Contemporary research on cyberbullying has shifted from a simple bully-victim model to a three-part bully-victim-bystander framework by incorporating the role of bystanders (Rudnicki et al., 2023). Bystanders, as social participants, engage in interactions in cyberbullying and their actions are influenced by personal traits and the expectations of others involved. When cyberbullying occurs, bystanders often recognize the victim’s situation (Rudnicki et al., 2023), and their responses can serve as prosocial actions that help reduce the impact of bullying and discourage the bully (Torgal et al., 2023). However, research shows that over 60% of college bystanders who witness cyberbullying tend to remain neutral or avoid taking action (Pronk et al., 2019; Shen et al., 2024). This hesitation comes from the significant social costs of confronting bullies, including fears of retaliation or ostracism. Furthermore, some bystanders feel unprepared to intervene in cyberbullying situations, leading them to refrain from taking action (Blackwell et al., 2017).

1.1.4 Group Scenarios: The Moderating Role of Social Relationship Closeness (SRC)

Social contextual factors in cyberbullying may enhance bystanders’ cognitive recognition of incidents (Wang, 2022). Research has focused more on the immediate circumstances during cyberbullying incidents, considering the complex and evolving situations in the online environment, rather than exploring the influence of external social factors (Patel and Quan-Haase, 2024). Although some social situational factors have been identified as influential in behavioral development, these conclusions pose challenges when applied to measuring and assessing the behavior of university students (Herry et al., 2021).

Social-ecological theory highlights the crucial role of social situational factors, emphasizing the notion that the social environment constantly influences human behavior development. This suggests that institutions like schools, communities, and culture significantly shape individual cognition and behavior, thereby affecting the role and status of individuals within the social setting (Guo et al., 2021). College students’ cyberbullying behaviors, as a typical cognitive and behavioral issue, can be affected by social situational factors. McLeroy’s refined social-ecological model places individuals at the core of the ecological framework, forming the first level (intrapersonal) (McLeroy et al., 1988). At this level, individuals possess inherent personality traits and attitudinal predispositions. The second level includes interpersonal relationships, demonstrated through peer interactions among students. The third level involves institutional contexts, where higher education institutions serve as organizational settings for college students involved in cyberbullying incidents. Building on the social-ecological perspective to examine group-level factors (McLeroy et al., 1988) requires identifying assessment variables for social relationships. These variables should reflect interaction dynamics within societal frameworks (Barbalet, 2021a; Chen et al., 2013). Research in management and social psychology has indicated that social relationships serve as the foundational element in the relationship development process, with interpersonal interactions following thereafter (Hong et al., 2025; Peng, 1998). Due to group norms, scholars have increasingly focused on the meanings behind social relationships, incorporating concepts such as human relations and interpersonal ethics into their research. This has led to the development of the concept of social relationship closeness (Clarke and Saiket, 2023). Social relationship closeness reflects the norms and dynamics of interpersonal relationships, including cognitive-based trust and affective feelings (Hong et al., 2025). Greater social relationship closeness indicates stronger group ties.

It is important to clarify that in this study, “social relationship closeness” does not refer to specific interactions between individuals. Instead, it is understood as a broader social context, which should be interpreted differently from the concept used in a previous study (Clarke and Saiket, 2023). Social relationship closeness requires clear differentiation from interpersonal variables such as social proximity, social distance, and relational interdependence in its operational definition. Some scholarship focuses on individual-centered concepts of interpersonal relationships, emphasizing self-perceived or evaluated group social connections (Brewer, 1997). For example, social proximity refers to the psychological construction process of group identification, forming cognitive belonging to a social group through recognizing similarities between oneself and others (Bicchieri et al., 2022). Social distance is “the grades and degrees of understanding and intimacy which characterize pre-social and social relations generally” (Bogardus, 1992). Relational interdependence describes the tendency to conceptualize oneself through relationships with close others, highlighting evaluations of intimacy from an individual perspective (Rogoff, 1990). Although these concepts describe social dynamics within groups from varied angles, they focus on relational bond strength, power dynamics in social interactions, and identity alignment between individuals and groups (Brewer, 1997; Dixon, 2001).

