Abstract
Cyberbullying is an increasingly prevalent public mental health concern during adolescence and is associated with substantial emotional and interpersonal difficulties. In this cross-sectional, school-based study, we examined the interrelated associations of emotion regulation difficulties, maladaptive metacognitive beliefs, loneliness, perceived social support, and duration of social media use with cybervictimization and cyberbullying perpetration in a school-based sample of 600 adolescents aged 14 to 18 years. Participants completed the Revised Cyber Bullying Inventory–II, the Metacognitions Questionnaire for Children, the Difficulties in Emotion Regulation Scale–Short Form (DERS-SF), the UCLA Loneliness Scale–Short Form, and the Multidimensional Scale of Perceived Social Support (MSPSS). Cybervictimization was positively associated with DERS-SF (r = 0.36, P < .001), MCQ-C (r = 0.25, P < .001), and the UCLA Loneliness Scale–Short Form (r = 0.22, P < .001), and negatively associated with MSPSS-total scores (r = −0.19, P < .001). In hierarchical regression analyses, gender (β = −0.190, P < .001), DERS-SF total scores (β = 0.284, P < .001), MCQ-C total scores (β = 0.104, P = .02), and the MSPSS family subscale (β = −0.119, P = .004) significantly predicted cybervictimization (R2 = 0.186). Structural equation modeling showed indirect associations between emotion regulation difficulties, perceived social support, and cybervictimization through loneliness. Cyberbullying perpetration was positively associated with DERS-SF total scores (r = 0.22, P < .001), MCQ-C total scores (r = 0.11, P < .05), and negatively associated with the MSPSS family subscale (r = −0.17, P < .001). In regression analyses, gender (β = −0.212, P < .001), DERS-SF total scores (β = 0.267, P < .001), and the MSPSS family subscale (β = −0.111, P = .01) significantly predicted cyberbullying perpetration (R2 = 0.108). Structural equation modeling analyses further suggested that loneliness may play a role in the association between emotion regulation difficulties, perceived family support, and cyberbullying perpetration. Overall, the findings suggest that adolescents’ involvement in cyberbullying-related behaviors may be associated with a broader psychosocial framework involving emotional dysregulation, maladaptive metacognitive beliefs, loneliness, and perceived social support. These findings may contribute to the development of multidimensional prevention and intervention approaches targeting emotional, cognitive, and interpersonal processes in adolescents.
Keywords: adolescents, cyberbullying, emotion regulation, loneliness, metacognitive beliefs, perceived social support
1. Introduction
Cyberbullying is a universal public health concern, defined as deliberate and repeated aggressive behavior conducted through digital technologies against individuals who have difficulty defending themselves, with cybervictimization referring to exposure to such behaviors and cyberbullying perpetration referring to engagement in them.[1] Recent international data indicate that approximately 15% of adolescents report experiences of cybervictimization, while nearly 1 in 8 adolescents report engaging in cyberbullying behaviors, highlighting the growing global burden of cyberbullying among youth populations.[2] Unlike traditional bullying, cyberbullying is facilitated by digital technologies that allow perpetrators to conceal their identities, transcend temporal and spatial boundaries, and potentially reach wide audiences through persistent digital content.[3] Cyberbullying involvement is strongly linked to psychological, social, and physical difficulties in adolescents.[4] Given these wide-ranging adverse outcomes, identifying the factors associated with cyberbullying and implementing targeted preventive strategies are of critical importance.
Metacognition refers to beliefs and processes involved in monitoring and regulating thoughts and emotions. Maladaptive metacognitive patterns, particularly worry and rumination, are linked to ineffective coping and increased vulnerability to psychiatric disorders.[5] Maladaptive metacognitive beliefs may also interfere with adaptive emotion regulation processes, thereby increasing vulnerability to maladaptive interpersonal behaviors, including cyberbullying involvement.[5] In a large sample of adolescents, involvement in bullying and cyberbullying – both as victims and bullies – was negatively associated with the use of cognitive and metacognitive learning strategies.[6]
Difficulties in emotion regulation have also been highlighted as an important mechanism underlying adolescents’ involvement in cyberbullying.[7] Difficulties in emotion regulation involve difficulties in understanding, managing, and responding adaptively to emotional experiences, particularly under distress.[8] Previous studies indicate that adolescents with higher levels of difficulties in emotion regulation are more prone to interpreting others’ behaviors as hostile, relying on maladaptive coping strategies, and consequently displaying greater tendencies toward aggressive or victimized roles in bullying contexts.[7,9–11]
Perceived social support, defined as an individual’s appraisal of available external resources, has been recognized as a protective factor in adolescence.[12] Grounded in social control theory, it is suggested that strong social bonds and emotional connections may buffer against deviant behaviors and help adolescents cope more effectively with adverse experiences such as cyberbullying involvement.[13] Prior studies highlight the protective role of both family and peer support in reducing adolescents’ involvement in cyberbullying.[12] Conversely, low levels of perceived social support and self-esteem have been associated with a greater risk of cyberbullying involvement, while experiences of cyberbullying involvement may in turn diminish social support and self-esteem.[14]
Loneliness, defined as the perceived gap between desired and actual social relationships,[15] has also been linked to cyberbullying involvement. Prior research suggests that loneliness may increase adolescents’ engagement in online interactions and social media use, potentially increasing vulnerability to cyberbullying involvement.[16,17] Feelings of social isolation may therefore act as a driving factor that increases susceptibility to negative online experiences.[17]
Building on these lines of research, the present study integrates the duration of social media use, emotion regulation, metacognition, loneliness, and perceived social support within a single analytic framework to examine their associations with cyberbullying involvement during adolescence using structural equation modeling (SEM), thereby contributing to the literature by offering a comprehensive, multivariate perspective on these interrelated psychosocial factors. We hypothesized that greater emotion regulation difficulties, maladaptive metacognitive beliefs, loneliness, and a longer duration of social media use, as well as lower perceived social support, would be associated with higher levels of cybervictimization and cyberbullying perpetration.
