Bullying and Character Development: An Examination of Character Strengths Associated With Bullying and Cyberbullying in Postprimary Schools in Ireland Free

Bullying is a deliberately aggressive and repeated behavior that is generally unprovoked and demonstrates an imbalance of power on the perpetrator’s side. It includes behaviors such as verbal and physical abuse, exclusion, extortion, rumor spreading, and threats (Burke & Minton, 2016). Since the emergence of cyberbullying, this type of bullying has become known as “traditional bullying” (Corcoran et al., 2015). While early definitions of cyberbullying simply mimicked the definitional criteria for traditional bullying, the new medium for the behaviors to occur in (Web 2.0, internet mediated communication) has led to an interesting discussion regarding whether cyberbullying and traditional bullying are the same thing (e.g., see the debate between Smith, 2012, and Olweus, 2012). In reviewing the definitional criteria for cyberbullying, Corcoran et al. (2015) have argued that “cyber-aggression” is a better representation of the issue. They argue that the term “cyberbullying” has been problematic in that it was based on the criteria for traditional bullying (i.e., intent to harm, repetition, imbalance of power) and did not fully account for the unique nature of cyber-based aggression. This argument was supported by data regarding traditional bullying, cyberbullying, and cyber-aggression from 2,474 students attending schools in Ireland (Corcoran et al., 2015).

Mc Guckin and Lewis (2006) noted that “much of the research exploring bully/victim problems has been largely sporadic in nature, resulting in a smorgasbord of findings that in combination provide a rich tapestry of findings from which is beginning to emerge a picture of the nature, incidence, correlates, and management of bully/victim problems” (p. 350). In terms of prevalence, characteristics, and intervention approaches, Zych et al. (2015) have reported a useful systematic study of publications on bullying and cyberbullying, from 1978 to 2013, which showed that the prevalence of traditional bullying is 35%, whereas cyberbullying is 16% (Modecki et al., 2014). Smith and Berkkun (2017) developed this work with a more focused review of cyberbullying (up to 2015) indicating a rapid increase in academic articles relating to cyberbullying. The variations in how best to define the issue, and different methodological positions regarding how best to quantify the behaviors of concern (e.g., single item questions, scaled questionnaires), has resulted in contested prevalence rates and differing understandings of involvement in bully/victim problems. Numerous intervention approaches (e.g., Olweus, 1993; O’Moore & Minton, 2005; Salmivalli et al., 2010) have demonstrated (usually) small effects on reducing traditional bully/victim problems (e.g., see Farrington & Ttofi, 2009), and intervention knowledge regarding cyberbullying is still emerging (e.g., Mc Guckin & Corcoran, 2016; Vollink et al., 2016). Due to the differences between bullying and cyberbullying, in relation to the definition, prevalence and interventions, both concepts need to be explored separately.

With the exception of a few studies (e.g., Bannink et al., 2014; Sjurso et al., 2016), the majority of research on bully/victim problems reports little difference in the psychological impact of bullying behaviors on participants exposed to traditional and/or cyberbullying (e.g., Ortega et al., 2012; Tokunaga, 2010). Factors such as frequency (e.g., daily, weekly, monthly, rarely), intensity (e.g., low-high), and the severity of specific types of bullying (e.g., rumor spreading, exclusion, verbal, or physical violence) seem to have more bearing on young people’s mental health than the medium through which bullying occurred (Chen et al., 2013; Smith et al., 2008). The most severe type of bullying is physical bullying, followed by verbal and relational bullying, whereas the most frequent type of bullying is verbal, followed by physical (Chen et al., 2013). Cyberbullying often occurs in tandem with traditional bullying (Kadiroglu et al., 2018). Due to the similarities associated with the consequences of cyberbullying and traditional bullying, they will be discussed together.