By comparison, social relationship closeness refers to the interplay of individual interests, community values, and interpersonal connections (Chen et al., 2013). Social relationship closeness reflects levels of interpersonal interaction and emphasizes norms of fairness and reciprocity (Chen and Peng, 2008). This concept prioritizes collective relational interests, which help to shape social moral frameworks (Chen et al., 2013). Additionally, social relationship closeness encompasses non-utilitarian ways of relating to others, characterized by unconditional care and tolerance, even toward strangers (Hwang, 2009). In cyber incidents, social relationship closeness is strongly correlated with proactive intervention behaviors; college students demonstrate a greater willingness to protect victims and step in against bullying. Previous research has suggested that various social and cultural factors can influence bullying dynamics and bystander intervention (Wang, 2022). Therefore, examining how social relationship closeness affects bystander behaviors is of important theoretical value.

Research has mainly examined interactions among victims, bullies, and bystanders. Among youth populations, the difference between friendship interactions and social relationship closeness can be illustrated with the following scenario: When bystanders show low levels of dyadic friendship with cyberbullying victims (i.e., limited personal familiarity), higher social relationship closeness within peer groups still influences their emotional expectations to defend, support, or show concern for victims, rather than remaining outsiders, helping bullies, or reinforcing bullying behaviors (Clarke and Saiket, 2023; Wang, 2022). Even in completely unrelated relationships, bystanders adhere to principles of objectivity and fairness, which promote a more impartial way of thinking and expressing emotions, significantly influencing individuals’ emotional perceptions (Barbalet, 2021b). When social closeness is high, individuals tend to feel positively toward others, and this emotional tendency encourages deeper consideration of the group (Chen and Peng, 2008). Therefore, this study included hypotheses that these emotional tendencies, stemming from high social closeness, affect the likelihood of negative bystander behavior and shape individuals’ actions and responses in cyberbullying situations.

Hypothesis 6: Social relationship closeness negatively moderates the relationship between cyberbullying perpetration experience and reinforcing the bully.

Hypothesis 7: Social relationship closeness moderates the relationship between cyberbullying perpetration experience and assisting the bully.

Incorporating social relationship closeness into empirical research can provide a deeper understanding of how social relationships, as situational factors, influence bystander behavior in cyberbullying incidents. When bystanders confront cyberbullying involving strangers, they are likely to follow the behavioral guidelines and norms of social relationships, which may lead to more objective and fair decisions (Hamel et al., 2021). Individual actions are interpreted through the collective lens of group behavior. Empirical evidence has shown that adolescents who care more about group norms and collective interests tend to exhibit lower tendencies toward transgressive behavior (Chen et al., 2021). At high levels of social relationship closeness, prosocial behaviors driven by fairness and reciprocity are often more pronounced. In cyberbullying incidents, defending victims (i.e., acting as defenders in the BIM) and maintaining impartiality are prosocial choices, which are negatively associated with moral disengagement as a mediating factor (Bandura, 2024). Based on this, the current study suggests that social relationship closeness, as a form of public moral norms, can affect the relationship between online moral disengagement and bystanders engaging in prosocial behavior among university students. Therefore, the following research hypotheses were proposed (Fig. 1).

Fig. 1.

Research hypothesis model.

Hypothesis 8: Social relationship closeness moderates the mediating effect of moral disengagement between bystanders’ cyberbullying perpetration experience and assisting the bully.

Hypothesis 9: Social relationship closeness moderates the mediating effect of moral disengagement between bystanders’ cyberbullying perpetration experience and reinforcing the bully.

Thirty-two investigators were recruited to carry out the survey at 8 universities in Shanghai, China. All investigators conducted their investigations after receiving standardized training. The training systematically covered survey operation standards, academic ethics protocols, informed consent procedures, data confidentiality requirements, and ethical review guidelines. The first page of the questionnaire clearly explained this study’s purpose and data-use policies. All participants voluntarily signed the informed consent form. The questionnaire was distributed and collected online via the Wenjuanxing platform. Data collection took place from November 2023 to January 2024, with 617 college students participating. To ensure data quality, each investigator assigned a unique number to every respondent and conducted logical checks on the data. After completing the sample collection phase, we implemented rigorous quality control and screening procedures. Specifically, we excluded invalid responses characterized by excessively short response times, duplicate submissions from the same IP address, and excessive missing items in the questionnaire. This process resulted in a final valid sample size of 549 for analysis.

