Cyber-Bystander Behavior Among Canadian and Iranian Youth: The Influence of Bystander Type and Relationship to the Perpetrator on Moral Responsibility

Nagar, Pooja Megha; Leduc, Karissa; Khalili, Nadia; Talwar, Victoria

doi:10.3389/fcomm.2022.796146

ORIGINAL RESEARCH article

Front. Commun., 14 April 2022

Sec. Culture and Communication

Volume 7 - 2022 | https://doi.org/10.3389/fcomm.2022.796146

This article is part of the Research Topic Examining Bias-Based Cyberaggression and Cybervictimization from a Cross-Cultural Perspective View all 5 articles

Cyber-Bystander Behavior Among Canadian and Iranian Youth: The Influence of Bystander Type and Relationship to the Perpetrator on Moral Responsibility

$\nPooja Megha Nagar$ Pooja Megha Nagar^*

Karissa Leduc

Nadia Khalili

Victoria Talwar

The Department of Educational and Counselling Psychology, McGill University, Montreal, QC, Canada

The current study examines how social determinants influence the way youth from Canadian and Iranian contexts evaluate and morally disengage as bystanders of cyberbullying. While Iranian culture differs from other individualistic and collectivist cultures, Iranian youth have become just as technologically acculturated as their global peers. Despite this, less is understood about how Iranian youth respond to cyberbullying in comparison to youth from individualistic societies. Participants from Canada (N = 60) and Iran (N = 59) who were between the ages of 8-to-15 years old (N = 119, M = 11.33 years, SD = 1.63 years) read 6 cyberbullying scenarios that varied according to Bystander Relationship to Perpetrator (Acquaintance or Friend) and Bystander Response (Assists Cyberbully, Does Nothing, Defends Victim). After reading each scenario, participants were asked to evaluate the bystander's behavior. They were also asked how they would feel if they were the bystander. Similar to past research, these responses were coded on a continuous scale ranging from morally disengaged to morally responsible. Overall, Canadians were more critical of passive bystander behaviors and more supportive toward defending behaviors compared to Iranians. Iranians were more supportive of the behaviors of bystanders who were friends of perpetrators than Canadians were, and Iranians were more critical toward acquaintances of perpetrators. Significant interactions were also found between participants' country of origin, the bystander's relationship with the perpetrator and the bystander's behavior. Taken together, these findings highlight the importance of differentiating between negative judgments and moral attributions of bystander responses.

Introduction

Cyberbullying is an issue that is affecting teens world-wide. Similar to traditional bullying, cyberbullying can be characterized by a willful intent to harm another (Patchin and Hinduja, 2006; Nocentini et al., 2010; Hutson, 2016). However, cyberbullying is distinct from traditional bullying because hateful content can be created and shared anonymously across different online contexts (Menesini et al., 2011; Bryce and Fraser, 2013; Sticca and Perren, 2013; Hutson, 2016; Leduc et al., 2022). Thus, the current generation of youth are continuously exposed to cyberbullying, and they are becoming increasingly desensitized to it (Pabian et al., 2016).

Desensitization to online bullying is problematic given the key role that bystanders play in reducing the severity of bullying (e.g., Salmivalli, 2010; DeSmet et al., 2019; Ng et al., 2021). Positive bystander action is a prosocial construct that involves defending the victim (Pronk et al., 2019) while simultaneously diminishing the perpetrators' power (Salmivalli, 2010; Anderson et al., 2014). Nonetheless, past research (Quirk and Campbell, 2015; DeSmet et al., 2016; Erreygers et al., 2016; Olenik-Shemesh et al., 2017; Song and Oh, 2018) has demonstrated that bystanders in both elementary school and high school stand by and remain passive when they witness cyberbullying, which highlights the prominence of bystander inaction across childhood and adolescence. Additionally, social media engagement is becoming increasingly popular with younger children (Gomez-Garibello et al., 2012; Talwar et al., 2014; Conway et al., 2016). As such, the current study aims to examine the factors that facilitate positive bystander action in these age groups.

Since cyberbullying is a global phenomenon that can occur cross-culturally, it is also important to understand how positive bystander behavior can be facilitated across diverse cultural contexts. Nonetheless, most research that has investigated the determinants of positive bystander action has been limited to Western cultures (e.g., DeSmet et al., 2014; Machackova and Pfetsch, 2016; Domínguez-Hernández et al., 2018), and less is known about how these determinants vary in Eastern contexts, particularly in the Middle East. For instance, the Middle East holds values that are more collectivist (e.g., focusing on the collective good) than individualistic (e.g., focusing on individual well-being; Buda and Elsayed-Elkhouly, 1998; Minkov et al., 2017). Previous research suggests that collectivist societies who are more likely to conform with group norms are less likely to defend a single victim (Kogut et al., 2015; Liu and Tung, 2018). To illustrate, Liu and Tung (2018) found that Taiwanese junior high students with high levels of conformity were less likely to defend cyberbullying victims. The researchers also found that those who had positive peer relationships were just as likely to defend as they were to stay quiet about the cyberbullying. Thus, in collectivist societies, inaction may be perceived as an attempt to maintain group harmony (Huang and Chou, 2010), which may allow one to morally disengage (Roccas et al., 2006). However, it remains unclear whether the same pattern of results is found among Middle Eastern youth.

Other research suggests that the computer-mediated world is known to be a distinct community within itself, and the technological affordances of social media have resulted in disinhibited, individualistic behavior among Middle Eastern users (Ghanem et al., 2013). Nonetheless, it remains unclear how youth from collectivist societies perceive theory-driven determinants of positive cyber-bystander behavior. As such, the current study aims to better understand the dynamic relationship between the determinants of bystander action by investigating how behavioral determinants (e.g., the relationship between bystander and perpetrator) affects the moral disengagement process for cyber-bystander behavior among individuals from both Canadian and Iranian contexts.

