Adolescents' Cyber-Defending for Cyberbullying: A Socio-Emotional,Beliefs,and Past Experience Model

Abstract

Adolescents experiencing cyberbullying attacks (i.e., cyber-victims) can suffer severe psychological harm (e.g., suicide). To combat cyberbullying, people can defend the cyber-victims (cyber-defending). Unlike past studies, we proposed a comprehensive theoretical model of cyber-defending that includes socio-emotional aspects, beliefs, and past bullying experiences (as a bully, victim, and/or witness; face-to-face vs. online). We then empirically tested it among 817 students across seven secondary schools using structural equation modeling (SEM). Results revealed that participants with higher social competence, depression, affective empathy, or stronger pro-victim beliefs reported more cyber-defending. Furthermore, beliefs and past experiences mediated the relationships between socio-emotional factors and cyber-defending. These findings help build a theory of cyber-defending, provide practical implications, and offer future directions for promoting cyber-defending, which will ultimately reduce cyberbullying.

Introduction

Part of the Internet's risks and dark side,¹ Cyberbullying (“long-term, aggressive, intentional, and repetitive acts by one or more individuals using electronic means against an almost powerless victim”)² harms 14% to 58% of youths.³ These cyber-victims were over twice as likely as others to harm themselves (odds ratio [OR] = 2.35), engage in suicidal behaviors (OR = 2.10), or attempt suicide (OR = 2.57; see meta-analysis John et al.⁴). Unlike traditional bullying, cyberbullying has potentially unlimited witnesses (i.e., cyber-bystanders) who can amplify the bullying or defend the victim (cyber-defend).⁵ In traditional bullying, a person who intervened or defended the victim ended the bullying 57% of the time,⁶ so cyber-defending might likewise counter cyberbullying.

Past studies suggested that socio-emotional factors,⁷ beliefs,⁸ past (cyber) bullying experiences,⁹ cognitive interpretation/awareness,¹⁰ or other bystanders' behaviors/norms¹¹ were linked to cyber-defending. Therefore, this study integrated them to develop and test a comprehensive model of cyber-defending among 817 Chinese adolescents via structural equation modeling (SEM).

Antecedents of cyber-defending

Cyber-defending antecedents might include past cyber-victimization, cyberbullying awareness, negative attitudes toward cyberbullying, subjective norms, perceived behavioral control, responsibility, or self efficacy to intervene.⁵ Combining the five-step Bystander Intervention Model,^12,13 Reasoned Action Approach (TRA),¹⁴ and Social Cognitive Theory (SCT),¹⁵ socio-emotional factors can drive beliefs, awareness, and past experiences, which can explain cyber-bystanding. Likewise, we can examine whether socio-emotional factors, beliefs, awareness, or past experiences explain cyber-defending.

Socio-emotional: Social competence, empathy, self-esteem, life satisfaction, depression

People who can effectively communicate with others (social competence)¹⁶ or have more empathy (“experience the feelings or emotions of others”—affective empathy; or “understand others' emotional states”—cognitive empathy)¹⁷ were more likely to understand and appreciate cyber-victims and therefore cyber-defend.^18–23 However, studies testing whether both types of empathy increase cyber-defending have yielded mixed results.^24–26

Individuals with higher perceptions of their self-worth (self-esteem)²⁷ often have more agencies, so they might be more likely to defend others.^10,28 However, such individuals often have greater resources or resilience, so they might be less sensitive to cyberbullying and less likely to cyber-defend.²⁹

People with greater cognitive and affective evaluations of their lives (life satisfaction)³⁰ showed more prosocial behavior intentions,^31,32 less cyberbullying, and less cyber-victimization.^33,34 As no published study to date has determined the link between life satisfaction and cyber-defending, we tested it.

Depression negatively affects one's feelings, behaviors, and thinking,³⁵ yielding fewer prosocial behaviors,³⁶ and greater cyber-victimization (see meta-analysis Tran et al.³⁷). According to affect-cognition theories, however, sadder individuals valued social norms more³⁸ and were more helpful,³⁹ so we proposed that people with depression would be more likely to cyber-defend.

Beliefs, past experiences, and awareness

People's belief system (i.e., attitudes, subjective norms and perceived behavioral norms, theory of planned behavior)⁴⁰ or awareness might affect cyber-defending. Attitudes are individuals' general affective evaluation of something or someone. People who supported cyberbullying were less likely to cyber-defend.⁸ Subjective norms are behaviors supported by an individual or a group.⁴⁰ Perceived behavior control is the perceived difficulty or ease in performing a behavior.⁴⁰ We proposed that those who are more supportive of cyber-defending, perceive greater subjective norms toward cyber-defending, or perceive greater behavioral control would be more likely to cyber-defend.