Data analysis was conducted using IBM SPSS Statistics (Version 26.0; IBM Corp., Armonk, NY, USA) and Mplus (Version 8.3; Muthén & Muthén, Los Angeles, CA, USA). First, descriptive statistics were calculated to examine sample means, standard deviations, and distribution characteristics. Bivariate correlation analysis and multicollinearity diagnostics were then conducted. Results indicated that all variance inflation factors (VIFs) were below 5 (Menard, 2001; Allison, 1999), suggesting no significant multicollinearity. Common method bias was assessed using Harman’s single-factor test, which revealed six factors with eigenvalues over 1. The first factor accounted for 31.32% of the variance, below the 40% threshold, confirming no significant common method bias (Podsakoff et al., 2003). Confirmatory factor analysis (CFA) showed an excellent model fit: $\chi$²/df = 3.243 (5), Root Mean Square Error of Approximation (RMSEA) = 0.064 (0.08), Tucker-Lewis Index (TLI) = 0.922 ($>$0.9), and Comparative Fit Index (CFI) = 0.930 ($>$0.9), aligning with the fit criteria proposed by Hu and Bentler (1999). Model construction proceeded in two stages. First, a mediation model without moderators was established, with fit indices ($\chi$²/df, RMSEA, TLI, CFI) evaluated. After mediation testing, interaction terms were introduced to test moderation effects. To analyze moderation, three levels were defined: the mean (M), mean plus one standard deviation (M+1SD), and mean minus one standard deviation (M–1SD). Effect size variations were assessed using the bias-corrected bootstrap method (5000 iterations, 95% CI). To ensure robustness, demographic variables such as sex, age, and students’ place of origin were controlled, as prior research indicates these factors may affect bystander decision-making processes (Zhao et al., 2023). This approach enhanced the reliability of the results.

The results of the mediation analysis revealed that $\chi$²/df = 3.378, RMSEA = 0.066, CFI = 0.924, and TLI = 0.917, indicating good model fit (Hu and Bentler, 1999). Table 2 indicates that cyberbullying perpetration experience was positively associated with moral disengagement ($\beta$ = 0.46, 95% CI = [0.39, 0.54]). Furthermore, moral disengagement was positively associated with remaining an outsider ($\beta$ = 0.46, 95% CI = [0.37, 0.54]), reinforcing the bully ($\beta$ = 0.53, 95% CI = [0.47, 0.60]), and assisting the bully ($\beta$ = 0.47, 95% CI = [0.40, 0.55]).

The indirect effect of moral disengagement was significant, constituting a partial mediating effect in the relationship between cyberbullying perpetration experience and both reinforcing the bully ($\beta$ = 0.25, 95% CI = [0.19, 0.31]) and assisting the bully ($\beta$ = 0.22, 95% CI = [0.16, 0.27]). These indirect effects accounted for 49% and 44% of the mediating effect. The indirect effect of moral disengagement was significant in this relationship ($\beta$ = –0.27, 95% CI = [0.22, 0.33]). These findings partially supported Hypotheses 3 and 4.

The results indicate that gender, age, and place of origin do not exert significant moderating effects. In Hypotheses 6 and 7, the study proposed that social relationship closeness moderates the relationship between cyberbullying perpetration experience and various bystander behaviors. The results (Table 3) showed a positive correlation (p 0.001) between cyberbullying perpetration experience and remaining an outsider, assisting the bully, and reinforcing the bully. Social relationship closeness was positively related to the relationship between cyberbullying perpetration experience and assisting the bully ($\beta$ = –0.06, 95% CI = [–0.10, –0.01]) and reinforcing the bully ($\beta$ = –0.06, 95% CI = [–0.10, –0.02]). It had a significant moderating effect on the positive relationship between moral disengagement and assisting the bully ($\beta$ = 0.10, 95% CI = [0.05, 0.16]) as well as reinforcing the bully ($\beta$ = 0.09, 95% CI = [0.04, 0.14]) (Fig. 2).

Fig. 2.

Moderated mediation model. *** p 0.001.

To further explore the indirect role of social relationship closeness in cyberbullying perpetration experience, moral disengagements, and various bystander behaviors, the present study performed path analyses with different levels of moderation. The results (Table 4) indicated that at higher levels of social relationship closeness (M+1SD), the positive association between cyberbullying perpetration experience and reinforcing the bully ($\beta$ = 0.22, 95% CI = [0.14, 0.29]) and assisting the bully ($\beta$ = 0.23, 95% CI = [0.15, 0.31]) was weakened. Conversely, the positive relationships between cyberbullying perpetration experience and reinforcing the bully ($\beta$ = 0.57, 95% CI = [0.50, 0.64]) and assisting the bully ($\beta$ = 0.52, 95% CI = [0.44, 0.60]) were strengthened through the indirect effect of moral disengagement. The 95% confidence intervals do not include 0.