Theories of Bystander Intervention

While bystanders can take on several roles in the online world, there is a consensus that bystanders can be categorized into at least 3 distinct roles: outsiders (e.g., doing nothing), assistants (e.g., joining in on the cyberbullying), and defenders (e.g., actively intervening; Salmivalli, 1999; Quirk and Campbell, 2015; Song and Oh, 2018; Pepler et al., 2021). Nonetheless, research continues to demonstrate that most cyber-bystanders maintain their role as outsiders and remain inactive in intervening (Quirk and Campbell, 2015; Allison and Bussey, 2016; DeSmet et al., 2016; Erreygers et al., 2016; Olenik-Shemesh et al., 2017; Song and Oh, 2018).

To better understand the factors that facilitate positive bystander intervention in cyberbullying, DeSmet et al. (2018) merged behavior change theories (e.g., the Theory of Reasoned Action and the Theory of Planned Behavior) with the Social Cognitive Theory to examine how cyber-bystander behavior changes across moral, social, behavioral, and cognitive domains. This multi-faceted analysis of bystander intervention has allowed us to better understand the complex interplay between the various determinants of bystander intervention. The behavior change theories highlight how background variables (e.g., past behavior, stereotypes, stigma, personality, culture) and environmental factors influence attitudes, beliefs, behavioral intentions, and actual behavior (DeSmet et al., 2018). The Social Cognitive Theory adds the influence of moral disengagement, which is a set of psychological mechanisms that are used to justify immoral behavior to avoid negative emotions and social disapproval (Bandura, 1998; Bandura et al., 2002). The cognitive mechanisms associated with moral disengagement involve re-framing negative behavior in a positive way, minimizing one's responsibility, distorting consequences, and blaming the victim (Bandura et al., 2002; Gini et al., 2020). These mechanisms overlap with the mechanisms that are associated with passive bystander behavior, such as diffusion of responsibility, distortion of consequences, and attribution of blame (Lo Cricchio et al., 2021). As such, the role of moral reasoning in cyber-bystander intervention has received increased attention.

Moral Reasoning in Cyber-Bystander Behavior

Several studies have investigated the relationship between being a cyber-bystander and the cognitive mechanisms of moral disengagement (Bussey et al., 2015, 2020; Allison and Bussey, 2016; Conway et al., 2016; DeSmet et al., 2016; Leduc et al., 2018; Song and Oh, 2018; Luo and Bussey, 2019). These studies found that low levels of moral disengagement (i.e., high moral responsibility) were positively associated with cyber-bystander defending behavior, and this was particularly true when youth also had: high self-efficacy for defending (Bussey et al., 2015, 2020), high individual morality (Allison and Bussey, 2016), awareness of contextual cues (Luo and Bussey, 2019), lower anti-social conformity, and a bad relationship with the bully (Song and Oh, 2018). These findings highlight the additive influence of the various determinants of cyber-bystander behavior in combination with low moral disengagement in facilitating positive bystander behavior.

In addition to cognitive mechanisms, moral reasoning can also involve emotional processes (Decety et al., 2012; Malti and Ongley, 2014; Conway et al., 2016). Morally responsible emotions, such as guilt (i.e., self-attributions to one's behavior) and shame (self-attributions to one's self; Eisenberg, 2000), can cause individuals to experience negative self-evaluations to the point that they feel a sense of moral responsibility toward another, which then predicts engagement in prosocial behavior (Eisenberg et al., 2010). In the context of traditional bullying, youth who score higher in morally responsible emotions are more likely to defend bullied victims (e.g., Gini et al., 2007; Caravita et al., 2009; Barchia and Bussey, 2011). In contrast, morally disengaged emotions, such as hubristic pride (i.e., self-attributions to one's ability) and authentic pride (i.e., self-attributions to one's effort; Ttofi and Farrington, 2008), are negatively associated with defending behavior among bystanders (e.g., Menesini et al., 2015; Thornberg et al., 2015; Romera et al., 2019). However, few studies have examined moral emotions in relation to cyber-bystander behavior. The technological features of online contexts minimize context-specific emotional cues (e.g., text-based communication, anonymous users), which can distort one's interpretation of cyber-aggression and change the moral emotions that are activated (Pornari and Wood, 2010; Runions and Bak, 2015).

Despite these differences between traditional and online bullying, only a handful of studies have examined moral emotions in relation to cyber-bystander behavior. Overall, these studies found that youth experienced morally responsible emotions when adopting the perpetrator's perspective, but morally disengaged emotions when considering the perspective of a bystander (Shultz et al., 2014; Conway et al., 2016; Leduc et al., 2018; Shohoudi Mojdehi et al., 2019; Tong and Talwar, 2020). While some researchers have found developmental trends for the moral attributions of cyber-bystander behavior (e.g., Conway et al., 2016; Leduc et al., 2018), Tong and Talwar (2020) found that age does not predict moral judgments of cyber-bystander behavior (Tong and Talwar, 2020), and concluded that bystander behavior may be better explained by factors other than age.

However, fewer studies have examined how moral emotions change in relation to the other determinants of cyber-bystander behavior. To illustrate, Pabian et al. (2016) found that continued exposure to cyberbullying lowers levels of morally responsible emotions, for bystanders. However, Shultz et al. (2014) found that continued exposure did not affect cyber-bystander responses if the bystander socially identified with the victim, which resulted in increased levels of empathy. In contrast, Tong and Talwar (2020) found that youth who had been bully-victims in the past experienced greater morally disengaged emotions when taking on the perspective of the bystander in comparison to victims, which suggests that bully-victims can identify with both parties and feel positively that they are in a position of power as they are not the ones being victimized. Taken together, these studies suggest that group membership plays a role in the moral emotions experienced by cyber-bystanders. However, more research is needed to better understand the role of in-group membership in the emotional processes that influence moral engagement with cyber-bystander behavior.

The Role of In-group Membership in Cyber-Bystander Behavior

Relational Processes

A bystanders' willingness to intervene in cyberbullying can be influenced by their relationship with those involved (Jones et al., 2011; Desmet et al., 2012; DeSmet et al., 2014, 2016; Machackova et al., 2013; Bastiaensens et al., 2014; Price et al., 2014; Coyne et al., 2019). When an individual has a relationship with another person, they are considered as being part of the same “in-group”. Typically, individuals who share group membership tend to converge on their values, attitudes, beliefs, and even personality variables. Adhering to group norms and maintaining close in-group relations is especially critical in adolescence, as peer acceptance is a significant indicator of healthy development and overall well-being (Bukowski et al., 1993).