Past (cyber) bullying/victimization experiences might affect cyber-defending. Cyberbullies were less likely to empathize with cyber-victims, so they were less likely to cyber-defend.⁴¹ By contrast, cyber-victims were more likely to empathize with victims and hence more likely to cyber-defend.^5,23,41,42 Past studies have not examined whether past witnessing of cyberbullying (cyber-witnessing) is related to cyber-defending.

As people must notice an emergency before intervening,¹³ those with greater awareness of cyberbullying were more likely to cyber-defend. Whereas most adolescents mistook cyberbullying for fun,⁴³ those who noticed past (cyber) bullying were more likely to cyber-defend.^44,45 Also, individuals with more pro-bully beliefs and weaker pro-victim attitudes perpetrated more cyberbullying.^46,47 Likewise, we proposed that pro-victim attitudes would predict cyber-defending.

Next, we tested this socio-emotional, belief, past experience, and awareness model of cyber-defending. To reduce omitted variable bias,⁴⁸ we controlled for demographics,⁴⁹ school effects,^46,50 duration of daily online activities,⁵¹ and past academic test scores.⁵²

Methods

Participants and procedures

In this study, 817 students (54% female, 46% male; mean age = 15.1 years; 97% ethnic Chinese) from 7 secondary schools in Hong Kong, China, participated. Participants, their parents (for minors), and school principals gave informed consent. Students completed surveys voluntarily. The authors' university's Human Research Ethics Committee approved this study.

Measures

A higher score on the scales indicated greater frequency or endorsement of the construct (for details, see Table 1).

Table 1.

Details of Measurement

Variable	Measures	Likert scale/frequency	Sample item
Socio-emotional factors
Satisfaction with life	Satisfaction with Life Scale⁵³	5-Item Likert scale, “1” as strongly disagree, “7” as strongly agree	“In most ways my life is close to my ideal”
Depression	Center for Epidemiologic Studies Depression Scale⁵⁴	20-Item Likert scale rating on frequency, “1” to “4”; “1” = rarely or none of the time; “4” = Most or all of the time	“I was bothered by things that usually don't bother me.”
Self-esteem	The Rosenberg Self-Esteem Scale²⁷	10-Item Likert scale, “1” to “4”	“I feel that I have a number of good qualities.”
Social competence	The Social Competence subscale of The Perceived Competence Scale for Children, Harter⁵⁵	7-Item scale; Likert scale, “1” to “4”	“Most of my classmates like me.”
Affective empathy and cognitive empathy	The Basic Empathy Scale⁵⁶	20-Item Likert scale; “1” to “7”	“When someone is feeling ‘down’ I can usually understand how they feel,” and “After being with a friend who is sad about something, I usually feel sad.”
Face-to-face bullying experiences
Face-to-face victimization and bullying	The My Day at School self-report questionnaire of Aggression and Victimization Scales⁵⁷	10-Item scale with 2 subscales that measures frequency; “1” to “4”; “1” = never; “4” = everyday	“How often do other people tease or make fun of you?” (face-to-face victimization), and “How often do you tease or make fun of other people?” (face-to-face bullying).
Online bullying experiences
Past experience in cyber-victimization, cyber-bullying, and witnessing cyberbullying	Three 10-item scale adapted from Leung et al.¹¹	“1” to “5”; “1” = never; “5” = everyday	“Others gossip or say mean things about me in online games/on the internet.” (cyber-victimization), “I gossip or say mean things about others in online games/on the internet” (cyber-bullying), and “I've seen others gossip of say mean things in online games/on the internet” (witnessing cyberbullying).
Beliefs
Pro-victim belief	The Pro-Victim Scale⁵⁸	12-Item Likert scale; “1” to “7”	“I like it when someone stands up for people who are being bullied.”
Pro-bully belief	The attitudes toward bullying⁵⁹	9-Item Likert scale; “1” to “7”	“It is fun to encourage bullying.”
Positive attitudes toward cyberbullying	PACQ⁶⁰	8-Item Likert Scale; “1” to “7”	“It makes me feel good to send mean online messages that make fun of others.”
Awareness of cyberbullying	Attitude and Awareness of Cyberbullying⁶¹	8-Item Likert scale; “1” to “7”	“Cyberbullying is a problem at my school.”
Cyber-defending behavior	A subscale adapted from PRQ⁶²	3-Item scale to measure frequency; “1” to “5”; “1” = never; “5” = everyday	“Tells others to stop the cyberbullying.”

PACQ, Positive Attitudes toward Cyberbullying Questionnaire; PRQ, Participant Role Questionnaire.

Demographics, online activities, examinations

Students identified their gender, age, school year, nationality, place of residence, family/individual income, number of siblings, and numbers of electronic devices; their daily time on Internet activities (instant messaging, checking e-mail, surfing the web, using social network, doing homework, and gaming); and the mean score of their academic subjects in the previous term (academic report).