2. Materials and methods
This study was designed with a cross-sectional approach. Prior to data collection, ethical approval was obtained from the Ethics Committee of Necmettin Erbakan University (approval date, February 19, 2024; approval number, 2024/4809/17135). Permission to conduct the study in high schools was granted by the relevant provincial directorate of national education (decision number E-836688308-605.99-100596291). The study was carried out in accordance with the principles of the Declaration of Helsinki. The sample consisted of 600 adolescents aged 14 to 18 years, recruited using a convenience sampling approach between April 29 and May 20, 2024, from 3 different high schools. Written informed consent was obtained directly from participants aged 18 years. For participants under 18 years of age, written informed consent was obtained from their parents or legal guardians. In addition, written assent was obtained from the adolescents themselves prior to participation. All data were collected anonymously. The inclusion criteria were as follows: adequate cognitive capacity to understand the study instructions and questionnaires, possession of a personal mobile phone and at least 1 social media account, and absence of major physical or neurological disorders. Adolescents who did not meet these criteria were excluded.
2.1. Data collection tools
2.1.1. Sociodemographic data form
The sociodemographic data form, developed by the research team, was used to collect information on participants’ sociodemographic characteristics and patterns of smartphone use and social media engagement.
2.1.2. Revised Cyber Bullying Inventory–II
The Revised Cyber Bullying Inventory–II (RCBI-II) is a 10-item self-report measure assessing both cyberbullying perpetration and cybervictimization among adolescents.[18] The Turkish version has demonstrated good reliability (α = 0.80 for victimization, α = 0.79 for bullying).
2.1.3. Metacognitions Questionnaire for Children
The Metacognitions Questionnaire for Children (MCQ-C) is a 24-item self-report instrument assessing metacognitive beliefs in children.[19] It comprises 4 subscales: positive and negative meta-worry, superstitious/punishment/responsibility beliefs, and cognitive monitoring, each containing 6 items. The Turkish adaptation demonstrated acceptable reliability and validity, with Cronbach’s α = 0.73 and test–retest correlations between 0.76 and 0.82.[20]
2.1.4. Difficulties in Emotion Regulation Scale–Short Form
The Difficulties in Emotion Regulation Scale–Short Form (DERS-SF)[21] is a 16-item self-report measure developed as a brief version of the original DERS.[7] It assesses 5 domains of emotion regulation difficulties: clarity, goals, impulse, strategies, and nonacceptance. The original validation study reported excellent internal consistency (α = 0.92) and good test–retest reliability (r = 0.85). The Turkish adaptation demonstrated satisfactory psychometric properties.[22]
2.1.5. UCLA Loneliness Scale–Short Form
The UCLA Loneliness Scale–Short Form was originally developed as an 8-item measure[23] to assess subjective feelings of loneliness. The Turkish adaptation was revised to a 7-item form.[24]
2.1.6. Multidimensional Scale of Perceived Social Support
The Multidimensional Scale of Perceived Social Support (MSPSS) is a 12-item self-report instrument developed to assess perceived social support.[25] It consists of 3 subscales: family, friends, and significant other. The Turkish adaptation demonstrated good psychometric properties.[26]
As all measures were based on self-report questionnaires, the possibility of reporting-related biases, including social desirability bias, should be considered.
2.2. Statistical analysis
Data analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk) and AMOS (IBM Corp., Armonk). Because missing responses were minimal, missing item values were handled using item mean imputation prior to statistical analyses. Descriptive statistics were calculated as frequencies and percentages for categorical variables and as means and standard deviations for continuous variables. The assumption of normality was evaluated using the Kolmogorov–Smirnov and Shapiro–Wilk tests, as well as by examining skewness–kurtosis values and distributional properties. Associations between cyberbullying involvement and other parameters were examined using Pearson’s correlation and independent-samples t tests. To further investigate predictors of cyberbullying involvement, hierarchical regression analyses were performed, and model fit was assessed through residual and fit statistics. Additionally, SEM was used because it enables the simultaneous examination of direct and indirect associations among multiple interrelated psychosocial variables within a single comprehensive model. This approach was considered particularly appropriate for evaluating the potential interrelationships among emotion regulation difficulties, metacognitive beliefs, loneliness, perceived social support, and cyberbullying involvement. Model fit was evaluated using the relative chi-square (χ2/df), root mean square error of approximation (RMSEA), comparative fit index (CFI), Tucker–Lewis index (TLI), and Normed Fit Index (NFI). RMSEA values below 0.08 were considered indicative of acceptable model fit, whereas CFI and TLI values ≥ 0.95 were regarded as reflecting good fit. Because fit indices may have limited discriminating value in near-saturated models, they were interpreted together with model parsimony and substantive plausibility rather than in isolation. Path coefficients in SEM were reported as standardized regression weights (β), reflecting the strength of associations between predictor and outcome variables. Effect size estimates were interpreted according to conventional guidelines for standardized regression coefficients and explained variance measures proposed by Cohen.[27] Statistical significance was set at P < .05. An a priori power analysis was conducted (α = 0.05, power = 0.80). The required minimum sample size was 347 participants. The final sample of 600 adolescents therefore provided adequate statistical power for correlation, hierarchical regression, and SEM analyses.
3. Results
A total of 287 male (47.8%) and 313 female (52.2%) students participated in the study. The mean age of the participants was 15.58 ± 0.88 years. No significant gender difference was observed in cybervictimization total scores (males: 17.01 ± 5.48; females: 16.17 ± 5.08; t = 1.954, P = .051). However, cyberbullying perpetration scores were significantly higher among males (16.99 ± 5.57) than among females (15.42 ± 4.93; t = 3.657, P < .001). Sociodemographic characteristics and information on digital device use are presented in Table 1.
Table 1.
Descriptive statistics of sociodemographic variables, social media use, short-video viewing, and study measures.
