm their victim in person and so does not realize the repercussions of their harmful behavior. Cyber-bullying is becoming more frequent, making it more socially acceptable to perpetrate, and the perpetrator incorrectly perceives they have more privacy and anonymity when bullying online compared to in-person bullying (Tzani-Pepelasi, 2018a). However, some academics believe that cyber-bullying can be caused by previous engagement with in-person bullying. Cassidy et al. (2009) proposed that cyber-bullying is often a continuation of in-person bullying which began at school and has continued as cyber-bullying once at home. Whilst this is logical, given the close nature between in-person relational bullying and the types of harassment that occur in cyber-bullying, it does contradict the idea the perpetrator would not realize the consequences of their online behavior.

Predictors of cyber-bullying

As with in-person bullying, self-esteem is an important predictor for cyber-bullying. Brewer and Kerslake (2015) found that perpetration could be predicted by the individual’s levels of self-esteem and empathy, and cyber-bullying victimization could be predicted by the amount of loneliness felt. Consistent with the finding that victimization can be predicted by loneliness, Sahin (2012) found that loneliness is a common predictor for cyber-bullying victimization amongst secondary school children. Whilst both, Sahin (2012) and Brewer and Kerslake (2015) found no relationship between levels of perpetration and loneliness, Srabstin and Piazza (2008) suggested individuals who cyber-bully will often rely on social interaction with others, implying that perpetration could be extraneously linked to loneliness. As well as dispositional factors such as loneliness, cyber-bullying can also be predicted by previous in-person bullying encounters. In comparison with Cassidy et al. (2009), nearly three out of four children (72%) who had experienced in-person bullying at school between March 2019 and March 2020 also experienced cyber-bullying (ONS, 2020). This indicates that a previous history of in-person bullying can be used to predict the prevalence of cyber-bullying.

In addition, cyber-bullying bystanders can impact the prevalence rates of cyber-bullying. Bystanders are individuals who are present and observe bullying incidents (Thornberg & Jungert, 2013), with cyber-bullying bystanders being individuals who observe harmful cyber-bullying interactions online. There are three main types of bullying bystanders: the reinforcer, the defender, and the passive bystander (Sarmiento et al., 2019). According to Sarmiento et al. (2019), the reinforcer bystander acts in a supporting role to the bullying that they are observing. In cyber-bullying, this could be sharing a harmful post that the perpetrator has uploaded about the victim. In contrast, the defensive bystander will often try to intervene on the victim’s behalf, such as commenting on a post and asking the bully to take it down. Research exploring this has highlighted that the more responsibility an individual feels that they themselves and other users have when witnessing online harassment, they are more likely they are to intervene and act as the defensive bystander (Butler et al., 2022). However, if the individual feels the responsibility of intervening in online harassment and bullying is attributed to law enforcement, the more likely it is the individual will act as a passive bystander and perform no response (Barlett et al., 2021). Passive bystanders are individuals who simply observe the bullying behavior but choose not to get involved. In cyber-bullying, this could involve scrolling past the post without reacting to it. A potential explanation as to why individuals may choose to be passive and not intervene when witnessing bullying is the notion that they believe that the responsibility of intervening with incidents of online harassment are assigned mostly to the online platforms themselves. This is a reasonable association to make considering the claims made by online communication platforms regarding their strong safeguarding claims against hateful content and discussions on their platforms (Williams et al., 2023). This reaction of being passive and not responding to witnessed bullying is one of the most predominant behaviors exhibited by online bystanders, with Barlett et al. (2021) finding it to be the chosen response for over a third of their participant sample. Defensive bystanders are the least frequent type of cyber-bullying bystander (Sarmiento et al., 2019), with passive and reinforcing bystanders being the most prevalent. As Salmivalli et al. (2011) reported, bystanders who reinforce bullying can increase the prevalence of instances of bullying, potentially suggesting that bystanders can predict the frequency of cyber-bullying attacks.