The impact of bullying on victims of traditional and cyberbullying is usually reported as pejorative (e.g., Burke & Minton, 2016; Ortega et al., 2012). Victims of bullying and cyberbullying are perceived as having negative social and physical self-concept, and lower levels of self-esteem (Corcoran et al., 2012; O’Moore & Kirkham, 2001; Patchin & Hin- duja, 2010), and are considered less popular among their peers, as well as having a lesser number of friends (Fossati et al., 2012; Leung et al., 2018). Being a victim of bullying impacts negatively on an individual’s mental health, increasing their risk of anxiety (Hawker & Boulton, 2000), sleep problems (Van Geel et al., 2016), depression (Ttofi et al., 2011), and suicidal ideation (Holt et al., 2015). However, it has also been reported as being a source of posttraumatic growth for some (Burke, 2016; Ratcliff et al., 2017).

Bullies, on the other hand, have been reported as being impulsive (Olweus, 1991), lacking empathy (Olweus, 1993), provocative (Ehrler et al., 1999), selfish (Fossati et al., 2012), and Machiavellian (Sutton & Keogh, 2000). Furthermore, they have been reported as being more prone to psychiatric disorders (Kumpulainen et al., 1999), and score higher on measures of narcissism and psychoticism (Ivarsson et al., 2005) than their nonbullying counterparts. In summation, the research indicates that being a bully or a victim of bullying is associated with largely negative outcomes.

School bullying is a serious public health problem (Zych et al, 2020). Gaffney et al.’s (2019b) meta-analysis of 100 antibullying program evaluations, including cyberbullying, found that they reduced school bullying by 19%-20% and school bullying victimization by 15%-16%. Further analysis that isolated cyberbullying programs found that when directly targeted, they reduced cyberbullying by 10-15%, and cyber victimization by 14% (Gaffney et al., 2019a). While this finding is to be welcomed, it emphasizes the importance of designing programs that may have a higher level impact on the reduction of bully/victim problems. Most of the programs reviewed by Gaffney et al. (2019a) focused on creating awareness, outlining consequences of bullying and cyberbullying, and developing coping strategies to helping victims cope more effectively.

Only two programs mentioned by Gaffney et al. (2019b) offered a strength-based approach to reducing bully/victim problems and combating the effects. The first of these was the whole-school “strengths in motion” program (Rawana et al., 2011) that focuses on identifying and applying strengths to promote prosocial behavior and enable victims to cope more effectively. The second program, the “resourceful adolescent program,” is a cognitive-behavioral therapy-intervention which has a component that discusses an individual’s personal strengths (Stallard et al., 2013). Both programs demonstrated a statistically significant decrease in bully/victim problems, with the strengths in motion program indicating a long-term effect measured at 8 months postintervention. Thus, while strengths are an emerging focus of antibullying programs, relatively little is known about the relationship between strengths and bully/victim problems.

A promising development in antibullying interventions comes from the research on the social aspects of bullying, according to which, in over 85% of bullying cases there are bystanders present who have the power to intervene in bullying (Pepler & Craig, 1995). They include defenders of the victim, reinforcers and assistants of the bully, as well as outsiders, who choose to walk away from the situation (Salmivalli et al., 1996). A “Kiusaamista Vastaan” (KiVA)—meaning “antibullying” in Finnish—program was developed, the aim of which was to empower bystanders to defend targets, in order to reduce bullying and bully-reinforcing behaviors (Salmivalli et al., 2010). While the program is associated with a reduction in bullying, the effect size are lower than other programs (Gaffney et al., 2019b). Given that aim of the program was to develop prosocial behaviors, it is deemed a strength-based approach to reducing bullying.

A previous study in a different field of research showed that using a strengths-based approach to career-counseling resulted in a 20% increase of clients’ rate of employment 3 months after the intervention, in comparison to a traditional approach (Littman-Ovadia et al., 2014). Also, when participants with depression were introduced to a positive psychotherapy intervention, including a discussion about strengths, their depression levels were significantly lower than “the treatment as usual” group (Seligman et al., 2006). This indicates a precedent for a strength-based approach being more effective than a traditional approach. Therefore, exploring the relationship between bullying and strengths can act as the first step toward refining a strength-based antibullying intervention.