Socio-emotional factors

We used the Satisfaction with Life Scale,⁵³ Center for Epidemiologic Studies Depression Scale,⁵⁴ Rosenberg Self-Esteem Scale,²⁷ the social competence subscale of the Perceived Competence Scale for Children,⁵⁵ and the affective empathy and cognitive empathy subscales of the Basic Empathy Scale.⁵⁶

Past (cyber) bullying experiences, beliefs, and cyber-defending

My Day at School self-report questionnaire of Aggression and Victimization Scales measured face-to-face bullying and victimization.⁵⁷ Three adapted scales measured past cyberbullying, cybervictimization, and cyberwitnessing.¹¹

Pro-Victim Scale measured pro-victim beliefs.⁵⁸ Attitudes toward Bullying measured pro-bully beliefs.⁵⁹ We measured attitudes and awareness using Positive Attitudes toward Cyberbullying Questionnaire⁶⁰ and Awareness of Cyberbullying, respectively.⁶¹

Participant Role Questionnaire measured cyber-defending behavior.⁶²

Data analysis

We addressed the following analytic issues with specific statistics strategies (Table 2): (a) missing data with Markov Chain Monte Carlo multiple imputation,⁶³ (b) survey measurement errors (e.g., for depression construct) with confirmatory factor analyses (CFA),⁶⁴ (c) indirect mediation effects with a SEM,⁶⁴ (d) SEM interactions with residual centering,⁶⁵ (e) many hypotheses' false positives with the two-stage linear step-up procedure,⁶⁶ (f) compare effect sizes with Lagrange multiplier tests,⁶⁷ (g) consistency of results across data sets (robustness) with analyses of original (not estimated) data (for a detailed description of the whole analysis, see Ahn et al.).^68,69

Table 2.

Statistics Strategies to Address Each Analytic Difficulty

Analytic difficulty	Statistics strategy
Data set
Missing data	Markov Chain Monte Carlo multiple imputation⁶³
Measurement errors on surveys	Factor analysis⁶⁴
Explanatory variables
Indirect, mediation effects (X → M → Y)	Sobel test⁷¹ Structural equation modeling⁶⁴
Interaction in structural equation modeling	Residual centering⁶⁵
Many hypotheses' false positives	Two-stage linear step-up procedure⁶⁶
Consistency of results across data sets (Robustness)	Separate single outcome models Analyses of subsets of the data⁶⁸ Original (not estimated) data

Factor analyses

We tested each construct's survey items for internal validity (e.g., affective empathy) and minimized their measurement errors with CFA. Bartlett factor scores yield unbiased estimates of factor score parameters.⁶⁴ To assess the fit of the CFA or SEM, we used the comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error approximation (RMSEA), and standardized root mean square residual (SRMR), which minimize type I and type II errors in many simulations.⁷⁰ The total effect (TE) of an explanatory variable on the outcome is the sum of its direct effects and all of its indirect effects.

Explanatory model

We modeled cyber-defending with an ordinary least squares regression before applying an SEM.⁶⁸ $C y b e r - D e f e n d_{i} = β_{0} + e_{i}$ (1)

In vector Cyber-Defend, the outcome of student i has grand mean intercept β₀, with residual e_i. Explanatory variables were entered in sequential sets to estimate the variance explained by each set and to test for mediation effects.⁶⁸ Structural variables influenced malleable process variables, so the former preceded the latter. Hence, we entered explanatory variables as follows: school, demographics, daily activities, average score in previous examination, socio-emotional measures, face-to-face bullying situations, attitudes, cyberbullying, and their interactions. $\begin{matrix} C y b e r - D e f e n d_{i} = β_{0} + e_{i} + β_{q} S c h o o l_{i} + β_{r} D e m o g r a p h i c s_{i} + β_{s} {D a i l y}_{-} a c t i v i t i e s_{i} \\ + β_{t} A v g s c o r e_{i} + β_{u} S o c i o - E m o t i o n_{i} + β_{v} F a c e - t o - F a c e_{i} + β_{w} A t t i t u d e - A w a r e n e s s_{i} \\ + β_{x} C y b e r_{i} + β_{z} I n t e r a c t i o n s_{i} \end{matrix}$

School identification numbers (School) were entered first. Using an alpha level of 0.05, a nested hypothesis test (χ² log likelihood) determined the significance of each set of explanatory variables.⁶⁸ As omitting nonsignificant variables did not cause omitted variable bias, we safely removed them to increase precision and reduce multicollinearity.⁶⁸

Next, we applied this procedure to all vectors: Demographics, Daily_activities, Avg_score, Socio-Emotion, Face-to-Face, Attitude-Awareness, Cyber, and their Interactions. See variables in each vector in the above variables section.

Sobel mediation tests across the above vectors⁷¹ yielded a path analysis⁶⁸ as an initial SEM.⁶³ Removing nonsignificant links finalized the SEM. We analyzed residuals for influential outliers.