The 2019 global crisis and cyber-bullying

Social interactions have been damaged considerably since March 2020. Long et al. (2021) found that weaker friendships disappeared once in-person social interaction was no longer possible; this made friendship groups homogenous and smaller in size. This had a devastating impact on individuals’ ability to gain social support which is both outside of their immediate support network and varied in opinions and advice. Long et al. (2021) also state that UK national lockdowns have negatively affected the ability to display in-person intimate behaviors (e.g., physical touch) which can cause negative consequences for a person’s emotional well-being. Furthermore, a lack of in-person social interactions has also increased the amount of loneliness felt during lockdowns, especially by those who live alone. Killgore et al. (2020) found that loneliness rose to 43% during periods of lockdown in the US; this was further associated with higher rates of depression and suicidal ideation. The same pattern emerged during the UK national lockdowns as Groarke et al. (2020)found a rise in loneliness to 27% during the first national lockdown period (March 2020 – June 2020). Like in the US, increased loneliness in the UK was related to a greater risk of mental health disorders such as anxiety, depression, and suicidal ideation. Alongside an increase in damaged social interactions and loneliness, UK COVID-19 lockdowns have also had negative consequences for social media usage and levels of self-esteem. A report by Ofcom (2020) reported an increase of around 2–3 million users during March and April 2020 for the social media sites Tiktok, Twitter, Houseparty, and WhatsApp. Such increases in social media usage were attributed to higher levels of low self-esteem and body dissatisfaction during the COVID-19 lockdowns too (Vall-Roqué et al., 2021). This suggests that individuals who were restricted in their movements during the national lockdowns had more time to dwell on negative thoughts about their own self-image. Overall, it is clear that some of the main predictors for cyber-bullying perpetration and victimization have increased during the UK-wide lockdowns.

Due to the increase in these predictors, it is fair to assume that the prevalence of cyber-bullying might rise as a result. Similar trends have been found in other countries. For example, Barlett et al. (2021) found an increase in cyber-bullying perpetration in individuals who had experienced COVID-19 infection within their immediate social network: they attributed to an increase in overall stress. Similarly, a Florida based study on the prevalence of cyber-bullying tweets and the COVID-19 pandemic found a substantial increase in the prevalence of cyber-bullying during the beginning of the pandemic, especially during March 2020 (Karmakar & Das, 2021). This finding is relevant considering the rise in Twitter usage during March 2020 reported by Ofcom (2020). Similarly, Jain et al. (2020) reported an increase in COVID-19 related cyber-bullying during the national COVID-19 lockdowns in India. Jain et al. (2020) found that, as in the USA, an increase in social media usage was a key predictor of the increased prevalence of cyber-bullying during national lockdowns. A specific increase in cyber-stalking (as a form of cyber-bullying) during the pandemic was also reported. This is notable considering that in-person stalking was unattainable during lockdowns.

Aims

In response to the reports that the prevalence of cyber-bullying increased during national lockdowns in 2020, the current study aimed to assess whether there was a direct increase in cyber-bullying during the UK national lockdowns. The present study also aimed to explore the prevalence of cyber-bullying perpetration, victimization, and the cyber-bullying bystander behavior.

Method

Participants

This study utilized a within-subjects design with participants completing questions regarding their cyber-bullying perpetration, victimization, and bystander experiences. The sample (N = 230) consisted of both male (N = 42, 18.2%) and female (N = 188, 81.7%) volunteers, whose ages ranged from 18 to 73 years (M = 24.6, SD = 10.2). Participation in this research conformed to the ethical requirements. To be eligible to participate in this study, participants needed to be over 18, currently residing in the United Kingdom, and have access to an electronic device. The most frequent ethnicity was White British/Irish (65.6%) whereas the least common ethnicity was Black African/British/Caribbean (4%). Overall, the most frequent ethnicity by gender type was White British/Irish women (48.2%) whilst the least frequent ethnicity by gender type was Black African/British/Caribbean men (1.3%).

Procedure and materials

The questionnaire was disseminated online via a link, and participants completed the questions anonymously. The questionnaire was available, and data was collected throughout the covid lockdown period that occurred between January 2021 to July 2021. This was the third lockdown the UK administered, therefore it is possible that participants may have related their answers to a combination of both their experiences to the most recent lockdown at the time and to other previous lockdowns. Prior to completing the cyber-bullying measurement scales, participants completed 10 questions related to their background which covered topics such as age, gender and social media and electronic device usage.