The first established model of strengths was a Clifton StrengthsFinder, which views strengths as refined talents (Rath, 2007). The second model, StrengthsScope considered strengths within the workplace and divided them into four categories of strengths; emotional, relational, executive and thinking strengths (Brook & Brewerton, 2006). Another model of strengths, realized strengths, which is now referred to as the strengths profile sees strengths as having an energizing quality and differentiates between realized strengths, unrealized strengths, weaknesses and learned behaviors, which referred to overused strengths or nonstrengths refined by habitual use (CAPP, 2010). Of the three models, the StrengthsFinder is the only one that has an assessment specific to youth, therefore could be potentially used as a strength measure (e.g. Hodges & Harter, 2005). However, the current study applied the “Values in Action Inventory of Strengths for Youth” (VIA-Youth: Park & Peterson, 2005), which is yet another model of strengths, as it has a more robust empirical basis than the StrengthFinder (Burke & Passmore, 2019) and has been used extensively in the context of positive education (i.e., the study of well-being and other positive outcomes in schools; e.g., Noble & McGrath, 2008; Norrish, 2015; Seligman et al., 2009).

VIA stands for values in action, therefore identifies what values individuals use at various frequency that best describes them and helps them build character. According to the VIA-Youth, good character is a “family of positive traits reflected in how people think, feel and behave” (Park & Peterson, 2010, p. 638). In order to measure good character, otherwise referred to as character strengths, Peterson and Seligman (2004) created a handbook, classification, and measurement of it, in a form of a Values in Action Inventory of Strengths (VIA-IS: Peterson & Seligman, 2004), which has been subsequently adapted to adolescents (VIA-Youth: Park & Peterson, 2005) and recently updated by McGrath (2017).

The validated measure of character strengths was developed following strict empirical guidelines (Peterson & Seligman, 2004). The strengths were created by reviewing the world’s most influential traditions (e.g., from Christianity through to Taoism and Islam; Dahlsgaard et al., 2005), as well as the science of psychiatry, psychology, and education. Included too were popular songs, greeting cards, bumper sticker, children books, and others (Peterson, 2006). They are organized into six virtues (courage, justice, humanity, temperance, wisdom, and transcendence) with a strict 10-item inclusion criteria, such as that each strength needed to contribute to the good life, be morally valued, not diminish other people, or manifest itself in participant’s thoughts, feelings, and behaviors (Peterson & Seligman, 2004). The VIA character survey is also referred to as an anti-Diagnostic and Statistical Manual of Mental Disorders survey (e.g., Boniwell, 2006), as it measures an individual’s good character, rather than what is wrong with them.

Strengths have been tested extensively with a young population. When compared with adults, young people scored higher on the strengths of hope, teamwork, and zest (Park & Peterson, 2006). However, as they grow and transition through primary and postprimary education, they develop a wider range of character strengths, allowing their character structures to become systematically more differentiated in adolescence (Shubert et al., 2019). This also provides evidence that strengths are not static and can be consciously altered by the person and their environment.

Teaching youth to apply their character strengths is beneficial. Using strengths is associated with heightened levels of well-being (Burke & Minton, 2019; Oppenheimer et al., 2014), self-esteem, and life satisfaction (Freire et al., 2018). Interventions used to create awareness and teach young people how to apply strengths more effectively in their lives resulted in better class cohesion and relatedness (Quinlan et al., 2014), improved school performance, decrease of negative classroom behavior, and enhancement of social relationships and academic motivation (Grinhauz & Castro Solano, 2014). Therefore, character strengths not only change throughout young people’s lives, but can also be taught.