In-group relations may also influence teenagers' perceptions of cyberbullying and determine their cyber-bystander behavior (Jones et al., 2011). Teen bystanders are more likely to join in on the cyberbullying when other bystanders were their friends (Bastiaensens et al., 2014; Coyne et al., 2019). Similarly, DeSmet et al. (2014) found that 12- to-16-year-old participants expected bystanders who were both a friend of the victim and high in popularity to take on a defending role. To our knowledge, only Machackova et al. (2013) has looked at how the bystander's relationship to the perpetrator influences defending behavior among adolescents. Specifically, they found that bystanders who were friends with the bully were less supportive of victims, regardless of their emotional response to the cyberbullying (Machackova et al., 2013). However, it remains unclear how one's relationship to the perpetrator affects the mechanisms of moral disengagement. The present study will further examine this relationship in a cultural context.

Cultural Context

The research examining the relationship with involved others provides insight into how the norms of a specific sub-culture (i.e., the peer-group) influence cyber-bystander behavior. However, online behavior is influenced by several sub-cultures, many of which have conflicting norms (Brody, 2021). One sub-culture that is often neglected in the literature is ethnicity, with only a few studies considering the norms and values of specific ethnic groups. There has been some investigation of cyber-bystander behavior in East Asian and South Asian cultural contexts (Huang and Chou, 2010; Park et al., 2014; Yang et al., 2014; Bhat et al., 2017; Ma and Chen, 2019; Leung, 2021), South American cultural contexts (Ferreira et al., 2016; Souza et al., 2018), European cultural contexts (DeSmet et al., 2014, 2016, 2018; Bastiaensens et al., 2016; Machackova and Pfetsch, 2016; Piccoli et al., 2020; Thornberg et al., 2021), and North American cultural context (e.g., Taylor et al., 2019; Pepler et al., 2021). This research demonstrates the differences in norm expectancies of bystander inaction across cultures. For example, Ma and Chen (2019) found that “doing nothing” (i.e., being an outsider as a bystander) aligned with the perceived collectivistic values of Asian cultures; whereas, bystander intervention in individualistic contexts relies on personal factors, such as individual attitudes and social factors, such as one's relationship to the victim (e.g., Bastiaensens et al., 2014). Only a handful of studies have conducted cross-cultural analyses (Lapidot-Lefler and Barak, 2012; Ma and Bellmore, 2016; Wright et al., 2016; Lee et al., 2017; Wachs et al., 2019; Cook et al., 2021); however, none were in relation to how the socio-cultural context (e.g., culture and relationship between bystander and perpetrator) specifically influences the processes involved with moral disengagement.

Cyber-Bystander Behavior in Middle Eastern Cultures

Middle Eastern cultures have been neglected in these investigations, which is problematic given the increased usage of technology in Middle Eastern countries (Shadmanfaat et al., 2019). In contrast to communist countries in East Asia, such as China, Middle Eastern countries' political and cultural construction have been influenced by Islam (Rabiei, 2013; Rezapour et al., 2019). Language diversities and political conflicts in the Middle East also have shaped socio-cultural differences within this region (Rabiei, 2013). Therefore, collectivistic roots and values in the Middle East can be different from other parts of Asia.

Only a handful of studies have examined cyber-bystander behavior in Middle Eastern communities. For instance found, Lapidot-Lefler and Barak (2012) that adolescent girls who were Israeli-Arab were more likely to bully others online and less likely to intervene as a bystander than adolescent girls who were Israeli-Jewish. The authors attributed this finding to the lack of supervision and control in online contexts, which allows Israeli-Arab girls to remove themselves from their collectivist identity and exercise their freedom of speech. Similarily, Shohoudi Mojdehi et al. (2019) found that Iranian youth attributed less morally responsible emotions (shame and guilt) in comparison to Canadian and Chinese youth. These findings highlight how Iranian youths' persectives may be shaped by different cultural influences than other Eastern cultures like China. However, more research is required to fully understand how the determinants of cyber-bystander behavior dynamically interact to influence the moral attributions assigned by Middle Eastern Youth.

Since Middle Eastern communities are high in collectivism, they tend to be emotionally interdependent and prioritize the welfare of the community, societal structure, and power differences (Schwartz, 1994). Many Middle Eastern countries overlap in collectivist values such as: (a) their high-power dynamics where one does not question authority figures (Samovar et al., 1981; Singhapakdi et al., 1999); (b) their emphasis on tradition and preservation of the past (Kluckhohn and Strodtbeck, 1961); (c) and their conformity to cultural norms and Islamic law (e.g., well-defined societal roles and group harmony; Ghanem et al., 2013). Previous research (Nesdale and Naito, 2005) has found that similar collectivist values have contributed to less bystander action in traditional bullying, especially when the bystander had a relationship with the perpetrator.

However, as technology evolves, these cultural characterizations are less pronounced in computer-mediated environments. The low-context communication style of online environments (e.g., instant messaging, clicking like) are a mismatch with high-context, collectivist cultures, which typically rely on non-verbal communication and contextual cues such as background variables, status, and hierarchy (Hall, 1989; Croucher et al., 2012). With the rise of the technological era, youth from high-context collectivist cultures are now engaging online in more depersonalized and uninhibited ways (Ghanem et al., 2013). Accordingly, that there has been a shift in communication style among online communities in the Middle East (Ghanem et al., 2013). Accordingly, researchers have been challenged with the task of re-examining how culture plays a role in online contexts, especially in relation to other contextual cues, such as one's relationship to those involved.

While there are several Middle Eastern communities, the current study will focus on Iranian culture. As one of the world's oldest civilizations in this area (Barrington, 2012), Iran has several religions, languages, and revolutions in its history (Abrahamian, 2021). Thus, research from this area has a specific context (secular-religious) and could add new pieces of literature to cross-cultural studies. In addition, both cyberbullying rates and passive bystander behavior are steadily increasing in Iran (Razjouyan et al., 2018). Understanding how one's context influences perceptions of bystander intervention can ensure that cyberbullying intervention programs are both culturally sensitive and inclusive.