Cyber-bullying questionnaire

The cyber-bullying questionnaire utilized was an adapted version of the “Cyber-bullying and Online Aggression Questionnaire” (Patchin & Hinduja 2006; Hamburger et al., 2011) used by. This questionnaire included 25 of the original questions plus 16 new COVID-19 lockdown related statements and was spread across two sections: cyber-bullying perpetration and cyber-bullying victimization. The questionnaire involved a variety of question types including a 5-point Likert-scale for perpetration frequency assessment and binary “Yes/No” questions such as “Has anyone ever used social media to hurt you?.” The cyber-bullying perpetration section of the questionnaire demands responses to statements like “How often have you taken a picture of someone and posted it online without their permission?” on a 5-point Likert-scale ranging from “Never” to “Everyday”. Whilst the victimization section of the questionnaire also utilized the same style of Likert-scales, it also used binary questioning to assess victimization predictors.

The cyber-bullying bystander scale

For this study the Cyber-bullying Bystander Scale (Sarmiento et al., 2019) was shortened to include 18 of the 40 statements from the original questionnaire. Eighteen statements were added to the original statements to compare prevalence rates before and during COVID-19 lockdowns. Responses to each statement were measured on a 5-point Likert-scale ranging from “Very Frequently” to “Never”. The 18 statements were split across three subsections: Passive Bystander, Defender of the Victim, and Reinforcer of the Bully. The Passive Bystander sub-section included statements like “I read on the internet hurtful messages from some people against others, but I do not say anything to defend them,” whereas the Defensive Bystander section included statements like “I tend to defend people attacked or insulted on social networks or on the internet,” and the Reinforcer Bystander section included statements such as “I share hurtful posts (photos, videos, or messages) that were uploaded by others.”

Results

Table 1 shows the comparisons of cyber-bullying perpetration before and during the COVID-19 lockdown. Before the lockdowns, 48.7% percent of all participants responded “never” to the statement “How often have you posted something online about someone else to make others laugh?” but during the lockdowns the frequency of “never” responses rose 71.3%. A paired samples t-test showed that this increase in “never” statements was statistically significant (t (229) = 7.857, two-tailed p = .002). The same was seen when comparing “never” responses for the statement “How often have you sent someone a computer text message to make them angry or to make fun of them?” as the frequency of “never” significantly increased from 71.7% before March 2020 to 88.3% during the lockdowns (t (229) = 5.786, two-tailed p = .002). Likewise, the amount of “never” responses to the statement “How often have you taken a picture of someone and posted it online without their permission?” also significantly rose from 76.5% pre-lockdown to 88.3% during the lockdown, (t (229) = 5.688, two-tailed p = .002). As a result, the prevalence of cyber-bullying perpetration has decreased since the UK-wide lockdowns began.

Table 1. Comparisons of cyber-bullying perpetration before and during the COVID-19 lockdown.

Question and Responses	Before COVID-19%)	During COVID-19%)	Change (%)	Significance (p < .05)
How often have you posted something online about someone else to make others laugh? - Never	48.7	71.3	+22.6	p = .002
How often have you sent someone a computer text message to make them angry or to make fun of them? - Never	71.7	88.3	+16.6	p = .002
How often have you sent someone an e-mail to make them angry or make fun of them? - Never	95.7	97.0	+1.3	Not sig at 0.05 level
How often have you posted something on someone’s social media page or other web page to make them angry or make fun of them? - Never	87.0	93.9	+6.9	Not sig at 0.05 level
How often have you taken a picture of someone and posted it online without their permission? - Never	76.5	88.3	+11.8	p = .002

Comparison of cyber-bullying victimisation

Of the 230 people who participated in this study, 105 (45.6%) admitted to experiencing cyber-bullying before March 2020, whereas 125 (54.3%) stated they had not experienced any form of cyber-bullying victimization. Since the first national UK lockdown in March 2020, 30 (28.5%) of those who had previously admitted to being victimized by cyber-bullying stated that the bullying had continued throughout the lockdown period, whereas 75% (71.4%) stated that they had not been victimized by cyber-bullying since the national lockdown had started. A paired-samples t-test was conducted to compare the prevalence of cyber-bullying before and after March 2020. There was a significant difference in cyber-bullying prevalence between the before and during national lockdown periods (t (104) = −16.125, two-tailed p = .002). From this, it can be seen that, there was a decrease in overall prevalence of cyber-bullying since the start of the UK-wide lockdowns in March 2020.