| Mean ± SD | Count n (%) | |
|---|---|---|
| Sociodemographic variables | ||
| Gender | ||
| Female | 313 (52.2%) | |
| Male | 287 (47.8%) | |
| Age | 15.58 ± 0.88 | |
| Digital media use | ||
| Duration of daily social media use (min) | 152.65 ± 87.02 | |
| Duration of daily short-video viewing (min) | 86.14 ± 64.01 | |
| Cyberbullying involvement | ||
| RCBI-II-CV | 16.57 ± 5.29 | |
| RCBI-II-CB | 16.18 ± 5.30 | |
| Metacognition | ||
| MCQ-C-total | 62.40 ± 11.28 | |
| MCQ-C-PMW | 12.70 ± 4.28 | |
| MCQ-C-NMW | 16.26 ± 4.59 | |
| MCQ-C-SPR | 16.07 ± 4.16 | |
| MCQ-C-CM | 17.35 ± 3.78 | |
| Emotion regulation | ||
| DERS-SF-total | 42.17 ± 15.48 | |
| DERS-SF-clarity | 5.48 ± 2.34 | |
| DERS-SF-goals | 9.85 ± 3.49 | |
| DERS-SF-impulse | 7.60 ± 3.70 | |
| DERS-SF-strategies | 12.37 ± 5.53 | |
| DERS-SF-nonacceptance | 6.86 ± 3.48 | |
| Loneliness and perceived social support | ||
| ULS-8-total | 13.66 ± 4.96 | |
| MSPSS-total | 55.98 ± 16.04 | |
| MSPSS-family | 19.80 ± 6.65 | |
| MSPSS-friends | 20.96 ± 6.46 | |
| MSPSS-significant others | 15.21 ± 8.06 | |
CB = cyberbullying perpetration, CM = cognitive monitoring, CV = cybervictimization, DERS-SF = Difficulties in Emotion Regulation Scale–Short Form, MCQ-C = Metacognitions Questionnaire for Children, MSPSS = Multidimensional Scale of Perceived Social Support, NMW = negative meta-worry, PMW = positive meta-worry, RCBI-II = Revised Cyber Bullying Inventory–II, SD = standard deviation, SPR = superstition, punishment, and responsibility beliefs, ULS-8 = UCLA Loneliness Scale–Short Form.
RCBI-II cybervictimization scores showed significant positive correlations with daily social media and short-video viewing time, loneliness, metacognitive beliefs, and difficulties in emotion regulation (all P < .01). In contrast, cybervictimization was negatively correlated with perceived social support, particularly family support (all P < .01). Detailed correlation analyses are presented in Table 2. Similarly, cyberbullying perpetration scores were positively correlated with daily social media and short-video viewing time, metacognitive beliefs, and difficulties in emotion regulation, whereas they were negatively correlated with perceived family support (all P < .05). Detailed correlation analyses are presented in Table 2.
Table 2.
Correlation analysis results between RCBI-II cybervictimization/cyberbullying total scores and social media use, short-video viewing, and DERS-SF, ULS-8, MCQ-C, and MSPSS scale scores.
| RCBI-II-CV | RCBI-II-CB | |
|---|---|---|
| Digital media use | ||
| Duration of daily social media use | 0.13** | 0.14** |
| Duration of daily short-video viewing | 0.13** | 0.18*** |
| Metacognition | ||
| MCQ-C-total | 0.25*** | 0.11* |
| MCQ-C-PMW | 0.10* | 0.05 |
| MCQ-C-NMW | 0.24*** | 0.08 |
| MCQ-C-SPR | 0.27*** | 0.16*** |
| MCQ-C-CM | 0.03 | −0.03 |
| Emotion regulation | ||
| DERS-SF-total | 0.36*** | 0.22*** |
| DERS-SF-clarity | 0.29*** | 0.14** |
| DERS-SF-goals | 0.28*** | 0.14** |
| DERS-SF-impulse | 0.32*** | 0.30*** |
| DERS-SF-strategies | 0.32*** | 0.21*** |
| DERS-SF-nonacceptance | 0.26*** | 0.10* |
| Loneliness and perceived social support | ||
| ULS-8-total | 0.22*** | 0.04 |
| MSPSS-total | −0.19*** | −0.05 |
| MSPSS-family | −0.25*** | −0.17*** |
| MSPSS-friends | −0.09* | 0.04 |
| MSPSS-significant others | −0.10* | 0.01 |
Values represent Pearson correlation coefficients (r).
CB = cyberbullying perpetration, CM = cognitive monitoring, CV = cybervictimization, DERS-SF = Difficulties in Emotion Regulation Scale–Short Form, MCQ-C = Metacognitions Questionnaire for Children, MSPSS = Multidimensional Scale of Perceived Social Support, NMW = negative meta-worry, PMW = positive meta-worry, RCBI-II = Revised Cyber Bullying Inventory–II, SPR = superstition, punishment, and responsibility beliefs, ULS-8 = UCLA Loneliness Scale–Short Form.
P < .05.
P < .01.
P < .001.
A hierarchical regression model was tested to predict cybervictimization scores, including gender, age, loneliness, difficulties in emotion regulation, metacognition, and perceived family support. The final hierarchical regression model was statistically significant and explained 18.6% of the variance in cybervictimization (R2 = 0.186, adjusted R2 = 0.178, F[6, 593] = 22.63, P < .001). Assumption checks indicated no evidence of multicollinearity (tolerance values > 0.57, variance inflation factor values < 1.73), and inspection of residuals suggested that the assumptions of normality were met. The results indicated that gender (P < .001), the MCQ-C total score (P = .02), the DERS-SF total score (P < .001), and the MSPSS family subscale (P = .004) significantly predicted cybervictimization scores. Loneliness remained a significant predictor of cybervictimization until the final step of the model. However, after the inclusion of family support, the association between loneliness and cybervictimization was no longer statistically significant, suggesting that a substantial proportion of the variance previously attributed to loneliness may be shared with family support. Given the observed changes in the predictive strength of loneliness after the inclusion of family support, further analyses were conducted using SEM to clarify the interrelationships among these variables. The findings of this hierarchical regression analysis are presented in Table 3.
Table 3.
Results of the hierarchical regression analysis predicting RCBI-II cybervictimization total score.