Current Study

The current study examined how Iranian youth evaluate cyber-bystander behavior in comparison to Canadian youth. This study also investigated how the interaction between background variables (e.g., cultural group) and contextual variables (e.g., relationship to the perpetrator) influences the moral disengagement process in cyber-bystander behavior. Consistent with previous research (Salmivalli, 1999; Quirk and Campbell, 2015; Song and Oh, 2018; Pepler et al., 2021), the types of bystander action that are presented in this study will be: (a) positively defending the victim; (b) assisting the cyberbully; and (c) being an outsider (i.e., doing nothing). A 3 (Bystander Type: defend, assist, outsider) x 2 (Relationship: acquaintance, friend) within subjects design was used among youth in both Canadian and Iranian cultures. The current study controlled for previous experiences of victimization and aggression because of previous research (e.g., Cao and Lin, 2015) that suggests previous experiences are a determinant of bystander behavior.

Hypotheses were made for moral evaluations (hypotheses 1a, 1b, and 1c) and the moral disengagement process (hypotheses 2a, 2b, and 2c) in Canadian and Iranian youth according to the three types of bystander action (i.e., assistants, outsiders, defenders; e.g., Pepler et al., 2021), and the relationship of the bystander to the perpetrator of cyberbullying (i.e., acquaintance, friend).

First, based on previous research (Shohoudi Mojdehi et al., 2019), we hypothesized that both Canadians and Iranians will morally evaluate assisting behavior as negative, regardless of whether the bystander is a friend or acquaintance to the perpetrator (hypothesis 1a). Based on Shohoudi Mojdehi et al. (2019), we also predict that Canadians will evaluate outsider behavior more negatively than Iranians, regardless of their relationship to the perpetrator (hypothesis 1b). Finally, we predict that both cultural groups will evaluate defending behavior positively when the bystander is friends with the perpetrator, since both cultures strive toward reducing in-group conflict (e.g., Nesdale and Naito, 2005; Ghanem et al., 2013; Machackova et al., 2013; Bastiaensens et al., 2014; hypothesis 1c).

With regard to moral disengagement, we predict that Canadians will feel a stronger sense of moral responsibility for assisting behavior, as previous research suggests Canadians tend to be more confrontational than Iranians (Hall, 1989; Ma and Bellmore, 2016; Hamelin et al., 2018; hypothesis 2a). Moreover, since both Eastern and Western cultures have shown to be morally disengaged in response to outsider behavior (e.g., see Zhao and Yu, 2021 for a review), we expect less moral responsibility for outsider behavior among both cultural groups (hypothesis 2b). Finally, we expect Iranians to be morally disengaged toward defending behavior when the bystander is an acquaintance to the perpetrator since collectivist cultures tend to be conflict avoiding (e.g., Huang and Chou, 2010; Ghanem et al., 2013; Ma and Chen, 2019; hypothesis 2c).

Method

Participants

One hundred and nineteen participants from Canada (N = 60) and Iran (N = 59) completed the study. Participants were between the ages of 8-to-15 years old (M = 11.33 years, SD = 1.63 years; 63.3% female). Participants were matched for age and gender across country of origin. For both Iranian and Canadian participants, 63.3% of the sample was female.

Selection of participants was done through convenience sampling. In Iran, participants were recruited from schools in [city name removed for confidentiality purposes]. In Canada, participants were recruited from university research recruitment database of families with children and adolescents in the [city name removed for confidentiality purposes] area of Quebec, Canada.

Materials

Vignettes

Participants read a total of six gender-matched vignettes (Appendix) depicting cyberbullying scenarios that involved a perpetrator, a victim and a bystander. Scenarios varied according to two independent variables. First, the bystander's relationship with the perpetrator was manipulated as either being acquaintances (3 scenarios) or friends (3 scenarios). Second, the bystander's response to the cyberbullying situation varied according to whether they assisted the cyberbully (2 scenarios), acted as an outsider (e.g., doing nothing) in response to the cyberbullying (2 scenarios), or defended the victim (2 scenarios). As in previous cyberbullying research (e.g., Conway et al., 2016; Leduc et al., 2018), vignette methodology was chosen to allow better control and consistency over the independent variables and scenarios that participants are asked to evaluate (Aguinis and Bradley, 2014).

Evaluations of Bystander Behaviors

After reading each scenario, participants responded to the question “What do you think about [bystander's] behavior?” They answered on a scale from −2 (very bad) to 2 (very good) to indicate how negatively or positively they perceived the bystander's behavior in the vignette.

Moral Emotions Attributed to Bystander Behaviors

Participants also responded to the question “How would you feel if you were [bystander]?” Participants answered on an ordinal scale of 0 (bad), 1 (neither good or bad) to 2 (good). Similar to past research (e.g., Menesini et al., 2015; Conway et al., 2016), participants were provided with example emotions for feeling “bad” of feeling sad, angry, or guilty, and for feeling “good” as feeling happy, or proud. For vignettes in which bystanders who assisted the cyberbully or who were passive (i.e., did not get involved), feeling bad was coded as morally responsible and feeling good was coded as morally disengaged. Attributed emotions were coded on the same ordinal scale of 0 (morally disengaged) to 2 (morally responsible), with 1 (feeling neither good or bad) indicating moral indifference or neutrality. Coding was reversed for vignettes where the bystander engaged in positive bystander behavior (i.e., supported the victim).

Cyber-Aggression and Cyber-Victimization Scale

This scale was used to measure participants previous experiences as perpetrators of cyberbullying (Cyber-Aggression subscale) and victims of cyberbullying (Cyber-Victimization subscale). Each subscale included 9 items for which participants were asked to reflect on their experiences in the last 2 months. Example items include “I sent insults on social media” (Cyber-Aggression) and “I received rude comments on a website” (Cyber-Victimization). Responses were scored on a scale of 0 (never) to 4 (several times a week) for a possible total score of 36 per subscale. All values were anchored, and the subscale showed good reliability (α = 0.80).