This trend continues when comparing the incidence of cyber-bullying both before and during UK lockdowns. For example, before March 2020, 37.8% of females and 7.8% of males stated that they had been cyber-bullied, whereas 22.8% of these females and 7.8% of these males reported that the cyber-bullying continued during this period. Moreover, cyber-bullying prevalence declined amongst all ethnicities after March 2020. For example, only 19% of the 74 White British/Irish respondents who admitted to previous cyber-bullying victimization admitted that such behaviors continued during the pandemic.

In addition to the decrease in cyber-bullying prevalence during the 2020 UK national lockdowns, an overall decline in specific cyber-bullying behaviors was seen too. All 105 participants who had experienced cyber-bullying before March 2020 rated the frequency of each type of cyber-bullying incidents. All nine of the statements showed an increase in the frequency of “never” responses to cyber-bullying incidents during the lockdown (see Table 2). However, only eight of these were statistically significant. For example, responses to the statement “How often have you received an email from someone you know that made you really mad?” showed a significant decrease in the frequency of times this incident occurred during lockdown as the amount of “never” responses significantly rose from 67.6% pre-lockdown to 80% during lockdown (t (104) = 3.309, two-tailed p = .002).

Table 2. Comparisons of “never” responses before and during UK lockdowns pertaining to cyber-bullying victimization.

Question and Responses	Before COVID-19%)	During COVID-19%)	Change (%)	Significance (p < .05)
How often have you been made fun of by another person in a chat room? - Never	30.5	61.9	+31.4	p<.001
How often have you received an e-mail from someone you know that made you really mad? - Never	67.6	80.0	+12.4	p = .002
How often have you received an e-mail from someone you didn’t know that made you really mad? This does not include “spam mail.” - Never	79.0	81.9	+2.9	p<.001
How often has someone posted something on a social media site that made you upset or uncomfortable? - Never	13.3	45.7	+32.4	Not sig at .05 level
How often has someone posted something on another web page that made you upset or uncomfortable? - Never	35.2	49.5	+14.3	p<.001
How often have you received an instant message that made you upset or uncomfortable? - Never	9.5	38.1	+28.6	p<.001
How often have you been bullied or picked on by another person while online? - Never	20.0	59.0	+39.0	p<.001
How often have you been afraid to go online? - Never	48.6	69.5	+20.9	p<.001
How often has anyone posted anything about you online that you didn’t want others to see? - Never	24.8	70.5	+45.7	p<.001

Comparison of cyber-bullying bystander roles

Comparison of passive cyber-bullying bystanders

Since the beginning of UK-wide lockdowns, the frequency of passive cyber-bullying behaviors has increased (see Table 3). Of the five behaviors measured, three showed a statistically significant decrease in the amount of “never” responses provided. For example, the statement “When I browse the internet and/or social networks, I see how some people make fun of others, but I do not do anything to avoid it” saw a significant decrease from 25.2% to 13.9% (t (229) = −3.189, two-tailed p = .004). The same was seen for the statement “When I interact in social networks, I see people attack others who cannot defend themselves, but I do not get involved” as the amount of “never” responses significantly decreased from 24.8% to 18.7% (t (229) = −3.437, two-tailed p = <.002). The statement “When I see that someone is attacking a person on the internet (insulting, mocking, or hurting them), I choose not to do or say anything” also showed an increase in prevalence of this behavior as the amount “never” responses dropped significantly from 25.7% to 19.1% (t (229) = −2.494, two-tailed p = .026).

Table 3. Comparison of passive cyber-bullying bystander behaviors from before and during COVID-19 lockdowns.

Question and Responses	Before COVID-19%)	During COVID-19%)	Change (%)	Significance (p < .05)
When I browse the internet and/or social networks, I see how some people make fun of others, but I do not do anything to avoid it.- Never	25.2	13.9	−11.3	p=.004
When I interact in social networks, I see people attack others who cannot defend themselves, but I do not get involved. - Never	24.8	18.7	−6.1	p<.001
I see on the internet and/or social networks how some people upload photos or videos that harm others, but I do not say anything to defend them. - Never	27.4	26.5	−0.9	Not sig at .05 level
I read on the internet hurtful messages from some people against others, but I do not say anything to defend them. - Never	20.9	20.0	−0.9	Not sig at .05 level
When I see that someone is attacking a person on the internet (insulting, mocking, or hurting them), I choose not to do or say anything. - Never	25.7	19.1	−6.6	p=.026

Comparison of defensive cyber-bullying bystanders

As with passive bystander behaviors, defensive bystander behaviors have also increased. The frequency of “never” responses decreased across all defensive bystander statements (see Table 4) with four of the six statements showing a statistically significant decline. For example, the statement “When I see on the internet that some people upload photos or videos that are offensive to others, I tell them that this is wrong” had a statistically significant reduction from 37% of “never” responses before lockdown to 27% of “never” responses during lockdown (t (229) = −3.592, two-tailed p = <.001). This suggests that participants tended to defend cyber-bullying victims more during the national lockdown.