In a 2011 theoretical paper, Tweed and colleagues (2011) reviewed the concept of VIA character strengths as well as other positive psychological research and suggested a relationship between youth violence and gratitude, forgiveness, sense of meaning, altruism, humility and prudence. The authors claimed that positive psychology research could lend itself to reducing youth violence. Similarly, of all examined strengths, research indicated that the strengths of prudence, honesty, persistence, and love were associated with fewer externalizing problems and more prosocial behaviors (Park & Peterson, 2008), which is why these strengths may also be related to bully/victim problems. Therefore, the past research indicated that certain character strengths may indeed play a role in protecting individuals from bullying. Considering that there is no empirical evidence of the association between involvement in bully/victim problems and specific character strengths, the current research did not offer any hypotheses, instead, aimed to explore the strengths associated with nonviolent bullying behaviors. As such, our research question is: What character strengths predicted traditional bullying, cyberbullying, and being a victim of traditional and cyberbullying?

The sample included 2,799 participants aged 12-19 (M = 15.05, SD = 1.66), 50.8% of whom were females, from 13 secondary schools located across Ireland (both rural and urban areas). Strategic sampling was applied to ensure all types of schools were represented (i.e., secondary, vocational, comprehensive, and community), across all the provincial regions of the Republic of Ireland (Connaught, Leinster, Munster, and Ulster), and various denominational affiliations (Roman Catholic, Church of Ireland, multi-, and nondenominational), as well as language use (English and Irish). Informed consent was obtained from school principals, who acted in loco parentis. The questionnaire was administered by school teachers in a paper-and-pen format, half way through the academic year (January-February). Each school was requested to distribute questionnaires to a maximum of 30% of its students’ population and ensure that each year was represented equally.

Two measures were used in the current study: The Values in Action Inventory of Strengths for Youth (VIA-Youth; Park & Peterson, 2005), and Corcoran’s (2013) version of the netTEEN bullying and cyberbullying questionnaire (Machmutow et al., 2012).

Character strengths were assessed using a 96-item VIA-Youth survey, which had a reliability of α = .87 in past research, thus demonstrating acceptable internal consistency levels (VIA, 2014). The respondents were asked to read statements and decide how much each statement was “like them.” Sample statement included: I am a very loyal member of my group/team and My temper often gets the best of me. Responses were recorded on a 5-point Likert scale ranging from 1 (very much like me) to 5 (not like me at all). The results were analyzed by the VIA Institute, which provided respondent’s mean scores for all character strengths. Sample strengths included: creativity, curiosity, bravery, honesty, kindness, leadership, and self-regulation.

Traditional and cyberbullying was assessed using Corocran’s (2013) modified 32-item version of the netTEEN questionnaire (Machmu- tow et al., 2012). Responses were recorded on a 5-point Likert scale, ranging from 1 (never) to 5 (almost daily), and included an additional response option of 6 (no answer). For the analysis, the occurrence of at least one bullying behavior in the highest frequency of bullying and cyberbullying was deemed as the actual frequency. Therefore, if a participant had never experienced bullying behavior—except for one form of bullying (rumors: experienced almost daily)—their frequency was recorded as being bullied almost daily. The questionnaire measured bullying behaviors that occurred in the previous 3 months. Seven questions assessed the frequency of displaying behaviors associated with traditional bullies. Sample questions included: Have you laughed at anybody, or did you say bad things to them? Have you deliberately excluded anybody so that they could not join in? A further seven were associated with the victims of traditional bullying. Sample questions included: Has anybody said bad things about you (e.g., spreading rumors)? Has anybody hit you, tripped you, or hurt you in some way? Nine questions inquired about being cyberbullies. Sample questions included: Have you sent mean or threatening messages to anyone (e.g. text messages, MSN, Facebook, etc.)? Have you sent mean or embarrassing messages or spread rumors about anyone to your friends (e.g. text messages, MSN, Facebook, etc.)? Finally, nine questions were related to experiences of being a victim of cyberbullying. Sample questions included: Has anybody used your username and password (e.g., breaking into your Facebook or email account to do mean or embarrassing things? Has anybody excluded you (e.g., blocked you on Facebook, or stopped you from joining a game online)? Past internal reliability for all of the four subscales for the current study (traditional bully, traditional victim, cyber bully, cyber victim) ranged between α = .63 and α = .78 (Corcoran, 2013), and for the current study it ranged between α = .84 and α = .96.