Procedure

This study was approved by [name removed for confidentiality purposes] University's Research Ethics Board (Tier III). The study was conducted in-person. After their parents provided informed consent, participants provided assent prior to completing the Cyber-Victimization and –Aggression questionnaire, and the vignettes. The order of measures and vignettes was completely counterbalanced across subjects, and the questions asked for the vignettes followed the fix order presented above.

Results

Country differences in moral evaluations of bystander responses were examined through a repeated measures analysis of covariance (ANCOVA), while controlling for previous experiences of cyber-aggression and cyber-victimization. Friedman's test was used to examine participants' attributions of moral responsibility and disengagement as a function of their country of origin (Canada or Iran). Friedman's test is a non-parametric alternative to a repeated-measures ANOVA that was chosen because our dependant variable (attributions of moral responsibility and moral disengagement) is measured on an ordinal scale (Gignac, 2019). Both analyses are described in detail below. Descriptive statistics for moral evaluations and moral disengagement scores are in Table 1. Since developmental differences were not significant, the subsequent analyses controlled for age.

TABLE 1

Table 1. Means (SDs) for moral evaluation and moral disengagement scores according to country of origin, bystander response and bystander's relationship to the perpetrator.

The results are presented according to the hypotheses presented above. First, findings from the repeated measures ANCOVA for moral evaluations (hypotheses 1a, 1b, and 1c) of different types of cyber-bystander action according to country of origin and the bystander's relationship to the perpetrator are presented. Second, findings from Friedman's non-parametric repeated measures test for participants' attributions of moral responsiblity and disengagement (hypotheses 2a, 2b, and 2c) toward different types of bystanders according to country of origin and the bystander's relationship to the perpetrator are presented.

Moral Evaluations of Bystander Behaviors

A three-way mixed repeated measures ANCOVA examined participants' moral evaluations of bystander behaviors in cyberbullying (dependent variable) with bystander response (assist, defend, be an outsider) and the bystander's relationship with the perpetrator (friend, acquaintances) as within-subject variables (independent variables), and the participants' country of origin (Canada or Iran) as a between-subject variable (independent variable). To control for the possible bias of participants' lived experiences, past involvement in cyberbullying as victims and/or perpetrators was entered as a covariate. Mauchly's test of sphericity, $χ_{(2)}^{2}$ = 8.85, p = 0.01 was violated and the more conservative Greenhouse-Geisser was used when assessing tests of within-subject effects. All other statistical assumptions were met, including normality and Levene's test for equality of variances (p < 0.05). Bonferroni corrections were applied to alpha levels for all follow-up tests.

There was a statistically significant three-way interaction between the bystander's response, the bystander's relationship to the perpetrator and country of origin on participants' moral evaluations of bystander behaviors in the scenarios, F_{(1.86, 206.07)} = 3.41, p = 0.04, partial eta² = 0.03, regardless of participants' previous experiences with cyber-aggression and cyber-victimization. There were no statistically significant differences between Iranians and Canadians for bystanders that assisted the perpetrator, regardless of their relationship. In support of hypothesis 1a, they were both more negative in their evaluations. Moreover, hypothesis 1b was partially supported. Canadians evaluated the bystander's behavior more negatively than Iranians in their evaluations of bystanders who remained passive in response to the cyberbullying situation (i.e., they were an outsider), p < 0.01, but this was when the bystander was a friend to the perpetrator (M = −0.90; SD = 0.82). Conversely, Iranians were more neutral or indifferent in their evaluations of the same bystanders (M = −0.07; SD = 1.30). Finally, with regard to bystanders who defended targets of cyberbullying, hypothesis 1c was also partially supported. The relationship with the perpetrator did not have a significant effect on the evaluations of Canadians, p = 0.132, but Iranians provided more positive moral evaluations, p < 0.01, when the bystander was friends with the perpetrator (M = 1.47; SD = 1.02) than when they were acquaintances (M = 0.45; SD = 1.15).

Findings from the repeated measures ANCOVA revealed additional findings. There were statistically significant two-way interactions between the country of origin and the bystander's response, F_{(1.89, 209.24)} = 4.53, p = 0.01, partial eta² = 0.04. Specifically, Canadians evaluated bystander behaviors more negatively than Iranians when the bystander did nothing in response to the cyberbullying. However, when the bystander defended the victim, Canadians provided more positive evaluations than Iranians, p < 0.01. There was also a two-way interaction between the participants' country of origin and the bystander's relationship with the perpetrator, F_{(1, 111)} = 30.67, p < 0.01, partial eta² = 0.22. While there were no significant differences in Canadians' moral evaluations, p = 0.132, Iranians gave more positive evaluations to the behaviors of bystanders who were friends with the perpetrator than those that were acquaintances, p < 0.01. Iranians were more negative toward bystanders who were acquaintances to the perpetrator.

In addition, there were significant main effects of the bystander's response, F_{(1, 209.24)} = 294.79, p < 0.01, partial eta² = 0.73, and of the bystander's relationship with the perpetrator, F_{(1, 111)} = 14.39, p < 0.01, partial eta² = 0.12, on participants' moral evaluations. For bystander responses, regardless of country of origin and relationship with the perpetrator, participants provided more negative evaluations of bystanders who assisted the perpetrator and those who did nothing and more positive evaluations of bystanders who defended victims of cyberbullying, p < 0.01. With regard to the bystander's relationship with the perpetrator, regardless of the bystander's response to the cyberbullying event and country of origin, participants provided more negative evaluations to acquaintances depicted in the cyberbullying scenarios, than they did to friends, p < 0.01.

No statistically significant interaction between the bystander's response and their relationship with the perpetrator, F_{(1.86, 206.07)} = 2.33, p = 0.10, partial eta² = 0.02.

Moral Emotions Toward Bystander Behaviors

A series of Friedman's non-parametric repeated measures ANOVA tests were run to examine differences in moral emotion attributions (dependent variable) across bystander responses and the bystander's relationship with the perpetrator (independent variables). Given the nature of our variables, data was split between country of origin (Canada or Iran) (independent variable). For both sets of analyses, Bonferroni corrections for multiple comparisons, and post-hoc tests were ran when appropriate. Data was also tested for normality and the assumption was met. The following results are presented by country of origin of the sample (Canada and Iran) given that the file was split in this way for analyses. Table 2 summarizes the 9 comparisons per country of origin and median ranks.