Table 4. Comparison of defensive cyber-bullying bystander behaviors from before and during COVID-19 lockdowns.

Question and Responses	Before COVID-19%)	During COVID-19%)	Change (%)	Significance (p < .05)
When I am on social media and I see some people harass others who cannot defend themselves, I tell them to stop. -Never	35.7	27.0	−8.7	p=.03
When I see on the internet that some people upload photos or videos that are offensive to others, I tell them that this is wrong. -Never	37.0	27.0	−10.0	p<.001
If I read hurtful messages on the internet, of a person toward another weaker person, I write to the person who wrote the message so that it stops. -Never	38.3	37.4	−0.9	Not sig at .05 level
When I see on the internet that some people post offensive messages about others who cannot defend themselves, I write to them to say to that this is wrong. -Never	39.6	34.3	−5.3	Not sig at .05 level
I tend to defend people attacked or insulted on social networks or on the internet. -Never	33.0	25.2	−7.8	p=.01
When I see someone upload photos or videos that are offensive to others, I write to them to ask them not do it more or to remove it. -Never	39.1	33.9	−5.2	p=.18

Comparison of bystander behaviours which reinforce cyber-bullying

All seven of the behaviors that reinforce cyber-bullying became slightly more frequent during lockdown (see Table 5). However, only two of the statements showed a statistically significant decrease in the amount of “never” responses from before March 2020. The first of these statements, “When I browse the internet, I click ‘like’ on publications such as photos, videos, messages and hurtful rumours about others who cannot defend themselves” had a significant decrease from 88.3% of “never” responses to 80.4% of responses (t (229) = −2.346, two-tailed p =.04). The second statement, “When I browse the internet, I support with comments the hurtful publications such as photos, videos, messages, and rumours towards others who cannot defend themselves”, showed a significant reduction from 94.8% of participants stating they had “never” engaged in this behavior pre-lockdown to just 88.7% during lockdown (t (229) = −3.226, two-tailed p =.002). This indicates a significant increase in passive supporting behaviors compared to direct encouragement.

Table 5. Comparison of reinforcing cyber-bullying behaviors from before and during UK COVID-19 lockdowns.

Question and Responses	Before COVID-19%)	During COVID-19%)	Change (%)	Significance (p < .05)
I share hurtful posts (photos, videos, or messages) that were uploaded by others. -Never	93.5	90.4	−3.1	Not sig at .05 level
When I browse the internet, I click “like” on publications such as photos, videos, messages, and hurtful rumours about others who cannot defend themselves-Never	88.3	80.4	−7.9	p=.04
When I browse the internet, I support with comments the hurtful publications such as photos, videos, messages, and rumors toward others who cannot defend themselves.-Never	94.8	88.7	−6.1	p=.002
When interacting in social networks on the internet, I see people who harass others and let them know that I find it funny. -Never	94.3	91.7	−2.6	Not sig at .05 level
I am interested and enjoy being part of groups in social networks that publish hurtful photos, gossip, and insults about others. -Never	93.9	93.0	−0.9	Not sig at .05 level
I follow profiles of people who post hurtful photos, gossip and insults about others and I make them see that I like them. -Never	91.7	89.1	−2.6	Not sig at .05 level
I like to encourage or support people who post hurtful photos, gossip, and insults about others. -Never	96.5	95.7	−0.8	Not sig at .05 level

Discussion

The results of the present study indicate that there was a decrease in cyber-bullying during the lockdown period. This study also explored frequency rates of bystander types in cyber-bullying to ensure an accurate picture of cyber-bullying prevalence. Previous research has indicated that protective bystanders are the least prevalent form of bystanders and that bystanders who reinforce bullying behaviors can increase the incidence of cyber-bullying. There were many significant findings, indicating that cyber-bullying incidence decreased during the lockdown.