SPSS (Version 25, IBM Corp., 2019) was used to conduct the statistical tests. The mean scores varied across all four bullying scales, with cyber bullies yielding the lowest scores (M = 2.09, SD = 1.42), and “traditional victims” showing the highest scores (M = 2.98, SD = 1.50). Similarly, the mean values for strengths were different across the groups, with the lowest values for “spirituality” (M = 2.79, SD = 1.16) and the highest value for “gratitude” (M = 4.08, SD = .72). See Table 1 for all descriptive statistics.

Stepwise regression was used to assess the ability of all character strengths to predict the frequency of behaviors associated with a traditional bully, traditional victim, cyber bully, and cyber victim. Preliminary analyzes were conducted to ensure there were no violations of the assumptions of normality (residuals), linearity, multicollinearity, and homoscedasticity. Bonferroni correction was applied to adjust for alpha inflation.

The first stepwise regression revealed that the frequency of traditional bullying was best predicted in the final model by the strengths of “fairness” (β = -.19), followed by “prudence” (β = -0.9). The eight-predictor model accounted for 12% of variance in bullying (R² = .12, F[8, 2,790] = 45.88, p < .001). See Table 2 for details of the analysis.

The second stepwise regression demonstrated that the frequency of being a traditional bullying victim was best predicted by the strengths of “love”, followed by “prudence”. The 10-predictor model accounted for 7% of variance in being a victim of bullying (R² = .07, F[10, 2,788] = 22.22, p .001, love [β= -10], prudence [β= -.09]). See Table 3 for details of the analysis.

The third stepwise regression revealed that the frequency of cyberbullying was best predicted by fairness, followed by prudence. The 7-predictor model accounted for 9% of variance in cyberbullying (R² = .09, F[7, 2,791] = 36.83, p < .001, fairness [β= -.14], prudence [β= -.12]). See Table 4 for details of the analysis.

The final stepwise regression revealed that the frequency of being a victim of cyberbullying was best predicted by prudence, followed by fairness. The 9-predictor model accounted for 5% of the variance in scores for being a cyber victim (R² = .05, F[9, 2,789] = 14.45, p < .001, prudence [β= -.07], fairness [β= -.07]). See Table 5 for details of the analysis.

Previous work in the area of prevention and intervention in traditional and cyberbullying has neglected to understand the importance of strengths, even though slow progress has been made in the last decade to incorporate this approach (e.g., Rawana et al., 2011; Salmivalli et al., 2010). The results from this research has added much needed new knowledge regarding the efficacy of the strengths approach and how it should be useful in the further (re)developments of such approaches. The findings from the current study demonstrate that the strength of prudence is a common denominator across the top-two strengths that predict traditional bullying and cyberbullying behaviors. Specifically, lower self-reports of prudence was associated with a higher likelihood of being a traditional bully and a cyber bully, as well as a victim of both traditional and cyberbullying. While previous studies have suggested that there may be a link between prudence and violent behavior (Park & Peterson, 2008; Tweed et al., 2011), this is the first research to confirm this association. Considering that there are various definitions of prudence, it is important to further define it in the context of the current study.

Prudence is considered one of the risk attitudes that goes beyond risk aversion (Denuit & Eeckhoudt, 2016). It is a regulator of human values based on choices that people make (Larrivee & Gini, 2014)—therefore, it is seen as a meta-strength regulating other strengths. It is also associated with both self-control and conscientiousness (Furnham et al., 2016). However, as a VIA strength, it indicates a “cognitive orientation to the personal future, a form of practical reasoning and self-management that helps to achieve the individual’s long-term goals effectively” (Peterson & Seligman, 2004, p. 477). The results from the current study relate to the VIA definition of prudence, rather than its other meanings. Therefore, when designing any future interventions, the VIA definition needs to be considered. Also, in order to avoid generalization relating to the findings about prudence, further research needs to explore how other definitions of prudence predict bullying.

