Cyberbullying in students: forms of aggression, risk factors, and educational responses in digital environments

Eligibility criteria

The establishment of eligibility criteria was a deliberate effort to guarantee the relevance, quality, and pertinence of the studies incorporated within the review. The establishment of four inclusion criteria was fundamental to the study. Initially, articles published between 2015 and 2025 were selected to ensure contemporary coverage of knowledge. Secondly, the selection of studies was constrained to those published in peer-reviewed scientific journals, a practice that was implemented to ensure the rigour and methodological validity of the research. Thirdly, research that directly addressed cyberbullying in students, its risk factors, or educational responses was included. Fourthly, publications in both Spanish and English were accepted in order to expand the geographic and linguistic reach of the findings.

The exclusion process was executed in three successive phases. The initial phase of the process involved the elimination of records that exhibited indexing errors. These records were then further filtered based on their relevance to the designated topic or the required document type. This filtration process was implemented with the intention of precluding the inclusion of irrelevant documents from the outset. In the second phase of the process, articles for which the full text was unavailable were excluded from the study, even in cases where institutional access channels had been exhausted. This measure ensured a comprehensive review of the content and avoided biased interpretations based on abstracts or titles. The third phase entailed the exclusion of data at the researcher’s discretion, with the objective of ensuring thematic and methodological relevance. At this stage of the process, works were eliminated that, despite meeting the criteria set out in the previous stage, presented tangential approaches or had little relevance to the objectives stated.

Sources of information

The information sources employed in this review were Scopus and Web of Science (WoS), selected on the basis of their international recognition and academic coverage. Scopus is a multidisciplinary database providing access to relevant scientific publications in the social sciences, humanities and education. The broad nature of this subject area facilitates the identification of studies on cyberbullying from pedagogical, psychological and sociological perspectives, thereby fostering a comprehensive understanding of the phenomenon. The Web of Science (WoS) database is distinguished by its inclusion of high-quality, peer-reviewed academic publications.

This database provides access to established research in various disciplines, including specific studies on educational environments and digital issues. The combination of both sources ensures thematic diversity and methodological rigor. The combined use of Scopus and WoS was a response to the need to construct a contemporary, rigorous, and representative corpus of available knowledge. This integration served to mitigate potential biases arising from disparities in editorial coverage between regions or disciplines, as noted by Asubiaro, Onaolapo, and Mills (2024).

Search strategy

The search strategy was defined on the basis of equations designed specifically for each database, in alignment with the established inclusion criteria. In Scopus, an equation based on title fields was used, with keyword combinations such as “cyberbullying,” “online harassment,” “students,” “school,” “university,” and “higher education” integrated using Boolean operators “AND” and “OR.” In Web of Science, an equivalent equation was applied, adapted to its syntax, using the TS = operator to perform thematic searches. Both platforms incorporated filters based on publication period (2021–2025), subject areas related to social sciences and education, and languages Spanish and English. Furthermore, a manual search for additional references was performed on the selected studies.

Selection process

The selection process was systematically carried out in three consecutive phases. The initial stage of the process entailed the review of titles to identify and discard records deemed irrelevant. The second phase entailed the perusal of abstracts to identify studies that were congruent with the research objectives. The third approach entailed a comprehensive analysis of the texts to implement exclusion criteria. Each phase of the study was carried out by the researcher in accordance with a uniform protocol, thereby ensuring consistency in the selection process. The entire process flow was meticulously documented using the PRISMA diagram, thereby ensuring optimal traceability and transparency in each decision made. Figure 1 presents the flowchart recommended by the PRISMA 2020 declaration. The diagram herein illustrates the comprehensive study selection process, which is delineated into the identification, selection, eligibility, and inclusion phases.

Figure 1. PRISMA flowchart.

Prepared by the authors based on Scopus and Web of Science.

Data processing

The data processing was performed using Microsoft Excel as the designated recording tool. The selected articles were then organised according to three thematic axes: forms of aggression, risk factors, and educational responses. For each record, the relevant variables were coded according to the review objectives. This configuration facilitated the systematic organisation of the collected information. The procedure facilitated qualitative and comparative analysis, ensuring a clear interpretation of the results and the identification of patterns, similarities, and gaps in available knowledge.