| Standardized β coefficient | t | P | R 2 | ΔR2 | |
|---|---|---|---|---|---|
| Step 1 | 25.845 | <.001 | 0.006 | ||
| Gender | −0.080 | −1.954 | .051 | ||
| Step 2 | 3.883 | <.001 | 0.007 | 0.001 | |
| Gender | −0.078 | −1.917 | .056 | ||
| Age | 0.028 | 0.694 | .488 | ||
| Step 3 | 2.405 | .016 | 0.163 | 0.056 | |
| Gender | −0.110 | −2.745 | .006 | ||
| Age | 0.058 | 1.448 | .148 | ||
| ULS-8-total | 0.240 | 5.954 | <.001 | ||
| Step 4 | 2.744 | .006 | 0.168 | 0.105 | |
| Gender | −0.189 | −4.860 | <.001 | ||
| Age | 0.029 | 0.771 | .441 | ||
| ULS-8-total | 0.100 | 2.430 | .015 | ||
| DERS-SF-total | 0.366 | 8.689 | <.001 | ||
| Step 5 | 2.359 | .019 | 0.175 | 0.007 | |
| Gender | −0.199 | −5.095 | <.001 | ||
| Age | 0.022 | 0.591 | .555 | ||
| ULS-8-total | 0.099 | 2.401 | .017 | ||
| DERS-SF-total | 0.316 | 6.622 | <.001 | ||
| MCQ-C-total | 0.098 | 2.188 | .029 | ||
| Step 6 | 3.032 | .003 | 0.186 | 0.011 | |
| Gender | −0.190 | −4.886 | <. 001 | ||
| Age | 0.019 | 0.507 | .612 | ||
| ULS-8-total | 0.063 | 1.471 | .142 | ||
| DERS-SF-total | 0.284 | 5.829 | <.001 | ||
| MCQ-C-total | 0.104 | 2.330 | .020 | ||
| MSPSS-family | −0.119 | −2.862 | .004 | ||
β (Beta) = standardized regression coefficient, CV = cybervictimization, DERS-SF = Difficulties in Emotion Regulation Scale–Short Form, MCQ-C = Metacognitions Questionnaire for Children, MSPSS = Multidimensional Scale of Perceived Social Support, P = significance level, R2 = coefficient of determination, ΔR2 = increase in explained variance at each step of the hierarchical regression model, RCBI-II = Revised Cyber Bullying Inventory–II, t = Student t test.
The SEM analyses identified loneliness as a central relational factor associated with cybervictimization. SEM was conducted to examine the associations among emotion dysregulation, perceived social support, duration of social media use, loneliness, and cybervictimization. Whereas family support was examined as a specific protective factor in the hierarchical regression analysis, perceived social support was included as a global construct in the SEM to reflect adolescents’ overall support environment. This framework enabled a broader examination of the interrelationships among psychosocial factors associated with cybervictimization. The cybervictimization model yielded the following fit indices: RMSEA = 0.016, CFI = 0.999, NFI = 0.991, TLI = 0.994, P = .327, χ2 = 3.455, df = 3, and χ2/df = 1.152. Higher levels of emotion dysregulation were associated with greater loneliness (β = 0.27, P < .001), whereas perceived social support was negatively associated with loneliness (β = −0.41, P < .001). Emotion dysregulation was also positively associated with the duration of social media use (β = 0.14, P < .001). Regarding the outcome variable, cybervictimization was positively associated with loneliness (β = 0.09, P = .030) and with the duration of social media use (β = 0.35, P = .001). In addition, emotion dysregulation and perceived social support showed a significant negative covariance (r = −0.31, P < .001). The detailed SEM path diagram for cybervictimization is presented in Figure 1.
Figure 1.
Relationships among factors predicting cybervictimization according to the structural equation modeling (SEM) results. DERS-SF = Difficulties in Emotion Regulation Scale–Short Form, MSPSS = Multidimensional Scale of Perceived Social Support, RCBI-II = Revised Cyber Bullying Inventory–II, ULS-8 = UCLA Loneliness Scale–Short Form.
A hierarchical regression model was also conducted to predict cyberbullying perpetration scores, including gender, age, loneliness, difficulties in emotion regulation, metacognition, and perceived family support. The final hierarchical regression model predicting cyberbullying perpetration was statistically significant and explained 10.8% of the variance in cyberbullying perpetration (R2 = 0.108, adjusted R2 = 0.098, F[6, 593] = 11.91, P < .001). Assumption checks indicated no evidence of multicollinearity (tolerance values > 0.57, variance inflation factor values < 1.73), and inspection of residuals suggested that the assumptions of normality were met. The results showed that gender (P < .001), the DERS-SF total score (P < .001), and the MSPSS family subscale (P = .011) significantly predicted cyberbullying perpetration scores. Detailed results of this hierarchical regression analysis are presented in Table 4.
Table 4.
Hierarchical regression analysis of variables predicting the RCBI-II cyberbullying total score.
| Standardized β coefficient | t | P | R 2 | ΔR2 | |
|---|---|---|---|---|---|
| Step 1 | 27.015 | <.001 | 0.022 | ||
| Gender | −0.148 | −3.657 | <.001 | ||
| Step 2 | 4.156 | <.001 | 0.023 | 0.001 | |
| Gender | −0.147 | −3.620 | <.001 | ||
| Age | 0.025 | 0.626 | .531 | ||
| Step 3 | 3.638 | <.001 | 0.027 | 0.004 | |
| Gender | −0.155 | −3.806 | <.001 | ||
| Age | 0.034 | 0.822 | .412 | ||
| ULS-8-total | 0.066 | 1.605 | .109 | ||
| Step 4 | 3.927 | <.001 | 0.098 | 0.071 | |
| Gender | −0.220 | −5.437 | <.001 | ||
| Age | 0.010 | 0.252 | .801 | ||
| ULS-8-total | −0.049 | −1.131 | .259 | ||
| DERS-SF-total | 0.300 | 6.839 | <.001 | ||
| Step 5 | 3.853 | <.001 | 0.098 | 0.000 | |
| Gender | −0.220 | −5.409 | <.001 | ||
| Age | 0.010 | 0.242 | .809 | ||
| ULS-8-total | −0.049 | −1.132 | .258 | ||
| DERS-SF-total | 0.297 | 5.957 | <.001 | ||
| MCQ-C-total | 0.005 | 0.109 | .914 | ||
| Step 6 | 4.401 | <.001 | 0.108 | 0.010 | |
| Gender | −0.212 | −5.217 | <.001 | ||
| Age | 0.007 | 0.166 | .868 | ||
| ULS-8-total | −0.082 | −1.836 | .067 | ||
| DERS-SF-total | 0.267 | 5.239 | <.001 | ||
| MCQ-C-total | 0.011 | 0.226 | .821 | ||
| MSPSS-family | −0.111 | −2.559 | .011 | ||
β (Beta) = standardized regression coefficient, CV = cybervictimization, DERS-SF = Difficulties in Emotion Regulation Scale–Short Form, MCQ-C = Metacognitions Questionnaire for Children, MSPSS = Multidimensional Scale of Perceived Social Support, P = significance level, R2 = coefficient of determination, ΔR2 = increase in explained variance at each step of the hierarchical regression model, RCBI-II = Revised Cyber Bullying Inventory–II, t = Student t test, ULS-8 = UCLA Loneliness Scale–Short Form.