The study tested the hypothesis that cyberbullying perpetration experiences would be positively related to negative bystander behaviors. The results supported this hypothesis, consistent with the previous findings (Giumetti et al., 2022). Smith (2019) argued that cyberbullying arises from a power imbalance, with bullying behaviors among college students possibly serving as a way to boost their social status and gain peer recognition. A recurring pattern of bullying indicates stability in the bully’s role. From the bully’s perspective, successfully engaging in bullying boosts their self-esteem (e.g., emotional satisfaction, approval from others, social privilege), which then strengthens their belief in their ability to bully again in the future (Chu et al., 2018). When the bully becomes a bystander in subsequent incidents, this sense of agency might lead them to make negative behavioral choices, such as supporting the bully or becoming an accomplice. Compared to traditional bullying, the online environment’s disinhibition and anonymity are more intense, which can decrease bystanders’ awareness of the possible consequences of their actions and result in less social responsibility (Runions et al., 2018).

In the present study, many bystanders tended to remain outsiders, and their experiences with bullying perpetration were not significantly or directly linked to this tendency. The study also indicated that higher levels of social relationship closeness led individuals to prioritize maintaining group relationships. When witnessing a cyberbullying incident, college students consider how their actions affect group perceptions and follow group moral norms, often choosing to stay silent to avoid alienating others or damaging relationships (Rovira et al., 2021). These findings raise further questions about the mechanisms within BIM models. Specifically, the present study suggested exploring potential factors such as social relationship closeness, which might influence the role of bystander experience in behavioral decisions, prompting bystanders to prefer inaction.

In the present study, students with a history of bullying perpetration exhibited higher levels of moral disengagement, which partially plays a mediating role in the relationship between bullying perpetration and two kinds of negative bystander actions. The analysis further indicated that moral disengagement served as a mediator in the association between bullying behavior and the tendency to adopt an outsider role. According to social cognitive theory, moral disengagement refers to an individual’s failure to self-regulate their moral conduct, leading them to engage in aggressive and antisocial behaviors (Shohoudi Mojdehi et al., 2019). Moral disengagement allows individuals to justify their negative actions and dismiss feelings of guilt. Consequently, cyberbullying situations are especially likely to trigger moral disengagement in bystanders, mainly because of the combination of bullying behavior and the anonymity the internet provides. Our findings supported previous research that identified moral disengagement as a key risk factor for cyberbullying (Gao et al., 2020).

Bandura et al. (2017) noted that moral disengagement is ultimately translated into behavior through self-regulatory mechanisms, a process that interacts closely with external environments. In bystander behavior in bullying, moral disengagement enables individuals to “rationalize their deviant behaviors”, making this process inherently an internal cognitive regulation. The present study explored how moral disengagement, as a psychologically cognitive attribute, relates to external situational factors that may serve as antecedents to moral disengagement. Specifically, first, the internet environment significantly reduces perceptions of bullying severity and victim suffering, lowering perpetrators’ guilt and empathy (Sun et al., 2025). Second, the online context weakens perceived group norms and reduces constraints on negative behaviors, facilitating aggressive and retaliatory actions (Colella et al., 2024). Therefore, as a unique situational cue, the online environment may more easily activate tendencies toward moral disengagement, justifying engagement in bullying and defending inappropriate actions.

Bystanders who have previously been involved in cyberbullying tend to have experiences that implicitly shape their thoughts about these events and help them rationalize their actions. When witnessing cyberbullying again, these experiences activate internal cognitive evaluations, leading them to perceive passive or reinforcing interventions (e.g., reinforce/assist the bully) as legitimate strategies. This history of participation strengthens social cognition and influences intervention decisions. Consequently, they may keep engaging in immoral behaviors, emphasizing the significant mediating role of moral disengagement between bullying perpetration and negative bystander actions.

This study highlighted the importance of social situational factors in cyberbullying. The moderating roles of offline social relationships and contextual factors in cyberbullying and bystander behavior warrant further exploration, given the diversity of bystander responses in different environments. Luo and Bussey (2019) recommended that cyberbullying research should prioritize individual-group relationship factors. The present study extended their argument, emphasizing the moderating role of social relationship closeness in the relationship between cyberbullying and bystander behavior. Higher levels of social relationship closeness were associated with stronger group bonds and moral norms with prosocial traits. Although cyberbullying perpetration was associated with negative behaviors, social relationship closeness weakened the connection between the factor and negative behaviors, highlighting the important moderating role of social situational factors. Specifically, high levels of social relationship closeness reduced the chances of future bullying incidents.