Risk of bias

The risk of bias was addressed through a systematic review of the inclusion criteria and the controlled application of filters. Nevertheless, potential biases persist due to the exclusive use of Scopus and Web of Science, as well as the selection of specific search terms. The limitations imposed by the language of publication are acknowledged by restricting the analysis to publications in Spanish and English. Furthermore, the potential for reporting biases, associated with the editorial tendency to favour positive or significant results, was considered.

Analysis of results

The results indicate that studies on types of cyberbullying primarily focus on the analysis of victimisation. This phenomenon aligns with the findings of preceding studies that have highlighted the deleterious impact of this experience on student mental health (Thumronglaohapun et al., 2022). The concept of the cyberbully-victim figure reflects complex dynamics where roles intertwine, as reported by Lam et al. (2022), linking victimization and perpetration with social anxiety. This classification serves to reinforce the notion that instances of cyberbullying in educational settings manifest in multiple forms, with varying consequences for each student depending on their individual experiences. The results of the study demonstrate that the most frequently discussed vulnerability factors are related to mental health, gender, and social media use. This is indicative of the diversity in roles and consequences. This thematic focus is consistent with research that has documented significant impacts of cyberbullying on self-esteem and emotional well-being (Alsawalqa, 2021). Furthermore, it has been demonstrated that prolonged exposure to digital media combined with experiences of victimisation increases mental health risks, including suicidal ideation (Mantey et al., 2023). The findings indicate that the vulnerabilities associated with cyberbullying vary according to individual, familial, and social conditions that increase the likelihood of being affected.

In this context, social media has been identified as the most frequently mentioned digital environment in relation to student cyberbullying, consistent with studies that associate these platforms with risky behaviours among young people (Pichel et al., 2022). However, the presence of channels such as instant messaging and email, as well as the frequent omission of specific platforms in the reports, reveals a wide range of digital environments involved. This diversity suggests that experiences of cyberbullying are mediated by the virtual ecosystem, influenced by elements such as peer pressure and limited moral regulation in students (Lee et al., 2021a).

The consequences of these digital interactions are primarily reflected in psychological maladjustment, along with depressive symptoms and suicidal ideation. The emotional impact is further compounded by personal and family circumstances that exacerbate the psychological consequences (Lee et al., 2021a). In addition to the emotional effects, behavioural, academic and social repercussions emerge that reflect the cross-cutting nature of the harm associated with cyberbullying. In this context, recent studies have emphasised the value of school cohesion as a protective factor against the effects on mental health, by reducing levels of depression and suicidal ideation (Lucas-Molina et al., 2022).

In response to these effects, the reviewed studies propose various educational strategies to address cyberbullying. These include public policy recommendations, psychological counselling, and comprehensive prevention programmes. Furthermore, they advocate for the establishment of a conducive school environment and the fostering of a collaborative partnership between families and educational institutions. These actions are complemented by peer support, awareness campaigns, emotional intelligence training, strengthening of coping skills, digital citizenship education, and early intervention mechanisms. The strategies outlined herein seek to address the psychological and social factors that shape aggressive behaviour in digital environments. This approach is informed by research focusing on social norms and self-esteem in the context of cyberbullying (Shaikh et al., 2021; Pomytkina et al., 2021).

Comparison of results with other studies

The results obtained were organised around forms of aggression, vulnerability factors, digital platforms, consequences, and educational responses, providing a comprehensive view of cyberbullying in students. This structure is consistent with the multidisciplinary perspective of Ahmed, Chaudhary, and Shahzad (2025), who approached the phenomenon from the perspectives of criminology and law, highlighting the influence of anonymity, low empathy, and social tension on aggressive digital behaviour. It is acknowledged by both studies that aggressive behaviour online is a complex phenomenon, and that a comprehensive approach is required which considers both individual and social factors. With regard to educational responses, the present study emphasises the implementation of policies, psychological counselling, and comprehensive prevention programmes.

These findings are consistent with those reported by Bussu et al. (2025), who emphasised the role of family and social support as protective factors against cyberbullying in university communities. Nevertheless, while the present study addresses educational responses at multiple levels, Bussu et al. focus on higher education and victim support strategies, highlighting gaps in institutional responses and the need for better practices. These are aspects that are less explored in the present analysis. With regard to risk factors, Tabuk and Akbaş (2025) provide a specific perspective on the relationship between physical activity, aggression, and cyberbullying in young university students. The researchers identify a weak correlation between aggression and cyberbullying, and highlight the role of digital anonymity in reducing gender-based aggression patterns.