TABLE 2

Table 2. Summary of comparisons (and median ranks) of moral emotion attributions to bystander behavior according to country and the bystander's relationship with the perpetrator.

Canadians' Moral Emotion Attributions

When the bystander was an acquaintance of the perpetrator, distributions of moral emotion scores were significantly different between bystanders who assisted the perpetrator and those who did nothing, $χ_{(1)}^{2}$ = 5.00, p = 0.03. Specifically, in partial support of hypotheses 2a and 2b, Canadians' moral emotions scores ranked higher in moral responsibility for bystanders who assisted the perpetrator (Mdn = 1.54) than those who did nothing (Mdn = 1.46). There were also significant differences when comparing bystanders who did nothing and those who defended victims. Specifically, in partial support of hypotheses 2b and 2c, Canadians' attributed moral emotion scores ranked higher in moral responsibility toward bystanders who defended victims (Mdn = 1.55), than those who did nothing (Mdn = 1.45), $χ_{(1)}^{2}$ = 6.00, p < 0.01. In additional analyses comparing distributions of moral emotion scores between bystanders who assisted the perpetrator and those who defended the victim, there were no significant differences among Canadians, $χ_{(1)}^{2}$ = 1.00, p = 0.32.

Regarding the influence of relationship with the perpetrator (hypothesis 2c), when bystanders were friends of the perpetrator, there were no significant differences among Canadians in their moral emotion attributions between bystanders who assisted the perpetrator and those who did nothing, $χ_{(1)}^{2}$ = 3.00, p = 0.08, those who assisted the perpetrator and those who defended the victim, $χ_{(1)}^{2}$ = 2.67, p = 0.10, and between those who did nothing and those who defended the victim, $χ_{(1)}^{2}$ = 0.33, p = 0.56.

In additional analyses comparing distributions of moral responsibility scores across bystander responses (assisting the perpetrator, doing nothing, or defending the victim), and the bystander's relationship with the perpetrator (acquaintance or friend), no significant differences were found for bystanders who assisted the perpetrator among Canadians' moral emotion scores, $χ_{(1)}^{2}$ = 0.67, p = 0.41, or for bystanders who defended victims, $χ_{(1)}^{2}$ = 0.33, p = 0.56. However, significant differences were found for bystanders who did nothing. Specifically, Canadians' moral emotion scores ranked higher in moral responsibility when bystanders who were friends with the perpetrator and did nothing (Mdn = 1.55), than when bystanders were acquaintances with the perpetrator and did nothing (Mdn = 1.45), $χ_{(1)}^{2}$ = 4.50, p = 0.03.

Iranians' Moral Emotion Attributions

For Iranian participants, hypotheses 2a and 2b were not supported. When bystanders were acquaintances of the perpetrator, there were no differences among the moral emotion scores for bystanders who assisted the perpetrator and bystanders who did nothing, $χ_{(1)}^{2}$ = 0.43, p = 0.51. Moreover, as in the Canadian sample, there were significant differences when comparing bystanders who did nothing and those who defended victims. However, hypothesis 2c was not fully supported for Iranians, $χ_{(1)}^{2}$ = 4.26, p = 0.04, who also ranked higher moral responsibility scores toward bystanders who defended victims (Mdn = 1.58), than bystanders who did nothing (Mdn = 1.42). In additional analyses, distributions of moral emotion scores between bystanders who assisted the perpetrator and those who defended the victim, were not significant among Iranians, $χ_{(1)}^{2}$ = 1.92, p = 0.17.

Regarding the influence of relationship with the perpetrator (hypothesis 2c), when bystanders were friends with the perpetrator, significant differences were found between distributions of moral emotion scores toward bystanders who assisted the perpetrator and bystanders who did nothing, $χ_{(1)}^{2}$ = 6.40, p = 0.01. Iranians' moral emotion scores ranked higher in moral responsibility for bystanders who did nothing (Mdn = 1.58), than for bystanders who assisted the perpetrator (Mdn = 1.42). There were also significant differences between distributions of moral emotion scores toward bystanders who assisted the perpetrator, and those who defended the victim, $χ_{(1)}^{2}$ = 7.36, p = 0.01. Iranians' moral emotion scores ranked higher in moral responsibility toward bystanders who defended the victim (Mdn = 1.58), than those who assisted the perpetrator (Mdn = 1.42). However, there were no differences between distributions of Iranians' moral emotion scores for bystanders who did nothing and bystanders who defended the victim, $χ_{(1)}^{2}$ = 0.33, p = 0.56.

Finally, in additional analyses, when comparing distributions of moral responsibility scores across bystander responses (assisting, defending, outsider) and the bystander's relationship with the perpetrator (acquaintance, friend), no significant differences were found for bystanders who assisted the perpetrator among Iranians' moral emotion scores, $χ_{(1)}^{2}$ = 0.20, p = 0.66. However, as among Canadians, significant differences were found for bystanders who did nothing. Specifically, Iranians' moral emotion scores, $χ_{(1)}^{2}$ = 13.25, p < 0.01, ranked higher in moral responsibility for bystanders who were friends with the perpetrator and did nothing (Mdn = 1.64), than those who were acquaintances with the perpetrator and did nothing (Mdn = 1.36). Finally, for bystanders who defended victims, while no significant differences were found for Canadians, there were significant differences for Iranians, $χ_{(1)}^{2}$ = 5.44, p = 0.02. Iranians' moral emotion scores ranked higher in moral responsibility for friends who defended victims (Mdn = 1.56) than acquaintances who defended victims (Mdn = 1.44).

Discussion

The current study aimed to examine cross-cultural differences in the way Canadian and Iranian youth morally evaluate cyber-bystander behavior depending on the bystander's behavior (assist, defend, be an outsider) and the bystander's relationship to the perpetrator (acquaintance, friend). In addition, participants were asked to reflect on the emotions they experienced in relation to the bystander's behavior, and then these emotional attributions were coded as morally engaged, neutral, or morally disengaged.