The findings indicated that social relationship closeness significantly weakened the direct positive link between bullying perpetration and negative bystander behavior. However, the indirect connection, mediated by moral disengagement, was strengthened. According to social-ecological theory, cyberbullying occurs within a multi-layered framework of social and ecological structures, including individuals, groups, and institutions, each influencing bystander behavior. High levels of social relationship closeness emphasize an individual’s self-perception within social relationships, often leading to the view that their behavioral decisions focus on the well-being of others or the organization (Chen and Peng, 2008). This construct is also linked to the prioritization of collective interests over individual ones (Chen et al., 2013). In high social relationship-closeness contexts, individuals are more inclined to adhere to moral norms of fairness and reciprocity (Chen and Peng, 2008). Even when possessing cognitive factors related to bullying involvement, bystanders may still engage in positive behaviors due to group pressure or moral constraints. For example, when bystanders do not have direct close relationships with victims, a high level of social relationship closeness at the group level still guides them toward protective/supportive tendencies toward victims (Clarke and Saiket, 2023). This shows that social relationship closeness, as a social contextual factor, influenced bystanders’ responsibility attribution. It prompts them to consider actions within collective interests. Ultimately, this weakens the direct relationship between bullying involvement and negative behaviors.

On the other hand, moral disengagement is a psychological process in which individuals justify negative behaviors through mental restructuring, offering cognitive reasons for actions that go against norms (Bandura, 2002). In situations with high social relationship closeness, although group norms reduce the direct influence of cyberbullying experience on negative behaviors, they may also activate the cognitive pathway for moral disengagement. For example, individuals may rationalize their own behaviors through “diffusion of responsibility” (blaming the group instead of the individual for bullying) or “moral justification” (interpreting their negative intervention behaviors as necessary to keep group harmony), thus reinforcing the mediating role of moral disengagement between bullying experiences and negative behaviors (Fissel et al., 2025). The research conclusion can further explore the aspect of “self-responsibility minimization” in moral disengagement (Fissel et al., 2025). Since engaging in negative intervention behaviors may cause individuals to violate group norms and face moral criticism, bystanders can rationalize their engagement in negative intervention behaviors. They do this by reducing their sense of responsibility and attributing it to others (Fissel et al., 2025). The present study enhanced understanding of situational influences and provided a detailed analysis of how concerns related to group affiliation can influence individuals’ behavioral choices.

The findings indicated that situational factors in cyberbullying are more varied than previously believed. In situations where social relationships are closer, bystanders’ decisions might be influenced by deeper considerations of social bonds and moral principles. Compared to Forsberg’s conclusion that “bystanders judge each other as ‘friends or non-friends’ to make intervention decisions” (Forsberg et al., 2014), University students seem to prioritize maintaining relationships within the group rather than relying solely on judgments of individual connections. As the results showed, this focus on group cohesion appears to reduce the likelihood of negative bystander behavior. Therefore, this study broadens the understanding of how interpersonal relationships influence bystander behavior in the context of cyberbullying among university students.

The theoretical perspective of positive and negative risk-taking behaviors helps develop a more targeted and practical framework for this study (Zhang and Wang, 2025a). The present study confirms that bystander behavior in cyberbullying is essentially a form of situated risk decision-making. Individuals with prior bullying experiences are more likely to activate moral disengagement as a cognitive mechanism, leading them to choose negative risk-taking behaviors such as assisting the bully or reinforcing the bullying—a finding consistent with the traditional tendency to pathologize risk-taking. However, a key insight from this study is that social relationship closeness can effectively mitigate this negative pathway and potentially redirect behavioral tendencies toward more constructive directions. This suggests that future theoretical models of cyberbullying should move beyond a simplistic “risk–suppression” binary and instead adopt a developmental perspective. Such a perspective would recognize bystander intervention as a form of positive risk-taking behavior (e.g., defending the victim), driven not merely by risk avoidance but by rational considerations grounded in group ethics and collective interests. On a practical level, these findings provide a scientific basis for constructing multi-system intervention strategies. Greater emphasis should be placed on the constructive value of social relationship closeness as a cultural resource. Universities can foster a campus culture that emphasizes trust, reciprocity, and collective responsibility, thereby strengthening students’ sense of group identity. In such an environment, individuals may be more inclined to channel inherent risk-taking impulses into positive interventions, such as protecting victims. Ultimately, interventions should be tailored to the individual and precise. For high-risk students with bullying experiences, efforts should focus on addressing moral disengagement. For the general student population, strengthening social bonds and group norms can encourage positive risk-taking behaviors, such as standing up for victims. This intervention paradigm, based on the concept of positive risk-taking, not only helps reduce the negative impact of cyberbullying but also guides adolescents’ natural risk-taking tendencies toward pathways that support personal growth and healthy social adaptation.