These findings partially coincide with the present study, which also demonstrates the multifactorial nature of cyberbullying. However, the current analysis delves into demographic differences and psychological factors such as self-esteem and social norms, aspects also addressed by Shaikh et al. (2021). Tintori, Ciancimino, and Cerbara’s (2025) study identifies the vulnerability of adolescents who act as both victims and aggressors simultaneously (victim-perpetrators), linking this condition with low self-esteem, anxiety, and reciprocal aggressive behaviours. This finding serves to broaden the scope of the study, highlighting a vulnerable group that requires differentiated attention and suggesting a line of future research in national and international educational contexts. Finally, Lu’s (2025) systematic review corroborates the patterns observed in this study, emphasising the intricate etiology of cyberbullying and its manifold psychological and academic ramifications. The review emphasises the necessity for multidimensional interventions that integrate individual, community, and policy levels, thus supporting the thematic structure applied in this analysis. However, the report also draws attention to the dearth of longitudinal studies and rigorous evaluations of preventive programmes, a paucity that is also evident in the national literature.

Proposed conceptual framework

As illustrated in Figure 7, the conceptual model integrates the primary components of the study, including the categories of aggression, vulnerability factors, digital platforms, consequences, and educational responses. The diagram illustrates the interaction between individual and social variables with the forms of cyberbullying in various digital environments, generating psychological and academic effects on victims and aggressors. Furthermore, the text presents an array of educational interventions meticulously designed to mitigate the aforementioned impacts, emphasising the intricate relationship between prevention, support, and the technological context. This model synthesises the complexity of the phenomenon in a clear and structured manner.

Figure 7. Conceptual model of student cyberbullying.

Prepared by the authors.

Implications

The study’s results provide theoretical, policy, and practical implications that improve the understanding and approach to cyberbullying in digital environments. Theoretically, the multidimensional complexity of the phenomenon is confirmed, integrating individual, social, and technological variables. The precise classification of types of aggression and the identification of vulnerability factors, such as low self-esteem, social norms, and digital anonymity, expand explanatory models by demonstrating the interaction and mutual conditioning in the development of aggressive behaviour online. This systematic approach serves to strengthen interdisciplinary conceptual frameworks, encompassing psychological, sociological, and technological aspects. The incorporation of particular digital platforms serves to enhance the theoretical analysis, thereby demonstrating that the digital environment functions as a mediating factor in aggressive dynamics, rather than as a neutral setting.

This finding necessitates the incorporation of the technological context into future models to address the phenomenon in a contextualised manner. From a policy perspective, the necessity to formulate clear, specific, and comprehensive educational policies that respond to students’ digital reality is emphasised. The heterogeneity of forms of aggression and their psychological, social, and academic consequences necessitate regulatory measures that proactively prevent and sanction cyberbullying, with particular emphasis on vulnerable groups. It is imperative that policies encompass preventative measures, such as the provision of training in socio-emotional skills, education in digital citizenship, and the promotion of positive school climates. The importance of regulatory frameworks that promote collaboration between families, schools, and authorities to ensure coordinated and effective interventions is emphasised.

Public policies should establish standardised protocols for early detection, timely intervention, and psychological support, as well as create safe physical and virtual spaces. In the absence of adequate institutional support, educational responses are characterised by fragmentation and reduced effectiveness. This underscores the necessity for the allocation of resources and the provision of training for the effective implementation of educational responses. From a pragmatic standpoint, the findings direct institutions towards the implementation of targeted interventions that address particular forms of aggression, including insults, identity theft and social exclusion. This precision precludes generic responses that fail to consider the diversity of cyberbullying manifestations. The identification of vulnerability factors, including but not limited to low self-esteem, digital anonymity, and a paucity of social support, forms the basis for the design of programs that are intended to strengthen student resilience.

The promotion of socio-emotional skills, digital empathy, and coping mechanisms is imperative in order to prevent victimisation and reduce aggression. In order to achieve this, educational practices should incorporate workshops, counselling, and awareness campaigns that involve students, teachers, and families, thereby engendering collective commitment. Furthermore, the identification of specific digital platforms enables the tailoring of interventions to the technological contexts in which aggression occurs.