The results of the cyber-bullying perpetration rates indicated a decline in cyber-bullying behaviors, such as taking and posting photos online without the individual’s consent and sending inflammatory text messages. A decrease in perpetration behaviors, such as sending harmful messages, is a counterintuitive result, as the lockdown provided a decreased chance for perpetrators to bully in-person and an increased prevalence of predictor factors, such as anger regarding family illness. Furthermore, the results of the cyber-bullying victimization analyses also showed a decreased rate in victim prevalence too. With an increased rate of predictor factors such as loneliness (Sahin, 2012) and bystander reinforcement (Salmivalli et al., 2011), it was expected that victimization would also increase. However, an increase in cyber-bullying incidents did not occur. Where the frequency of individuals experiencing cyber-bullying and not experiencing cyber-bullying before March 2020 was roughly equal, there was a statistically significant decrease in individuals who experienced cyber-bullying incidents during the lockdown (approximately 28.5%). This finding was supported by the results of questions asking victims to rate frequency of specific cyber-bullying statements. Eight of the nine victimization statements showed a significant increase in the “never” responses during the pandemic. For instance, a significant increase of participants reported that they had “never” had someone post something online they did not want others to see, when compared to individuals during the pre-March 2020 period. These results show a noticeable decline in the amount of cyber-bullying incidents during lockdowns.

Support for this decrease in cyber-victimization prevalence during the lockdown can be seen by comparing crosstabulations of frequency of cyber-victimization by ethnicity and gender. One predictor of increased cyber-victimization is ethnicity, as individuals who define themselves as White or Black/Black British experience an increased risk for cyber-victimization compared to other ethnicities (ONS, 2020). Despite this, no one ethnicity in the present study experienced an increase in cyber-victimization during the lockdown. There was a greater decline in cyber-bullying victimization experienced by females than by males. Even though the ONS (2020) reported that there were no gender differences for cyber-victimization between March 2019 and March 2020, female cyber-victimization declined by a greater amount than male cyber-victimization. However, this could be a result of there being more females in the COVID-19 lockdown condition than males, rather than an actual effect of the lockdown. In sum, these results indicate that the COVID-19 lockdown did not increase the prevalence of cyber-victimization. In fact, the lockdown decreased cyber-victimization.

As mentioned above, Cassidy et al. (2009) reported that cyber-bullying is an extension of in-person bullying. This could suggest that cyber-bullying has decreased because no in-person bullying took place within the same time period in the present study. However, a failure to see an increase in cyber-bullying prevalence could suggest three things: (1) the predictor factors that increased during the COVID-19 lockdowns do not actually increase cyber-bullying prevalence, (2) having minimal interaction with the perpetrator in-person does not increase risk for online bullying, or (3) a presently unknown mediating variable was involved. Since there is a wealth of previous research regarding the predictive factors for cyber-bullying incidence and a few studies showing that both cyber-bullying perpetration and victimization has increased during lockdowns (Bartlett et al., 2021; Jain et al., 2020), it is possible that a mediating variable changed the present results. One potential mediator variable assessed in this study was bystander behaviors. The prevalence of behaviors amongst the three types of bystander roles (passive, defensive, reinforcer) were analyzed in relation to frequency before and during the COVID-19 lockdowns. Of the five statements for the passive bystander behaviors, four showed a decrease of the amount of “never” responses during the lockdown, three of which were statistically significant. This indicates that there was an increase in cyber-bullying prevalence as more participants reported seeing cyber-bullying behaviors but doing nothing about it during the lockdown than before March 2020. Similarly, two thirds of the statements regarding defensive bystander behaviors showed a statistically significant decrease in the number of participants who responded with the “never” option. This also indicates an increase in cyber-bullying prevalence as people have defended victims more over the lockdowns.

Again, all seven of the statements regarding reinforcing bystander behaviors showed a decrease in the amount of “never” responses during the lockdown, but only two were statistically significant. These findings indicate that reinforcing behaviors have increased. Interestingly, only 28.5% of the statements showed a significant decrease, compared to 66.67% of the defensive bystander and 60% passive bystander statements, indicating that, during lockdown, bystander behaviors that reinforced the bully were the least common. This is contradictory to previous research which states that the reinforcer-type bystander is usually the most common bystander reaction (Salmivalli et al., 2011) and could help to explain the findings for cyber-victimization and perpetration. As the reinforcer type of bystander was the least common type of bystander of the three roles, cyber-bullying may have decreased as a result of lack of support and perceived anonymity. A lack of social support implies a smaller social circle in which to become uninhibited, meaning that, if the perpetrators felt a lack of support for their behavior, they may lose the feeling of anonymity usually provided by group cyber spaces (Postmes, 2001; Willard, 2004). Therefore, decreases in cyber-victimization and perpetration could result from less reinforcement than usual.