In higher education, effective strategies to strengthen students’ moral norms and ethical oversight deserve further exploration. For example, proactive attitude education and encouraging students to intervene in bullying incidents to help victims can effectively reduce such behaviors. Research has shown that proactive, risk-taking behavioral interventions not only significantly decrease the likelihood of cyberbullying but also increase positive emotions and life satisfaction among youth (Zhang and Wang, 2025b). Advocating for “proactive bystander action to stop bullying after witnessing it” emerges as a key future educational practice. Such interventions reinforce group norms, foster prosocial tendencies in those who intervene, and decrease levels of moral disengagement (Yalçın and Aktaş, 2024). It is important to note that lower moral disengagement encourages proactive interventions, creating a positive cycle. Recent evidence has suggested that intervention-focused educational practices are effective and should be expanded. That approach helps reduce the risks of bullying and negative moral development by regulating individual cognitive processes.

As a situational factor, social relationship closeness requires collective maintenance by group members. School-led group activities, ideological education, and psychological counseling significantly decrease campus cyberbullying incidents. These efforts encourage bystanders to uphold positive environmental norms and reduce individual moral disengagement (Crooks et al., 2017).

Given the cultural specificity of variable measurement, cross-cultural comparative studies are recommended. Although cyberbullying among university students is prevalent worldwide, the behavioral and cognitive factors influencing such behavior in different contexts have yet to be fully elucidated. Expanding sample sources and investigating both social contextual and individual cognitive dimensions can deepen insights into cyberbullying intervention strategies. For instance, comparing how social relationship closeness (in collectivist cultures) differs from concepts like “social proximity” or “social distance” (used in Western contexts) can reveal cultural differences in norms. This comparison informs localized intervention practices by clarifying how these factors influence cyberbullying perceptions and intervention decisions. Specifically, the moderating role of social relationship closeness may be stronger in collectivist cultures, whereas individualist cultures may rely more on personal moral judgment than group norms — a hypothesis that requires large-scale validation.

As a cross-sectional study, this research only assessed the correlations among variables at a single time point and therefore could not determine causality. First, because this study can not determine causality, conducting follow-up research and developing a longitudinal framework are practically important. Specifically, longitudinal modeling methods, such as cross-lagged and random-intercept cross-lagged models, can be used to systematically examine the lagged association effects of bystander intervention behaviors, moral disengagement, and social relationship closeness over time at both the between-person and within-person levels. We are dedicated to systematic longitudinal and experimental research and will subsequently explore the causal mechanisms among cyberbullying involvement experiences, social relationship intimacy, and different bystander behaviors.

Additionally, the sample in this study was mainly composed of college students in Shanghai, where regional and age factors might influence. Future research should include a more diverse sample by gathering data from multiple regions and various age groups to strengthen the research’s conclusions, making them more applicable to a broader population. Moreover, given the cultural specificity of selected variables, cross-cultural validation is essential to confirm the universality of research conclusions, ensuring applicability beyond the current sample context.

From a methodological perspective, this study relies on validated scales to assess cognitive and behavioral dimensions. Specifically, we employed the “Behavioral Intentions Questionnaire” to evaluate distinct bystander intervention behaviors. While scale-based measurement is a widely adopted international approach, it is crucial to recognize its inherent limitations. Our assessments depend on participants’ subjective perceptions and self-reported data. This reliance introduces challenges in excluding the possibility that individuals may withhold truthful responses due to concerns, fear, psychological distress, or ethical considerations. Additionally, the behavioral assessment in this study relies on intentional reports within hypothetical scenarios. However, in real-world cyberbullying incidents, factors such as time pressure and concerns about social consequences may lead to discrepancies between actual behaviors and intentions.