Results

The factor analyses showed single dominant factors for the outcome variable (Cyber-Defend) and each of the factors in the explanatory model (Table 3), all with high reliability (Table 4). See Table 5 for summary statistics.

Table 3.

Eigenvalues Showing Single Dominant Factors in Each Set of Scales

	Eigenvalues					% of variance explained by …
Factor	First	Second	Third	First/Second	Second/Third	First eigenvalue	Second eigenvalue
Cyber-defending	2.542	0.383	0.075	6.637	5.107	85%	13%
Depression	8.969	1.127	0.901	7.958	1.251	56%	7%
Social competence	5.024	0.758	0.417	6.628	1.818	72%	11%
Affective empathy	4.365	1.488	0.896	2.933	1.661	44%	15%
Face-to-face victimization	3.519	0.52	0.437	6.767	1.190	70%	10%
Pro-victim belief	4.617	1.801	0.817	2.564	2.204	46%	18%
Cyberbullying awareness	3.283	1.096	0.782	2.995	1.402	47%	16%
Cyber-victimization	7.807	0.457	0.424	17.083	1.078	78%	5%
Witnessing cyberbullying	5.750	0.696	0.548	8.261	1.270	72%	9%

Note: Only variables that were included in the final model are listed above.

Table 4.

Confirmatory Factor Analysis Results for Each Factor

				CFA goodness of fit statistics
Factor	No. of items	Reliability	Cronbach's α	SRMR	CFI	IFI	TLI	RMSEA	Chi-square	df	p	AGFI	RFI
Cyber-defending	3	0.991	0.824	0.012	1.000	1.000	1.000	0.000	2.82	3	0.420	0.998	0.996
Depression	16	0.945	0.925	0.067	0.994	0.994	0.991	0.028	150.33	79	0.000	0.991	0.982
Social competence	7	0.976	0.902	0.061	0.993	0.993	0.988	0.051	47.22	12	0.000	0.992	0.984
Affective empathy	10	0.879	0.831	0.080	0.985	0.985	0.973	0.040	71.76	26	0.000	0.989	0.959
Face-to-face victimization	5	0.937	0.819	0.036	0.987	0.987	0.974	0.062	26.74	5	0.000	0.989	0.969
Pro-victim belief	10	0.910	0.829	0.055	0.979	0.979	0.969	0.047	103.29	30	0.000	0.987	0.956
Cyberbullying awareness	7	0.832	0.768	0.058	0.963	0.963	0.934	0.077	90.30	12	0.000	0.979	0.925
Cyber-victimization	10	0.983	0.937	0.073	0.998	0.998	0.997	0.026	54.67	31	0.005	0.996	0.993
Witnessing cyberbullying	8	0.965	0.920	0.072	0.989	0.989	0.977	0.075	100.94	14	0.000	0.984	0.974

Only variables included in the final model are listed above. For some scales, some items were trimmed to yield a better fit of the measurement model.

AGFI, adjusted goodness of fit index; CFA, confirmatory factor analyses; CFI, comparative fit index; IFI, incremental fit index; RFI, relative fit index; RMSEA, root mean square error approximation; SRMR, standardized root mean square residual; TLI, Tucker–Lewis index.

Table 5.

Summary Statistics

	Variable	Mean	SD	1	2	3	4	5	6	7	8	9
1	Cyber-defending	0.502	0.306	0.252	0.024	0.013	0.053	0.030	0.043	0.055	0.057	0.088
2	Depression	0.653	1.791	0.075^*	0.426	−0.167	0.092	0.126	−0.014	0.079	0.095	0.144
3	Social competence	0.601	2.655	0.044	−0.426^***	0.361	0.050	−0.075	0.037	0.012	−0.056	−0.023
4	Affective empathy	0.993	4.687	0.107^**	0.143^***	0.084	0.985	−0.019	0.326	0.260	−0.036	0.083
5	Face-to-face victimization	0.516	1.466	0.116^***	0.373^***	−0.241^***	−0.037	0.266	−0.070	0.031	0.099	0.096
6	Pro-victim belief	1.040	5.361	0.083^*	−0.021	0.060	0.316^***	−0.130^***	1.081	0.327	−0.132	−0.008
7	Cyberbullying awareness	1.000	5.227	0.110^**	0.121^***	0.019	0.262^***	0.061	0.315^***	0.998	−0.024	0.131
8	Cyber-victimization	0.549	1.312	0.207^***	0.265^***	−0.171^***	−0.067	0.349^***	−0.232^***	−0.045	0.301	0.213
9	Witnessing cyberbullying	0.855	2.037	0.206^***	0.258^***	−0.045	0.097^**	0.217^***	−0.009	0.153^***	0.455^***	0.730

Note: Correlations, variances, and co-variances are along the lower left triangle, diagonal, and upper right triangle of the matrix (N = 817). Only variables included in the final model are listed above.