Similarly, SEM analyses suggested that loneliness may play a central role in the associations among emotional dysregulation, family support, and cyberbullying perpetration. The cyberbullying perpetration model yielded the following fit indices: RMSEA = 0.000, CFI = 1.000, NFI = 0.997, TLI = 1.003, P = .357, χ2 = 0.847, df = 1, and χ2/df = 0.847. Because this model is near-saturated (df = 1), these indices should be interpreted with caution, as they may have limited discriminating value. Higher levels of emotional dysregulation were associated with higher levels of loneliness (β = 0.32, P < .001), whereas family support was negatively associated with loneliness (β = −0.27, P < .001). Loneliness, in turn, was associated with higher levels of cyberbullying perpetration (β = 0.50, P < .001). In addition, emotional dysregulation and family support showed a significant negative covariance (r = −0.35, P < .001). The detailed SEM path diagram for cyberbullying perpetration is presented in Figure 2.
Figure 2.
Relationships among factors predicting cyberbullying perpetration according to the structural equation modeling (SEM) results. DERS-SF = Difficulties in Emotion Regulation Scale–Short Form, MSPSS = Multidimensional Scale of Perceived Social Support, RCBI-II = Revised Cyber Bullying Inventory–II, ULS-8 = UCLA Loneliness Scale–Short Form.
4. Discussion
This study examined associations between cyberbullying involvement and the duration of social media use, emotion regulation, metacognition, loneliness, and perceived social support in an adolescent school sample. SEM findings suggest that cyberbullying involvement during adolescence is shaped by interrelated psychosocial processes rather than isolated risk factors.
Gender differences were consistently observed across both outcomes. Male gender predicted higher levels of cyberbullying perpetration and cybervictimization and remained a significant predictor across all steps of the hierarchical regression models. Previous research on gender differences in cybervictimization has yielded mixed findings.[28] While several studies among adolescents have reported no significant gender differences in cybervictimization,[29,30] others have found higher rates among females.[31,32] With regard to cyberbullying perpetration, previous findings are also inconsistent. While some studies have reported higher levels of cyberbullying perpetration among males,[33] others have indicated that females may engage in cyberbullying perpetration more frequently.[34] These inconsistencies may be related to differences in developmental stages, social media use, and the methodological characteristics of the studies. Future studies are needed to further clarify the role of gender in cyberbullying perpetration and victimization across different age groups and sociocultural contexts. Cultural characteristics related to family connectedness, gender roles, and patterns of adolescent social media use may also influence the observed associations and should be considered when interpreting the findings across different sociocultural contexts. Cultural norms regarding family relationships, peer interactions, emotional expression, and online communication practices may influence both the prevalence and psychosocial correlates of cyberbullying involvement. Therefore, caution is warranted when generalizing the present findings to adolescents from different cultural settings, and future cross-cultural studies are needed to determine whether the observed patterns are consistent across diverse sociocultural environments.
In the present study, both cybervictimization and cyberbullying perpetration were positively associated with social media use and short-form video viewing. Consistent with prior research, several features of social media environments – such as easy accessibility to targets, content editability, anonymity, and the pursuit of social visibility – may increase adolescents’ involvement in cyberbullying.[35–39] While the relationship between general social media use and cyberbullying perpetration has been well documented, empirical research specifically addressing short-form video consumption remains limited.[40] Given the highly immersive, algorithm-driven, and rapidly consumable nature of short-form video platforms, future studies are needed to clarify their potential role in shaping cyberbullying perpetration and experiences.
Maladaptive metacognitive beliefs were associated with both cybervictimization and cyberbullying perpetration. In hierarchical regression analyses, metacognitive beliefs remained a significant predictor of cybervictimization but not cyberbullying perpetration after the inclusion of other psychosocial variables, indicating a more pronounced role in victimization experiences. A study among high school adolescents has shown that maladaptive metacognitive beliefs – particularly negative metacognitive beliefs about worry – are associated with greater exposure to cybervictimization.[41] In a case–control study, Ünal-Aydin et al compared adolescents diagnosed with anxiety and depressive disorders with healthy controls and examined the role of metacognitive beliefs in cyberbullying involvement.[42] Their findings indicated that cybervictimization was higher among adolescents with internalizing disorders and that specific maladaptive metacognitive beliefs, particularly those related to superstition, punishment, and responsibility, were associated with increased cybervictimization, even after accounting for Internet and social media use.[42] Notably, the same study did not identify a significant association between metacognitive beliefs and cyberbullying perpetration. Additionally, Özkan et al reported that among treatment-naive adolescents with attention deficit and hyperactivity disorder, cybervictimization was significantly associated with negative metacognitive beliefs about worry, whereas no significant associations were observed for other metacognitive dimensions.[43] Together, these findings extend previous research by suggesting that metacognitive processes may be more strongly implicated in cybervictimization than in perpetration within community samples.
Emotion regulation difficulties emerged as a central psychosocial factor associated with cyberbullying involvement. Importantly, SEM findings further suggested that emotion regulation difficulties were indirectly associated with cybervictimization via 2 pathways: increased loneliness and longer duration of social media use, both of which were linked to higher levels of cybervictimization. This pattern is consistent with the notion that adolescents who struggle to manage negative affect may experience heightened feelings of social disconnection, which in turn may increase vulnerability to online victimization – potentially by reducing perceived social support, limiting effective coping, or increasing exposure to risky online interactions.[44–46] A similar pattern was observed for cyberbullying perpetration, with emotion regulation difficulties remaining a significant predictor in hierarchical models. Additionally, SEM findings indicated an indirect pathway whereby emotion regulation difficulties were associated with higher levels of loneliness, which in turn were associated with increased cyberbullying perpetration. In this framework, loneliness may reflect unmet belongingness needs and diminished offline social resources, which could increase the likelihood of maladaptive online behaviors, including retaliatory or instrumental aggression.[7] This pattern aligns with theoretical and clinical accounts emphasizing the role of emotion regulation difficulties in shaping aggressive responses in online interactions.[7,10,44,47] These findings may support developmental models suggesting that emotional dysregulation and interpersonal disconnection interact dynamically in adolescents’ online social experiences.