This involves the education of students in the responsible use of social media, messaging applications, and educational platforms, as well as the collaboration with technology developers to create safe environments and accessible reporting mechanisms. The integration of technology as an ally in the prevention and mitigation of problems represents a significant advance for institutional practice. The documented consequences on mental health and academic performance demonstrate the need to offer specialised psychological support and mentoring to victims and aggressors. Institutions must ensure that they have trained human resources capable of intervening effectively to promote school environments that foster recovery and comprehensive well-being.

It is imperative that this practice is complemented by the implementation of policies that ensure the continuity and evaluation of interventions in order to guarantee sustainable results. The study provides a comprehensive framework that articulates theory, policy, and practice to address student cyberbullying. In principle, this approach facilitates a more comprehensive understanding by offering a multidimensional and contextual perspective. From a political standpoint, it underscores the necessity for clear, comprehensive, and coordinated regulatory frameworks that are designed to safeguard students and nurture secure digital environments. In educational practice, it guides specific, contextualised, and evidence-based interventions, with an emphasis on prevention, support, and collaboration between educational stakeholders and families.

Limitations

The study presents methodological and analytical limitations that must be considered when interpreting the results. While the sample is representative of certain student groups, it does not encompass the full diversity of socioeconomic, cultural, and geographic contexts, which restricts the generalisability of the findings to broader populations. The research is primarily based on self-reported data, which has the potential to introduce biases due to subjective perception, memory, and the willingness to disclose experiences of cyberbullying. The cross-sectional nature of the study prevents the establishment of causal relationships between variables, thus limiting the analysis to associations at a specific point in time. Despite the integration of individual, social, and technological factors, a comprehensive analysis of specific psychological variables or the impact of family contexts was not undertaken. The absence of longitudinal follow-up data poses a significant challenge in evaluating the effectiveness and sustainability of educational responses.

Lines of future research

Future lines of research should concentrate on expanding and deepening knowledge about cyberbullying in students, taking into account the findings and limitations of this study. It is recommended that research be conducted with diverse and representative samples that include different socioeconomic, cultural, and geographic contexts. This will facilitate the generalisation of results and understanding of the variations in the phenomenon, allowing for the design of contextualised strategies. It is imperative to implement longitudinal designs that analyse the evolution of cyberbullying over time and establish causal relationships between risk factors, types of aggression, digital platforms, and consequences. The efficacy and durability of educational responses, in addition to the evolution of aggressive or victimising behaviours across various educational stages, will be illuminated by these designs.

A more thorough investigation is required into the specific psychological variables that influence the dynamics of cyberbullying, including self-esteem, resilience, self-control, and social norms. The integration of interdisciplinary approaches, combining psychology, sociology and technology, will foster a more complete understanding. In addition, it is pertinent to examine the impact of contextual factors, including family dynamics, school climate, and community influences, on the prevention and mitigation of the phenomenon. Another significant research trajectory that merits attention is the rigorous evaluation of the efficacy of educational interventions and public policies through the utilisation of experimental and mixed methods. This will facilitate the identification of good practices, the adaptation of interventions to specific needs, and the optimisation of resources. The progression of technology has necessitated the undertaking of research on novel digital platforms and emergent manifestations of cyberbullying.

In order to develop effective strategies for addressing bullying, it is essential to examine the influence of characteristics associated with applications, social networks, video games and virtual environments on the manifestation and perception of bullying behaviours. Collaboration with technology developers to design accessible prevention and reporting mechanisms represents a promising area of research. Finally, it is recommended that the cultural and gender dimensions be explored in greater depth, in order to understand their influence on the experience of cyberbullying and the institutional response. The integration of qualitative methodologies will facilitate the capture of diverse voices and the enhancement of knowledge pertaining to the phenomenon, thereby providing a foundation for the formulation of sensitive and effective policies. It is recommended that future research on the subject of cyberbullying adopt a multidimensional, longitudinal, and interdisciplinary approach. In order to achieve this, varied contexts, psychological and social variables, and rigorous evaluations of interventions should be integrated. This guidance will contribute to the design of effective and tailored strategies to prevent and address cyberbullying in digital educational environments.