SD, standard deviation.

p < 0.05; ^**p < 0.01; ^***p < 0.001.

Bold values represent variances.

Explanatory model

Students' socio-emotional variables, real life situation, attitude toward cyberbullying, and cyberbullying variables were linked to cyber-defending (Fig. 1). Students whose social competence exceeded the mean by 10% averaged 1.2% more cyber-defending (1.2% = 0.119 × 10%; from β × 10%). Meanwhile, students who were 10% more depressed than the mean averaged 0.6% more cyber-defending, an indirect effect mediated by 4.8% more face-to-face victimization. Furthermore, students with 10% higher affective empathy than the mean averaged 0.5% more cyber-defending, mediated by 3.8% higher pro-victim belief and 1.7% higher cyberbullying awareness.

FIG. 1.

A path diagram showing the SEM model predicting Defend students (SRMR = 0.071; CFI = 0.954; IFI = 0.954; TLI = 0.952; RMSEA = 0.045; χ² [7,146] = 16,268; AGFI = 0.713; RFI = 0.918). Solid lines indicate positive links. Dashed lines indicate negative links. Thicker lines indicate larger links. ***p < 0.001. AGFI, adjusted goodness of fit index; CFI, comparative fit index; IFI, incremental fit index; RFI, relative fit index; RMSEA, root mean square error approximation; SEM, structural equation modeling; SRMR, standardized root mean square residual; TLI, Tucker–Lewis index.

Students who experienced 10% more face-to-face victimization than the mean averaged 1.3% more cyber-defending, completely mediated by 4.3% more cyber-victimization, and 2.9% more cyber-witnessing. Moreover, students who had 10% higher pro-victim belief than the mean averaged 1.2% more cyber-defending via a direct effect and partially mediated by 1.9% lower cyberbullying awareness and 1.9% less cyber-victimization. Students with 10% higher cyberbullying awareness than the mean averaged 0.1% more cyber-defending, mediated by 1.2% more cyber-witnessing.

Students who experienced 10% more cyber-victimization than the mean averaged 2.3% more cyber-defending. Furthermore, students who witnessed 10% more cyberbullying than the mean averaged 1.1% more cyber-defending. Other variables and interactions were not significant.

Discussion

We proposed and showed via SEM that socio-emotional factors (social competence, depression, and affective empathy), beliefs (pro-victim belief and cyberbullying awareness), and past experience (face-to-face and cyber-victimization, and witnessing cyberbullying) factors explained cyber-defending. Moreover, belief and past experience mediated socio-emotional factors' links with cyber-defending.

Participants with greater social competence, depression, or affective empathy reported more cyber-defending. Aligning with past studies showing that people with greater social competence perceived greater control and agency,¹⁷ this study showed that such people cyber-defended more.

Next, people with greater depression reported more past face-to-face bullying victimization, more past cyber-victimization, more past cyber-witnessing, and hence more cyber-defending. These results aligned with past studies suggesting that those with more depression were more likely to suffer face-to-face bullying or cybervictimization. Thus, they might identify with victims more and recognize cyberbullying and cyber-witness more, so they might have both the empathy and awareness to cyber-defend them.⁷²

In addition, affective empathy indirectly explained cyber-defending through (a) greater pro-victim beliefs and (b) greater awareness of cyberbullying, which in turn positively explained witnessing cyber-bullying. These results were consistent with past studies showing that people with more affective empathy appreciated cyber-victims' suffering²⁵ and had greater awareness to recognize cyberbullying, both of which helped them cyber-defend.⁴⁰ Hence, affective empathy, not cognitive empathy, explained cyber-defending.^25,73

These results imply interventions that enhance adolescents' social competence or affective empathy might also increase cyber-defending⁷⁴ (e.g., a cyber-version of bullying bystander interventions) in text, audio, or virtual reality Internet environments.⁷⁵ Furthermore, future studies can determine the mechanism(s) by which negative moods (e.g., depression) facilitate cyber-defending.

Participants with more pro-victim beliefs or more cyberbullying awareness were more likely to cyber-defend, aligning with the view that such participants identified more with the victims and recognized cyberbullying more readily,⁸ which facilitated their cyber-defense.²³ Future studies can test the stability of pro-victim belief's negative links with cyberbullying awareness and cyber-victimization (and their potential mechanisms), which slightly reduced its substantial link with cyber-defending. More cyberbullying awareness explained more cyber-defending, aligning with the importance of awareness in the five-step intervention model.¹³ Future studies can test whether interventions that enhance pro-victim beliefs or cyberbullying awareness increase cyber-defending.

This study's limitations include a limited sample, few measures, and cross-sectional data. Future studies can include participants across more ages and more regions, other measures (e.g., negative attitudes toward cyberbullying), and longitudinal or experimental data to test causality.