The VIA-Youth questionnaire consists of four questions on the basis of which prudence is measured. The strength of prudence denotes individuals who are cautious, differentiate right from wrong, and who consider the consequences of their behavior (VIA-Youth, 2014). An important aspect of prudence is, therefore, its future time orientation that instigates impulse-control in the present. It is about not saying anything, or acting in any way that may be regretted later (Park & Peterson, 2006). This is particularly relevant in relation to bully/victim problems that is associated with impulsivity (Walters & Espelage, 2018), which implies reactivity and not thinking about the consequences of one’s actions. This could be one reason why participants who scored low in prudence were more likely to display bullying behavior in the current sample.

Furthermore, McGrath (2017) analyzed the content of the VIA scale and found that there was an overlap between prudence and other strengths, such as bravery, honesty, and self-regulation. These constructs are of central importance to any work that seeks to ameliorate the insidious effects of bully/victim problems. In this context, these strengths are associated with being an “upstander” (i.e., standing up to others) rather than a “bystander” (i.e., bravery), telling the other person how one feels (i.e., honesty), and making systematic choices that lead to long-term achievement of goals, such as reporting a bully to a teacher (i.e., self-regulation). Therefore, individuals scoring low in prudence may also be more passive in the way they deal with bullying behaviors directed at them, which is consistent with Olweus’ seminal work (1984, 1990), demonstrating that 80% of bullying victims tend not to react to bullies. This has been a consistent finding in cyberbullying research too - that children and young people are loathe to report incidents of victimization for fear that the adult response will be to remove their technology, rather than any action directed toward the perpetrator of the action (Mc Guckin & Corcoran, 2020). This could be another reason as to why participants who scored low in prudence were also more likely to be victims of bullying in the current study.

In addition to the strength of prudence, other strengths, such as fairness, love, spirituality, and gratitude have been identified as predicting one, two or three types of bullying (traditional bully, and/or traditional victim, and/or cyber bully, and/or cyber victim). While the findings from the current study focused on the character strengths predicting all four types of bullying, further research may endeavor to create profiles of nonbullies and nonvictims, because strengths do not exist in isolation and their effect on each other needs to be considered (Biswas-Diener et al., 2011).

However, there is also an argument for developing all other strengths, as they are associated with improvement in well-being, which can inadvertently reduce anti-bullying behavior, given that bullying is associated with lower levels of well-being (Burke & Minton, 2016). This fits well with curriculum developments at the international level, where there is increased focus on issues related to personal well-being and the more holistic development of the pupil. Moreover, the implications of the findings may also be extended to the issues of improving overall student safety in school and the role strengths play in accomplishing this. Further research needs to explore the links between the strengths in relation to bullying.

While caution is needed in interpreting results, the greatest implication of the current study is the potential it affords for creating a pro-social, strength-based program to intervene and prevent traditional and cyberbullying among children and young people. Previous studies have demonstrated that character strength education may alter young people’s behavior (e.g., Bates-Krakoff et al., 2016; Freire et al., 2018). Therefore, strength-development, specific to strengths associated with a nonbully profile may be created with a particular focus on developing the strength of prudence, which to date is reported as the least endorsed character strengths (Karris & Craighead, 2012; McGrath, 2015; Satoshi et al., 2006). Future, experimental research may explore whether developing the strength of prudence and other strengths associated with nonviolent behavior have a capacity to reduce bullying in schools. This new antibullying, or prosocial program may become a significant part of positive education and make a contribution to UNESCO’s (2016) ambitious objective of eradicating bullying in schools by 2030.