In addition, a rise in defensive bystander behavior could help to explain the decline in the cyber-bullying prevalence rates. Brodeur et al. (2020) found that boredom increased in Europe and the USA during lockdown periods, meaning that individuals did not have as many activities to occupy them as before COVID-19. Since boredom can increase prosocial behaviors (van Tilburd & Igou, 2016), such as defending cyber-bullying victims, it is likely that those who had less things to do during lockdowns engaged in more defensive bystander behaviors. This is an especially salient idea when considering that increased defensive bystander behaviors (i.e., peer support) contributes to decreased cyber-bullying prevalence (Holfeld & Baitz, 2020; Smith et al., 2003). Therefore, bystander support may act as an important predictor for prevalence rates.

Limitations, future research and conclusions

To gain further information of how bystander behaviors might have mediated the prevalence rates of cyber-bullying during the COVID-19 lockdowns, future research needs to explore how COVID-19 lockdowns increased prosocial behaviors which may guide future intervention tactics. The present study disseminated the utilized survey during the third lockdown the UK was subjected to, potentially influencing the results of the research as participants may have influenced the results if the participants experienced different interactions during different lockdown periods. Future research should also investigate the reasons behind decreased cyber-bullying prevalence during lockdowns as this also can influence intervention tactics. To summarize, both cyber-bullying victimization and perpetration decreased during the COVID-19 pandemics in the UK. This finding did not support the research hypothesis, as it was expected that this study would find an increase in cyber-bullying prevalence during the lockdown. Counterintuitively, bullying bystander behaviors all increased during the COVID-19 lockdowns, with passive and reinforcing behaviors being more prevalent. It is thought that the increase in prosocial behaviors and the decrease in support for perpetrators acted as a mediating variable. Overall, it can be inferred that the increased opportunity to act in an altruistic manner helped to decrease cyber-bullying during a difficult time.

Data availability of statement

The data were collected in a manner consistent with ethical standards for the treatment of human subjects and according to the Declaration of Helsinki 1964.

Ethics

Prior to data collection, the study was approved ethically by the ethics committee of the University of Huddersfield.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Helena Charlotte Peck

Helena Charlotte Peck, At the time of the project, Miss Helena Peck was an MSc Investigative Psychology student.

Calli Tzani

Dr. Calli Tzani, Dr Calli Tzani completed her BA in Psychology at University of Indianapolis. Dr Tzani joined the University of Huddersfield in 2017 after completing her PhD in Bullying and Cyberbullying. Presently Dr Tzani is a Senior lecturer for the University of Huddersfield, with experience in the field of psychology, human well-being, development, teaching and human support. Dr Tzani’s research is focused on wellbeing, bullying and cyber-bullying prevention, sextortion, terrorism prevention, fraud, cybercrime and criminal behaviour.

David Lester

Professor David Lester, Professor David Lester has received a BA in psychology from Cambridge University, two MA’s, from Cambridge and Brandeis and two PhDs from both Cambridge and Brandeis. Professor Lester is a distinguished psychology professor at the University of Stockton, where he has specialized in researching suicide and has published numerous books and over two thousand research papers on the subject.

Thomas James Vaughan Williams

Thomas James Vaughan Williams, Thomas James Vaughan Williams is a researcher at the university of Huddersfield. His research focuses on ideology and radicalisation but also explores topics like climate change, terrorism, and online behaviours.

Josefa Page

Josefa Page, At the time of the project submission Miss Josefa Page was a researcher and PhD candidate at the University of Huddersfield, studying the typology of Sexual Homicide Offenders (SHOs) of children relevant to UK offences, and exploring offender and victim characteristics, as well as behaviours exhibited at crime scenes, previous convictions and geography of the offences to establish an updated ‘profile’ of SHOs of children that may be useful to UK law enforcement in the event of future relevant offences.