Conclusions

We showed that socio-emotional factors (social competence, affective empathy and depression), beliefs (pro-victim belief and cyberbullying awareness), and past experiences (face-to-face and cyber-victimization, and witnessing cyberbullying) explained cyber-defending among these adolescents. Furthermore, these beliefs and past experiences mediated the links between socio-emotional factors and cyber-defending. Results inform future interventions (e.g., low-cost alternative interventions)⁷⁶ to promote cyber-defending.

Footnotes

Authors' Contributions

A.N.M.L. contributed to the conceptualization, design, data collection, and writing. M.M.C. contributed to the analysis and writing of the article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The work in this article was substantially supported by grants from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No.: EdUHK 18602917). The funding source was not involved in study design, collection, analysis, interpretation of data, writing of the article, or the decision to submit the article for publication.

References

Radanliev

, De Roure

. Epistemological and bibliometric analysis of ethics and shared responsibility—Health policy and IoT systems. Sustainability, 2021; 13(15):8355.

Dehue

Cyberbullying research: New perspectives and alternative methodologies. Introduction to the special issue. J Community Appl Soc Psychol, 2013; 23(1):1–6.

Zhu

, Huang

, Evans

, et al. Cyberbullying among adolescents and children: A comprehensive review of the global situation, risk factors, and preventive measures. Front Public Health, 2021; 9:634909.

John

, Glendenning

, Marchant

, et al. Self-harm, suicidal behaviours, and cyberbullying in children and young people: Systematic review. J Med Internet Res, 2018; 20(4):e9044.

Leung

ANM.

To help or not to help: Intervening in cyberbullying among Chinese cyber-bystanders. Front Psychol, 2021; 12:483250.

Hawkins

, Pepler

, Craig

. Naturalistic observations of peer interventions in bullying. Soc Dev, 2001; 10(4):512–527.

Allison

, Bussey

. Individual and collective moral influences on intervention in cyberbullying. Comput Hum Behav, 2017; 74:7–15.

Pabian

, Vandebosch

, Poels

, et al. Exposure to cyberbullying as a bystander: An investigation of desensitization effects among early adolescents. Comput Hum Behav, 2016; 62:480–487.

Allison

, Bussey

. Cyber-bystanding in context: A review of the literature on witnesses' responses to cyberbullying. Child Youth Serv Rev, 2016; 65:183–194.

10.

Ireland

, Hawdon

, Huang

, et al. Preconditions for guardianship interventions in cyberbullying: Incident interpretation, collective and automated efficacy, and relative popularity of bullies. Comput Hum Behav, 2020; 113:106506.

11.

Leung

, Wong

, Farver J

. You are what you read: The belief systems of cyber-bystanders on social networking sites. Front Psychol, 2018; 9:365

12.

DeSmet

, Bastiaensens

, Van Cleemput

, et al. Deciding whether to look after them, to like it, or leave it: A multidimensional analysis of predictors of positive and negative bystander behavior in cyberbullying among adolescents. Comput Hum Behav, 2016; 57:398–415; doi:10.1016/j.chb.2015.12.051

13.

Latané

, Darley

. The Unresponsive Bystander: Why Doesn't He Help?. Prentice Hall: Englewood Cliffs, NJ; 1970.

14.

Fishbein

, Ajzen

Predicting and Changing Behavior: The Reasoned Action Approach. Psychology Press: New York, NY; 2010.

15.

Bandura

Self-Efficiency, the Self-Efficacy Feeling. De Boeck University: Bruxelles; 2007.

16.

Rubin

, Rose-Krasnor

Interpersonal Problem Solving. In: Handbook of Social Development. ( Van Hassett

, Hersen

. eds.) Plenum: New York; 1992; pp. 283–323.

17.

Jolliffe

, Farrington

. Examining the relationship between low empathy and bullying. Aggress Behav, 2006; 32(6):540–550.

18.

Song

, Oh

. Factors influencing bystanders' behavioral reactions in cyberbullying situations. Comput Hum Behav, 2018; 78:273–282.

19.

Barlińska

, Szuster

, Winiewski

. Cyberbullying among adolescent bystanders: Role of the communication medium, form of violence, and empathy. J Community Appl Soc Psychol, 2013; 23(1):37–51.

20.

Barlińska

, Szuster

, Winiewski

. The role of short-and long-term cognitive empathy activation in preventing cyberbystander reinforcing cyberbullying behavior. Cyberpsychol Behav Soc Netw, 2015; 18(4):241–244.

21.

Kokkinos

, Kipritsi

. Bullying, moral disengagement and empathy: Exploring the links among early adolescents. Educ Psychol, 2017; 38(4):535–552.

22.

Lambe

, Della Cioppa

, Hong

, et al. Standing up to bullying: A social ecological review of peer defending in offline and online contexts. Aggress Violent Behav, 2019; 45:51–74.

23.