Moral Evaluations

We predicted that both Canadians and Iranians would negatively evaluate assisting behavior (e.g., helping the cyberbully), regardless of whether the bystander is a friend or acquaintance to the perpetrator (hypothesis 1a). Our findings were consistent with this predicition, as youth in both cultures do not support a peers' decision to join in on the cyberbullying, even if the bystander is friends with the perpetrator. Although past research (Desmet et al., 2012; DeSmet et al., 2014, 2016) has found that adolescents will stick up for a friend even if the circumstances around the interpersonal conflict are unclear, the current findings suggest that friends of the perpetrator may be doing so against their moral convictions. Since this pattern of results was demonstrated cross-culturally, educational bystander intervention programs could standardize a component that is aimed at acknowledging this discrepancy between adolescents' moral beliefs and their behaviors in order to facilitate actions that are consistent with one's beliefs.

Based on previous findings (e.g., Shohoudi Mojdehi et al., 2019), we predicted that Canadians would evaluate outsider behavior (i.e., doing nothing in response to the cyberbullying) more negatively than Iranians, regardless of their relationship to the perpetrator (hypothesis 1b). The results were slightly more nuanced than expected as we found that Canadians evaluated outsider behavior negatively when the bystander was a friend, which is consistent with previous research (DeSmet et al., 2016), and further suggests there are higher expectations from friends to intervene (Price et al., 2014). In contrast, Iranians were more neutral or indifferent toward the same bystanders (friend who acted as an outsider). These findings are consistent with the findings of Huang and Chou (2010), who found a similar effect among Chinese youth, and further supports the idea that bystander inaction may be more socially acceptable in collectivist societies than individualistic societies (Samovar et al., 1981; Singhapakdi et al., 1999; Ghanem et al., 2013). Again, given the collectivistic values of exercising forgiveness and tolerance (Shohoudi Mojdehi et al., 2019), outsider behavior may be another instance of groupthink, as passive responses can be perceived as a way of maintaining group harmony and avoiding further conflict.

We also predicted that both cultural groups would evaluate defending behavior (e.g., helping the victim) positively when the bystander is friends with the perpetrator (hypothesis 1c), since most cultures value the importance of reducing in-group conflict (e.g., Cuhadar and Dayton, 2011). While our prediction was true for Iranians, there was no difference found in evaluations of defending behavior for friends and acquaintances of the bystander among Canadians. These findings may further support other research (e.g., Palomera et al., 2021) that suggests a youth's decision to engage in confrontational cyber-bystander behavior (e.g., defending) is shaped by their social values. Perhaps there was no difference in defending behavior among friends and acquintances for Canadians since individualist cultures value confrontational behavior moreso than Iranians (Hall, 1989; Ma and Bellmore, 2016; Hamelin et al., 2018). Whereas, Iranian youth may only engage in confrontational behavior when it is in the best interest of their in-group.

Moral Emotions

Given the more confrontational nature of Canadians, we predicted that Canadian youth would feel a stronger sense of moral responsibility for assisting behavior, due to urges to intervene, irrespective of the bystander's relationship to the perpetrator (hypothesis 2a). However, our results showed that relationship to the bystander mattered for both Canadians and Iranians. For Canadians, youth experienced increased moral responsibility when the bystander was an acquaintance and assisted the perpetrator compared to when they acted as an outsider. Since North American cultures tend to be “information-based” (Urschler, 2016), it is possible that Canadians remain passive and become morally disengaged in situations where they do not fully understand the context and when they do not have group membership. For this reason, it is possible that Canadians experience cognitive dissonance when they witness a person from an “out-group” join in on a potentially “in-group” conflict, as previous research has found that disagreements within social groups can be a source of cognitive dissonance (Matz and Wood, 2005). Thus, feelings of moral responsibility toward intervening on acquaintances who help perpetrators may reduce this experienced cognitive dissonance.

In contrast, Iranians felt less moral responsibility when the bystander was a friend and assisted in comparison to when a friend acted as both an outsider and defender. Since Iranians value group membership, it is possible they are more tolerant of assisting behavior if it is considered a norm or expectation of the group to maintain group harmony. According to Janis (1991), highly cohesive groups can engage in “groupthink” which is when people strive toward unanimity and solidarity to the point that it encourages self-censorship; groupthink could be an important barrier of bystander intervention in collectivist communities.

We expected cross-cultural similarities when it came to the moral emotions associated with outsider behavior (hypothesis 2b). Since both Eastern and Western cultures have shown to be morally disengaged in response to outsider behavior (Allison and Bussey, 2016; Brody and Vangelisti, 2016; Razjouyan et al., 2018; Lo Cricchio et al., 2021; Shariatpanahi et al., 2021), we expected less moral responsibility for outsider behavior among both cultural groups. The results somewhat support this prediction. Both cultural groups felt less moral responsibility when the bystander was an acquaintance to the perpetrator and did nothing in comparison to when the acquaintance defended the victim. However, both cultural groups felt increased moral responsibility when the bystander was friends with the perpetrator and acted as an outsider. These findings suggest that the peer group as a “sub-culture” may better facilitate youths' moral responsibility to intervene as a cyber-bystander in comparison to the larger cultural context. Consistent with previous research (e.g., Machackova et al., 2013; Shultz et al., 2014; Brody and Vangelisti, 2016; DeSmet et al., 2016), youth in both cultures felt more responsible for the behavior of their peer in-group members than those who were in their peer out-group. Since adolescence is a time where friendships play a key role in psychosocial and behavioral development (Vitaro et al., 2009), as well as in identity formation (Jones et al., 2014), these findings highlight the importance of educational intervention programs that include components that strengthen the behavioral styles and interactions between same-aged peers. While previous researchers have recommended targeting friendship to strengthen ties within the community to reduce vulnerability online (Patterson et al., 2017), our findings add that targeting friendship characteristics can also facilitate a sense of moral responsibility for prosocial bystander intervention.