Findings indicated a differentiated pattern of social support across sources. Family support emerged as the most consistent protective factor across both cybervictimization and cyberbullying perpetration, while broader perceived social support appeared particularly relevant for victimization experiences. Moreover, SEM findings indicated that perceived social support was indirectly associated with lower levels of cybervictimization through loneliness and that family support was indirectly associated with lower levels of cyberbullying perpetration via loneliness. In a meta-analysis, deficits in family communication were linked to greater reliance on social media as a compensatory interactional context, increasing adolescents’ exposure to cyber-related risks and underscoring the importance of family- and community-level interventions.[48] In one study, perceived social support from multiple sources was found to be negatively associated with cybervictimization, with family support emerging as particularly influential in earlier developmental stages, while the protective role of peer support increased during adolescence.[49] In a large population-based study, higher levels of social support from family, peers, and teachers were found to be protective against both cyberbullying perpetration and victimization, even in the presence of problematic social media use, indicating that social support can buffer adolescents against engagement in cyber-aggressive behaviors.[50] Overall, the convergence of the present findings with prior research highlights the importance of strengthening social support within broader preventive frameworks, as social resources appear to buffer adolescents against cybervictimization and cyberbullying perpetration through reductions in loneliness. From a clinical and preventive perspective, these findings suggest that intervention programs targeting emotion regulation skills, family communication, and adolescents’ perceived social connectedness may help reduce vulnerability to cyberbullying involvement.
Loneliness was positively associated with cybervictimization at the correlational level. Although loneliness was not significantly associated with cyberbullying perpetration in bivariate analyses, SEM findings suggested that it may function as both a direct correlate of cyberbullying involvement and an intermediary psychosocial mechanism linking emotional and interpersonal factors with adolescents’ online experiences. A recent nationally representative study of Danish adolescents from the Health Behavior in School-aged Children study reported a robust and graded association between loneliness and both school bullying and cyberbullying involvement. Adolescents exposed to bullying across multiple contexts were particularly vulnerable, suggesting that cumulative victimization may intensify feelings of social isolation.[51] Although several studies have examined the association between loneliness and cybervictimization,[51–53] to our knowledge, there is a lack of research directly investigating the relationship between loneliness and cyberbullying perpetration.[54] In the present study, our findings suggest that loneliness may be linked not only to experiences of cybervictimization but also to engagement in cyberbullying perpetration, potentially through both direct and indirect pathways. These results highlight the need for further research to clarify the role of loneliness in cyberbullying perpetration and to better understand the underlying mechanisms connecting social disconnection with different forms of involvement in online aggression.
Several limitations of the present study should be acknowledged. First, the cross-sectional design precludes causal inferences regarding the directionality of the observed associations among cyberbullying involvement, duration of social media use, emotion regulation, metacognition, loneliness, and perceived social support. Accordingly, the findings should be interpreted as associative rather than causal, and the temporal ordering of the observed relationships cannot be determined. It is also possible that some of these relationships are bidirectional or mutually reinforcing over time. Longitudinal and cross-cultural studies are needed to clarify temporal ordering, potential reciprocal relationships, and sociocultural variability in psychosocial mechanisms associated with cyberbullying involvement. Future studies should also examine the cross-cultural applicability of the present findings and evaluate whether the proposed psychosocial model demonstrates similar patterns across diverse sociocultural settings and different adolescent populations. Validation of the model in different cultural contexts may help distinguish universal from culture-specific mechanisms underlying cyberbullying involvement and provide further evidence regarding the stability and generalizability of the proposed SEM model. In addition, the absence of test–retest assessments and longitudinal follow-up data limits the evaluation of the temporal stability and persistence of the observed relationships over time. Second, all measures relied on adolescents’ self-reports, which may be subject to recall bias and social desirability effects, particularly in the assessment of cyberbullying perpetration. These factors may have influenced participants’ responses and should be considered when interpreting the findings. Third, the study was conducted within a school-based adolescent sample, which may limit the generalizability of the findings to out-of-school youth or clinical populations. Finally, although social media use and short-form video viewing were assessed, platform-specific behaviors and qualitative features of online interactions were not examined and warrant further investigation.
Despite these limitations, the study has several notable strengths. This study extends existing research by concurrently examining cyberbullying victimization and perpetration in relation to the duration of social media use, emotion regulation difficulties, maladaptive metacognitive beliefs, loneliness, and perceived social support within a single SEM framework. The use of SEM allowed for the simultaneous examination of direct and indirect pathways, providing a more nuanced and integrative understanding of the psychosocial mechanisms underlying adolescents’ cyber experiences. Additionally, the inclusion of both intrapersonal (emotion regulation, metacognition) and interpersonal (family and peer support, loneliness) factors reflects a comprehensive, developmentally informed approach to cyberbullying involvement research.
5. Conclusion
The present study highlights the interconnected roles of emotion regulation difficulties, maladaptive metacognitive beliefs, loneliness, perceived social support, and social media use in adolescents’ involvement in cyberbullying. Within the SEM framework, loneliness emerged as a central psychosocial mechanism associated with both cybervictimization and cyberbullying perpetration. By integrating intrapersonal and interpersonal factors within a single multivariate model, the study contributes to the growing literature on the psychosocial mechanisms underlying adolescents’ online experiences. These findings highlight the need for prevention and intervention strategies that address emotion regulation skills, family and social support systems, and adolescents’ sense of social connectedness rather than focusing solely on online behaviors.
Acknowledgments
The authors wish to thank the participants and their families for taking part in this study.
Author contributions
Conceptualization: Fatma Coşkun, Hasibe Ağir, Hurşit Ferahkaya, Ömer Faruk Akça.
Data curation: Fatma Coşkun, Hasibe Ağir.
Formal analysis: Fatma Coşkun, Ömer Faruk Akça.
Methodology: Fatma Coşkun, Ömer Faruk Akça.
Supervision: Ömer Faruk Akça.
Visualization: Fatma Coşkun.
Writing – original draft: Fatma Coşkun.
Writing – review & editing: Fatma Coşkun, Ömer Faruk Akça.
Abbreviations:
- CFI
- comparative fit index
- DERS-SF
- Difficulties in Emotion Regulation Scale–Short Form
- MCQ-C
- Metacognitions Questionnaire for Children
- MSPSS
- Multidimensional Scale of Perceived Social Support
- NFI
- Normed Fit Index
- RCBI-II
- Revised Cyber Bullying Inventory–II
- RMSEA
- root mean square error of approximation
- SEM
- structural equation modeling
- TLI
- Tucker–Lewis index
Written informed consent was obtained directly from participants aged 18 years. For participants under 18 years of age, written informed consent was obtained from their parents or legal guardians. In addition, written assent was obtained from the adolescents themselves prior to participation. Written consent for publication was obtained from all participants and their parents or legal guardians.
Ethical approval for this study was obtained from the Ethics Committee of Necmettin Erbakan University on February 19, 2024, under decision number 2024/4809/17135.