Van Cleemput

, Vandebosch

, Pabian

. Personal characteristics and contextual factors that determine “helping,”“joining in,” and “doing nothing” when witnessing cyberbullying. Aggress Behav, 2014; 40(5):383–396.

24.

Schultze-Krumbholz

, Zagorscak

, Hess

, et al. The influence of school climate and empathy on cyberbystanders' intention to assist or defend in cyberbullying. Int J Bullying Prev, 2020; 2(1):16–28.

25.

Macháčková

, Pfetsch

. Bystanders' responses to offline bullying and cyberbullying: The role of empathy and normative beliefs about aggression. Scand J Psychol, 2016; 57(2):169–176.

26.

Barlińska

, Szuster

, Winiewski

. Cyberbullying among adolescent bystanders: Role of affective versus cognitive empathy in increasing prosocial cyberbystander behavior. Front Psychol, 2018; 9:799.

27.

Rosenberg

Rosenberg Self-Esteem Scale. Princeton University Press: Princeton, NJ; 1965.

28.

Bussey

, Luo

, Fitzpatrick

, et al. Defending victims of cyberbullying: The role of self-efficacy and moral disengagement. J School Psychol, 2020; 78:1–12.

29.

Macháčková

, Dedkova

, Sevcikova

, et al. Bystanders' supportive and passive responses to cyberaggression. J School Viol, 2018; 17(1):99–110.

30.

Diener

, Lucas

, Oishi

. Subjective well-being: The science of happiness and life satisfaction. Handb Posit Psychol, 2002; 2:63–73.

31.

Kushlev

, Radosic

, Diener

. Subjective well-being and prosociality around the globe: Happy people give more of their time and money to others. Soc Psychol Pers Sci, 2021; 13(4):849–861.

32.

Zhang

, Zhao

. Influence of urban residents' life satisfaction on prosocial behavioral intentions in the community: A multiple mediation model. J Community Psychol, 2021; 49(2):406–418.

33.

Arnarsson

, Nygren

, Nyholm

, et al. Cyberbullying and traditional bullying among Nordic adolescents and their impact on life satisfaction. Scand J Public Health, 2020; 48(5):502–510.

34.

Leung

ANM

, Wong

, Farver

. Cyberbullying in Hong Kong Chinese students: Life satisfaction, and the moderating role of friendship qualities on cyberbullying victimization and perpetration. Pers Individ Differ, 2017; 133:7–12.

35.

American Psychological Association. Depression. American Psychological Association. Available from: https://www.apa.org/topics/depression [Last accessed: July 24, 2022].

36.

Morabito

, Burani

, Hajcak

. Depressive symptoms prospectively predict peer victimization: A longitudinal study among adolescent females. Child Psychiatry Hum Dev, 2022; 53(1):39–47.

37.

Tran

HGN

, Thai

, Dang

NTT

, et al. Cyber-victimization and its effect on depression in adolescents: A systematic review and meta-analysis. Trauma Violence Abuse, 2023; 24(2):1124–1139.

38.

Forgas

JP.

Feeling and speaking: Mood effects on verbal communication strategies. Pers Soc Psychol Bull, 1999; 25(7):850–863.

39.

Forgas

JP.

Don't worry, be sad! On the cognitive, motivational, and interpersonal benefits of negative mood. Curr Dir Psychol Sci, 2013; 22(3):225–232.

40.

Ajzen

The theory of planned behavior. Organ Behav Hum Decis Process, 1991; 50(2):179–211.

41.

Kaiser

, Kyrrestad

, Fossum

. Help-seeking behavior in Norwegian adolescents: The role of bullying and cyberbullying victimization in a cross-sectional study. Scand J Child Adolesc Psychiatry Psychol, 2020; 8(1):81–90.

42.

Desmet

, Veldeman

, Poels

, et al. Determinants of self-reported bystander behavior in cyberbullying incidents amongst adolescents. Cyberpsychol Behav Soc Netw, 2014; 17:27–215.

43.

Runions

, Shapka

, Dooley

, et al. Cyber-aggression and victimization and social information processing: Integrating the medium and the message. Psychol Violence, 2013; 3(1):9–26.

44.

Dillon

, Bushman

. Unresponsive or un-noticed?: Cyberbystander intervention in an experimental cyberbullying context. Comput Human Behav, 2015; 45:144–150.

45.

Greitemeyer

, Fischer

, Kastenmüller

, et al. Civil courage and helping behavior: Differences and similarities. Eur Psychol, 2006; 11(2):90–98.

46.

Christian Elledge

, Williford

, Boulton

, et al. Individual and contextual predictors of cyberbullying: The influence of children's provictim attitudes and teachers' ability to intervene. J Youth Adolesc, 2013; 42(5):698–710.

47.

Wright

, Harper

. Cyberbullying: An Online Menace. In: Advances in Psychology Research. (Columbus AM. ed.) Nova Science Publishers, Inc.: New York; 2017; pp. 147–176.