In addition, even though Iranians attributed lower moral evaluations to outsider behavior among friends of the perpetrator, they still experienced emotions that elicited a sense of moral responsibility. Emotions are typically expected to be controlled in collectivist cultures; however, the experience of strong emotions among collectivist group members can be regarded as an interactive experience that mirrors the social context (Mesquita, 2001). Thus, when the Iranian participants exhibited emotions associated with moral responsibility, it may have provided them with a “situation specific clue” (p.5, Keshtiari and Kuhlmann, 2016) that oriented them to the contextual variable (i.e., relationship with perpetrator), which consequently influenced the way they responded. It is possible that the role of peer relationships as a contextual variable also cues empathy, which is an important determinant of prosocial bystander intervention.

For defending behavior (hypothesis 2c), we accurately predicted that Iranians felt increased moral responsibility for defending behavior when the bystander was a friend to the perpetrator in comparison to when they were acquaintances. Whereas, the same differences were not found among Canadians. Thus, despite the fact that Iranians felt less moral responsibility (and more tolerance) for assisting the cyberbully behavior within one's in-group, they simultaneously feel moral responsibility to intervene as a defender. This pattern of results further suggests that Iranians are willing to compromise group harmony within their in-group if there is a greater purpose, whereas this effect was not found among Canadian youth. Furthermore, the tolerance and forgiveness that is valued by Iranian culture could apply toward both the assistor and the defender, which would allow group members to feel comfortable enough to intervene. In contrast, Canadian youth who take on the perspective of a “friend of the perpetrator” may not agree with their friend's decision to cyberbully (as indicated by their moral evaluation), but they do not feel as morally responsible to address the cyberbullying as Iranians. It is possible that Canadians who are friends of perpetrators morally disengage when the cyberbullying occurs to allow the perpetrator to have autonomy over their choices, which is a value of individualistic cultures.

Limitations and Future Directions

The current study is not without its limitations. First, we did not examine the various cultural constructs that define ethnic groups. Future research could conduct cross-cultural comparisons in relation to cultural constructs that are on an ecological level (e.g., egalitarian commitment vs. conservatism; utilitarian involvement vs. loyal involvement; societal cynicism) and an individual level (e.g., uncertainty avoidance; power distance; time orientation; Matsumoto and Yoo, 2006). By investigating these constructs, we would achieve a richer understanding of the cultural determinants of bystander intervention. Researchers could also investigate how these cultural constructs relate to the mechanisms involved with moral disengagement to understand how the two processes interact to influence cyber-bystander behavior. Furthermore, the current study did not further examine the justifications behind youths' emotional attributions. A qualitative analysis of youths' moral justifications could allow for a better understanding of the social norms around bystander intervention in specific circumstances (e.g., when there is a relationship to those involved). Additionally, the current study relied on youths' evaluations of a hypothetical other; whereas future research could benefit from measuring youth's actual cyber-bystander behavior to these incidents either retrospectively (recording natural behavior online) or spontaneously (in-vivo/experimentally) and if their behavior varies among in-group and out-group members. Future research could also benefit from examining both constructive and aggressive forms of bystander intervention (e.g., Moxey and Bussey, 2020), as well as contradictory bystander roles (e.g., frenemies who change their allegiance according to peer context; Wójcik and Flak, 2021) across a wider range of cultural and gender groups. Finally, the current study was also limited in sample size, and future research could benefit from a sample with more statistical power.

Implications

The current study demonstrates that cultural values can be maintained in online interactions where the involved parties are known by one another. For this reason, it is critical that educational cyberbullying intervention programs that target bystander action are aware of these cross-cultural nuances and introduce strategies that are culturally sensitive. For example, educational curriculum designed as part of bystander intervention programs in collectivist societies could facilitate conversations about groupthink and the discrepancy between moral beliefs and moral actions to bring awareness to obstacles that prevent bystander action. Furthermore, since youth tend to interact with their same-aged classmates on social media, existing bystander intervention programs in educational settings could add content that strengthens the behavioral styles and interactions between same-aged peers in culturally appropriate ways. Many existing intervention programs fail to consider how cultural norms and peer norms influence the targeted outcomes (e.g., prosocial cyber-bystander action), and this study aims to encourage further research in this area.

Conclusion

Overall, the current study provides further insight into how peer-group norms and cultural norms dynamically interact to influence perceptions of cyber-bystander behavior in Middle Eastern and Canadian youth. Iranians morally evaluated assisting behavior similarly to Canadians; however, they felt less moral responsibility when the bystander was a friend to the perpetrator and assisted. On the other hand, Canadians felt moral responsibility when the bystander was an acquaintance and assisted. This pattern of results shows that the two cultural groups have different ways of interpreting negative cyber-bystander behavior, which may be largely influenced by their cultural values (e.g., information-based values in individualistic societies and solidarity-based values in collectivist societies). Furthermore, when the bystander was friends with the perpetrator and then acted as an outsider, Canadians assigned negative moral evaluations; whereas, Iranian youth are indifferent. Despite this difference in moral evaluations, both Canadians and Iranians felt morally responsible for the bystander's behavior, which suggest that the peer group as a “sub-culture” may better facilitate youths' moral responsibility in relation to passive bystander behavior, especially in comparison to the larger cultural context. Furthermore, being part of an “in group” among same-aged peers (e.g., friends with the perpetrator) may matter more for Iranians when it comes to defending behavior, whereas the relationship to the perpetrator mattered less for Canadians in defending behavior. This pattern of results suggests that Iranians may be more willing to compromise group harmony by defending the victim if it is among members of their own in-group. However, they will not facilitate inter-group conflict by defending against an out-group (i.e., acquaintance of the perpetrator). This study has important implications for the construction of culturally-appropriate bystander intervention programs for youth.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

The studies involving human participants were reviewed and approved by McGill Research Ethics Board-3 (REB-3). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

Author Contributions

VT and KL contributed to the conception and design of the study. KL and NK collected the data for this study and organized the database. PN and KL performed the statistical analysis. PN wrote the first draft of the manuscript. KL, NK, and VT wrote sections of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

1. ^Vignettes were gender-matched. Characters in the vignettes read to male participants included male characters.

References

Abrahamian, E. (2021). Iran Between Two Revolutions. Princeton, NJ: Princeton University Press.

Aguinis, H., and Bradley, K. J. (2014). Best practice recommendations for designing and implementing experimental vignette methodology studies. Organ. Res. Methods 17, 351–371. doi: 10.1177/1094428114547952