The authors have no funding and conflicts of interest to disclose.
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
How to cite this article: Coşkun F, Ağir H, Ferahkaya H, Akça ÖF. Cyberbullying involvement in adolescence: Associations with emotion regulation, metacognitive beliefs, loneliness, and perceived social support in a cross-sectional study. Medicine 2026;105:32(e50114).
Contributor Information
Hasibe Ağir, Email: hasibeagir4101@gmail.com.
Hurşit Ferahkaya, Email: drhursitferahkaya@gmail.com.
Ömer Faruk Akça, Email: dromerakca@gmail.com.
References
- [1].Smith PK, Mahdavi J, Carvalho M, Fisher S, Russell S, Tippett N. Cyberbullying: its nature and impact in secondary school pupils. J Child Psychol Psychiatry. 2008;49:376–85. [DOI] [PubMed] [Google Scholar]
- [2].Cosma A, Molcho M, Pickett W. A Focus on Adolescent Peer Violence and Bullying in Europe, Central Asia and Canada: Health Behaviour in School-Aged Children International Report from the 2021/2022 Survey. Vol 2. WHO Regional Office for Europe; 2024. [Google Scholar]
- [3].Santre S. Cyberbullying in adolescents: a literature review. Int J Adolesc Med Health. 2023;35:1–7. [DOI] [PubMed] [Google Scholar]
- [4].Bonanno RA, Hymel S. Cyber bullying and internalizing difficulties: above and beyond the impact of traditional forms of bullying. J Youth Adolesc. 2013;42:685–97. [DOI] [PubMed] [Google Scholar]
- [5].Weil LG, Fleming SM, Dumontheil I, et al. The development of metacognitive ability in adolescence. Conscious Cogn. 2013;22:264–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Solas-Martínez JL, De la Torre-Cruz MJ, Rusillo-Magdaleno A, Martínez-López EJ. Bullying and cyberbullying is associated with low levels of cognitive and metacognitive learning strategies in young people. Front Psychol. 2025;16:1569400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Jiang Q, Zhao F, Xie X, et al. Difficulties in emotion regulation and cyberbullying among Chinese adolescents: a mediation model of loneliness and depression. J Interpers Violence. 2022;37:NP1105–24. [DOI] [PubMed] [Google Scholar]
- [8].Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav Assess. 2004;26:41–54. [Google Scholar]
- [9].Estévez A, Macía L, López-González H, et al. Cyberbullying and gambling disorder: associations with emotion regulation and coping strategies. J Gambl Stud. 2023;39:1399–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Gomes SDS, Ferreira PDC, Pereira N, Simão AMV. A comparative analysis of adolescents’ emotions and emotion regulation strategies when witnessing different cyberbullying scenarios. Heliyon. 2024;10:e29705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Rieffe C, Camodeca M, Pouw LBC, Lange AMC, Stockmann L. Don’t anger me! Bullying, victimization, and emotion dysregulation in young adolescents with ASD. Eur J Dev Psychol. 2012;9:351–70. [Google Scholar]
- [12].Wu J, Shao Y, Zang W. Perceived social support and self-esteem predict cyberbullying among rural left-behind adolescents. Sci Rep. 2025;15:24676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Costello BJ, Laub JH. Social control theory: the legacy of Travis Hirschi’s causes of delinquency. Annu Rev Criminol. 2020;3:21–41. [Google Scholar]
- [14].Castaño-Pulgarín SA, Otero KLM, Echavarría AAA, Mendoza CER, Parra MCC, Tang JFC. Perceived social support and risk of cyberbullying in adolescents: a systematic review. Qual Rep. 2022;27:1290–304. [Google Scholar]
- [15].Motta V. Key concept: loneliness. Philos Psychiatry Psychol. 2021;28:71–81. [Google Scholar]
- [16].Brewer G, Kerslake J. Cyberbullying, self-esteem, empathy and loneliness. Comput Hum Behav. 2015;48:255–60. [Google Scholar]
- [17].Larrañaga E, Yubero S, Ovejero A, Navarro R. Loneliness, parent-child communication and cyberbullying victimization among Spanish youths. Comput Hum Behav. 2016;65:1–8. [Google Scholar]
- [18].Topcu C, Erdur-Baker O. RCBI-II: the second revision of the revised cyber bullying inventory. Meas Eval Couns Dev. 2018;51:32–41. [Google Scholar]
- [19].Bacow TL, Pincus DB, Ehrenreich JT, Brody LR. The Metacognitions Questionnaire for Children: development and validation in a clinical sample of children and adolescents with anxiety disorders. J Anxiety Disord. 2009;23:727–36. [DOI] [PubMed] [Google Scholar]
- [20].Irak M. Standardization of Turkish Form of Metacognition Questionnaire for Children: relationships with anxiety and obsessive-compulsive symptoms. Turk J Psychiatry. 2012;23:46–52. [PubMed] [Google Scholar]
- [21].Bjureberg J, Ljótsson B, Tull MT, et al. Development and validation of a brief version of the Difficulties in Emotion Regulation Scale: the DERS-16. J Psychopathol Behav Assess. 2016;38:284–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Yiğit I, Guzey Yiğit M. Psychometric properties of Turkish version of Difficulties in Emotion Regulation Scale-Brief Form (DERS-16). Curr Psychol. 2019;38:1503–11. [Google Scholar]
- [23].Hays RD, DiMatteo MR. A short-form measure of loneliness. J Pers Assess. 1987;51:69–81. [DOI] [PubMed] [Google Scholar]
- [24].Yildiz MA, Duy B. Adaptation of the short-form of the UCLA Loneliness Scale (ULS-8) to Turkish for adolescents. Dusunen Adam. 2014;27:194–203. [Google Scholar]
- [25].Zimet GD, Dahlem NW, Zimet SG, Farley GK. The Multidimensional Scale of Perceived Social Support. J Pers Assess. 1988;52:30–41. [Google Scholar]
- [26].Eker D. Çok boyutlu algilanan sosyal destek ölçeğinin gözden geçirilmiş formunun faktör yapisi, geçerlik ve güvenirliği. Turk Psikiyatri Derg. 2001;12:17–25. [Google Scholar]
- [27].Cohen J. Statistical Power Analysis for the Behavioral Sciences. Routledge; 2013. [Google Scholar]
- [28].Wong RYM, Cheung CMK, Xiao B. Does gender matter in cyberbullying perpetration? An empirical investigation. Comput Hum Behav. 2018;79:247–57. [Google Scholar]