48.

Cinelli

, Hazlett

. Making sense of sensitivity: Extending omitted variable bias. J R Stat Soc Series B Stat Methodol, 2020; 82(1):39–67.

49.

Topçu Ç

Erdur-Baker Ö

, Çapa-Aydin

. Examination of cyberbullying experiences among Turkish students from different school types. Cyberpsychol Behav, 2008; 11(6):643–648.

50.

Williford

, DePaolis

, Colonnieves

. Differences in school staff attitudes, perceptions, self-efficacy beliefs, and intervention likelihood by form of student victimization. J Soc Soc Work Res, 2021; 12(1):83–107.

51.

Aizenkot

Social networking and online self-disclosure as predictors of cyberbullying victimization among children and youth. Child Youth Serv Rev, 2020; 119:105695.

52.

Cook

, Williams

, Guerra

, et al. Predictors of bullying and victimization in childhood and adolescence: A meta-analytic investigation. School Psychol Quart, 2010; 25(2):65–83.

53.

Diener

, Larsen

, Levine

, et al. Intensity and frequency: Dimensions underlying positive and negative affect. J Pers Soc Psychol, 1985; 48(5):1253.

54.

Radloff

LS.

The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas, 1977; 1(3):385–401.

55.

Harter

The perceived competence scale for children. Child Dev, 1982; 51(1):87–97.

56.

Jolliffe

, Farrington

. Development and validation of the Basic Empathy Scale. J Adolesc, 2006; 29(4):589–611.

57.

Schwartz

, Chang

, Farver

. Correlates of victimization in Chinese children's peer groups. Dev Psychol, 2001; 37(4):520.

58.

Rigby

, Slee

. Bullying among Australian school children: Reported behavior and attitudes toward victims. J Soc Psychol, 1991; 131(5):615–627.

59.

Stevens

, Van Oost

, De Bourdeaudhuij

. The effects of an anti-bullying intervention programme on peers' attitudes and behaviour. J Adolesc, 2000; 23(1):21–34.

60.

Barlett

, Gentile

. Attacking others online: The formation of cyberbullying in late adolescence. Psychol Pop Media Cult, 2012; 1(2):123.

61.

Brewer

EA.

Fighting Fire with Fire: The Use of a Multimedia Webquest in Increasing Middle-School Students' Understandings of Cyberbullying (Doctoral Dissertation, Loyola Marymount University); 2011.

62.

Salmivalli

, Voeten

. Connections between attitudes, group norms, and behaviour in bullying situations. Int J Behav Dev, 2004; 28(3):246–258.

63.

Peugh

, Enders

. Missing data in educational research. Rev Educ Res, 2004; 74:525–556.

64.

Joreskog

, Sorbom

LISREL 10.1. Scientific Software International: New York; 2018.

65.

Little

, Card

, Bovaird

, et al. Structural Equation Modeling of Mediation and Moderation with Contextual Factors. In: Modeling Contextual Effects in Longitudinal Studies. ( Little

, Bovaird

, Card

. eds.) Routledge: Abingdon, England; 2012; pp. 207–230.

66.

Benjamini

, Krieger

, Yekutieli

. Adaptive linear step-up procedures that control the false discovery rate. Biometrika, 2006; 93:491–507.

67.

Bertsekas

DP.

Constrained Optimization and Lagrange Multiplier Methods. Academic: New York; 2014.

68.

Kennedy

Guide to Econometrics. Wiley-Blackwell: New York; 2008.

69.

Ahn

, Chiu

, Patrick

. Connecting teacher and student motivation: Student perceived teacher need-supportive practices and student need satisfaction. Contemp Educ Psychol, 2021; 64:101950.

70.

, Bentler

. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Eq Model, 1999; 6(1):1–55.

71.

Sobel

ME.

Asymptotic Intervals for Indirect Effects in Structural Equations Models. In: Sociological Methodology. ( Leinhart

. eds.) Jossey: San Francisco; 1982; pp. 290–312.

72.

Steger

, Kashdan

. Depression and everyday social activity, belonging, and well-being. J Counsel Psychol, 2009; 56(2):289–300.

73.

Zych

, Baldry

, Farrington

, et al. Are children involved in cyberbullying low on empathy? A systematic review and meta-analysis of research on empathy versus different cyberbullying roles. Aggress Violent Behav, 2019; 45:83–97.

74.

Salmivalli

Participant roles in bullying: How can peer bystanders be utilized in interventions?. Theory Pract, 2014; 53(4): 286–292.

75.

Ingram

, Espelage

, Merrin

, et al. Evaluation of a virtual reality enhanced bullying prevention curriculum pilot trial. J Adolesc, 2019; 71:72–83.

76.

Radanliev

, De Roure

. Alternative mental health therapies in prolonged lockdowns: Narratives from Covid-19. Health Technol, 2021; 11(5):1101–1107.