- [29].Álvarez-García D, Núñez Pérez JC, Dobarro González A, Rodríguez Pérez C. Risk factors associated with cybervictimization in adolescence. Int J Clin Health Psychol. 2015;15:226–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Hinduja S, Patchin JW. Cyberbullying: an exploratory analysis of factors related to offending and victimization. Deviant Behav. 2008;29:129–56. [Google Scholar]
- [31].Lozano-Blasco R, Quilez-Robres A, Latorre-Cosculluela C. Sex, age and cyber-victimization: a meta-analysis. Comput Hum Behav. 2023;139:107491. [Google Scholar]
- [32].González‐Cabrera J, Díaz‐López A, Caba‐Machado V, et al. Epidemiology of peer cybervictimization and its relationship with health‐related quality of life in adolescents: a prospective study. J Adolesc. 2023;95:468–78. [DOI] [PubMed] [Google Scholar]
- [33].Yudes C, Rey L, Extremera N. The moderating effect of emotional intelligence on problematic Internet use and cyberbullying perpetration among adolescents: gender differences. Psychol Rep. 2022;125:2902–21. [DOI] [PubMed] [Google Scholar]
- [34].Barlett C, Coyne SM. A meta-analysis of sex differences in cyber-bullying behavior: the moderating role of age. Aggress Behav. 2014;40:474–88. [DOI] [PubMed] [Google Scholar]
- [35].Barlett CP, Dewitt CC, Maronna B, Johnson K. Social media use as a tool to facilitate or reduce cyberbullying perpetration: a review focusing on anonymous and nonanonymous social media platforms. Violence Gend. 2018;5:147–52. [Google Scholar]
- [36].Barlett CP, Gentile DA, Chng G, Li D, Chamberlin K. Social media use and cyberbullying perpetration: a longitudinal analysis. Violence Gend. 2018;5:191–7. [Google Scholar]
- [37].Chan TKH, Cheung CMK, Wong RYM. Cyberbullying on social networking sites: the crime opportunity and affordance perspectives. J Manage Inf Syst. 2019;36:574–609. [Google Scholar]
- [38].Chan TKH, Cheung CMK, Lee ZWY. Cyberbullying on social networking sites: a literature review and future research directions. Inf Manag. 2021;58:103411. [Google Scholar]
- [39].Longobardi C, Settanni M, Fabris MA, Marengo D. Follow or be followed: exploring the links between Instagram popularity, social media addiction, cyber victimization, and subjective happiness in Italian adolescents. Child Youth Serv Rev. 2020;113:104955. [Google Scholar]
- [40].Yao Q, Zhu W, Gao Y, et al. The impact of bullying victimization on short video addiction in adolescents: the role of emotional distress and neural mechanisms. Addict Biol. 2025;30:e70038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].Güçlü-Aydogan M, Ünal-Aydin P, Aydin O. The effects of psychological resilience, self-efficacy and metacognition on cyberbullying among adolescents. Child Adolesc Soc Work J. 2023;42:57–65. [Google Scholar]
- [42].Ünal-Aydin P, Özkan Y, Öztürk M, Aydin O, Spada MM. The role of metacognitions in cyberbullying and cybervictimization among adolescents diagnosed with major depressive disorder and anxiety disorders: a case–control study. Clin Psychol Psychother. 2023;30:659–70. [DOI] [PubMed] [Google Scholar]
- [43].Özkan Y, Öztürk M, Tvrtkovic S, Aydin O, Ünal-Aydin P. Exploring the associations between symptom severity, metacognition, problematic social media use and cyberbullying in treatment-naïve adolescents with attention deficit and hyperactivity disorder. Addict Behav. 2025;160:108169. [DOI] [PubMed] [Google Scholar]
- [44].Arató N, Zsidó AN, Lénárd K, Lábadi B. Cybervictimization and cyberbullying: the role of socio-emotional skills. Front Psychiatry. 2020;11:248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [45].Rey L, Neto F, Extremera N. Cyberbullying victimization and somatic complaints: a prospective examination of cognitive emotion regulation strategies as mediators. Int J Clin Health Psychol. 2020;20:135–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [46].Schunk F, Zeh F, Trommsdorff G. Cybervictimization and well-being among adolescents during the COVID-19 pandemic: the mediating roles of emotional self-efficacy and emotion regulation. Comput Hum Behav. 2022;126:107035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [47].Rębisz S, Jasińska-Maciążek A, Grygiel P, Dolata R. Psycho-social correlates of cyberbullying among Polish adolescents. Int J Environ Res Public Health. 2023;20:5521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [48].Lozano-Blasco R, Barreiro-Collazo A, Romero-Gonzalez B, Soto-Sanchez A. The family context in cybervictimization: a systematic review and meta-analysis. Trauma Violence Abuse. 2024;25:2143–57. [DOI] [PubMed] [Google Scholar]
- [49].Rodríguez-Enríquez M, Álvarez-García D, Rodríguez-Alvarado S, Ares-Ferreirós M. Students’ perceived social support in the transition from primary to secondary education: grade-related trends and association with cybervictimization. Rev Psicodidáct (Engl Ed.). 2025;30:500172. [Google Scholar]
- [50].Borraccino A, Marengo N, Dalmasso P, et al. Problematic social media use and cyber aggression in Italian adolescents: the remarkable role of social support. Int J Environ Res Public Health. 2022;19:9763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [51].Madsen KR, Damsgaard MT, Petersen K, Qualter P, Holstein BE. Bullying at school, cyberbullying, and loneliness: national representative study of adolescents in Denmark. Int J Environ Res Public Health. 2024;21:414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [52].Cañas E, Estévez E, León-Moreno C, Musitu G. Loneliness, family communication, and school adjustment in a sample of cybervictimized adolescents. Int J Environ Res Public Health. 2020;17:335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [53].Espinoza G, Schacter HL, Juvonen J. School and cybervictimization across high school: normative developmental trajectories and bidirectional links with loneliness. J Res Adolesc. 2020;30:787–99. [DOI] [PubMed] [Google Scholar]
- [54].Han Z, Wang Z, Li Y. Cyberbullying involvement, resilient coping, and loneliness of adolescents during COVID-19 in rural China. Front Psychol. 2021;12:664612. [DOI] [PMC free article] [PubMed] [Google Scholar]