3.1. Research Design

Utilizing an explanatory multi-case approach, our study used Habermas (1984)’s Theory of Communicative Action principles and qualitative research to investigate the phenomena within its real-life setting (Yin, 2009). This design was selected based on examining intricate social phenomena and comprehending the subtle experiences of individuals in their natural settings (Denzin & Lincoln, 2011).

3.2. Case Selection

Following Yin (2009)’s recommendations, we conducted a detailed examination of eight cases conveniently selected from three universities (for ethical reasons, we named them Universities X, Y, and Z). The selection criteria were designed to ensure a rich diversity of experiences, aiming for depth over breadth (Creswell, 2007; Stake, 2006). According to Creswell and Creswell (2018), no specific minimum number of participants is needed for qualitative interviews. This is because qualitative research prioritizes the depth and richness of data rather than quantifiable validity indicators. For case study design, the Creswell and Creswell (2018) suggest that even a single case can be profoundly informative if it is well-chosen, critical, unique, or revealing. That will best help to understand, explore, and explain the phenomenon of interest.

Participants were selected through a purposive sampling method, targeting individuals from universities X, Y, and Z (which have been anonymized for confidentiality purposes) who have experienced cyberbullying. This selection aimed to capture diverse cyberbullying experiences by including participants from varied university settings. Our methodological approach was designed to yield in-depth insights into the multifaceted nature of cyberbullying across different educational environments. To ensure the inclusion of individuals who genuinely encountered cyberbullying, we relied on self-reported experiences confirmed through initial screening processes:

Initial identification of potential participants was conducted via online surveys distributed through university email lists. The surveys also served as a preliminary screening tool to assess the frequency and types of cyberbullying experiences among respondents. Following this, students who met our criteria based on their survey responses were contacted for further evaluation through an initial brief interview via email or phone, where we clarified their experiences and ensured their willingness to participate in more extended, semi-structured interviews.

After agreeing to participate, participants received invitation and consent forms, which detailed the study’s purpose, procedures, and ethical considerations. These forms also emphasized their rights as participants, including the right to withdraw from the study at any point without repercussion and the option to request the removal of any of their statements from the research findings.

3.3. Data Collection

This study collected data using interviews, surveys, and documentation (Fig. 2). The study’s validity is strengthened through data triangulation (Yin, 2017).

Fig. 2

Methods of data collection.

Semi-structured interviews allowed participants to express their cyberbullying experiences in depth. The data collecting method of choice was semi-structured online interviews conducted through platforms via Google Meet or Zoom for convenience. Follow-up inquiries were conducted via chat or email. The researcher prioritized the anonymity and privacy of the participants by employing pseudonyms to safeguard their identities. The interviews were recorded solely for data collection and were securely saved with password protection. The interview questions encompass a range of subjects about cyberbullying encounters, emotional effects, coping mechanisms, societal norms, and the influence of peers. Furthermore, using a semi-structured framework facilitated the process of probing and clarification, leading to a deeper understanding of the phenomena being investigated. Throughout the interview, participants were allowed to express any additional knowledge or concerns not covered by the questions. To enhance the reliability of the data, the researcher implemented member checking, a method wherein the participants were given access to their interview transcripts to validate the precision of the recorded data and the conclusions drawn from the study (Creswell, 2007). This methodology improves the veracity and reliability of the study.

Document analysis involves systematically collecting and examining diverse textual data derived from social media platforms such as Facebook. This includes scrutinizing social media posts, messages exchanged through messenger services, and user comments. The study of documentation entailed scrutinizing the language and context of cyberbullying interactions on Facebook, which was found to be the most prominent venue for such activities within our sample.

Qualitative research involves using surveys to identify and examine the variety within a particular population. However, the primary aim of this study is not to ascertain frequencies, means, and other statistical measures, but rather to evaluate the level of diversity related to a particular subject of interest within a given population (Jansen, 2010). Google Forms was used to conduct the survey. This method was chosen for its convenience and easy accessibility through email, making it suitable for efficiently reaching many respondents. Moreover, this approach required minimal time and cost for both the researchers and participants. The estimated completion time for the survey was less than 10 minutes. The survey questionnaire was distributed conveniently among students from three participating universities involved in the study.

3.4. Database Management

A case study database was established to organize and manage the data systematically, in alignment with Yin (2017)’s guidelines. This facilitated an efficient analysis process and ensured the traceability of conclusions back to the original data. For example, interview transcripts were called ‘Transcript_Case1,’ ‘Transcript_Case2,’ etc. This standardized naming approach made data administration easy.

3.5. Data Analysis of the Multiple Case Study

The analysis was conducted using a three-phase methodology, following Yin (2017)’s methodological structure. At first, a within-case analysis was conducted to emphasize the distinctive characteristics of each instance, focusing on their individuality. The researchers categorized the initial classifications into the main themes using categorical aggregation. Stake (2006) defines categorical aggregation as combining category data by addition. In this section, the themes identified through the analysis of the case studies are given. The relevance of the findings can be inferred from the themes identified. Subsequently, a cross-case analysis was conducted to identify repeating patterns and variations across the different cases, providing a comparative viewpoint. Then, the process of assertion analysis was employed to incorporate the data into the existing theoretical framework, making it easier to evaluate the main themes and derive implications from the study. The analysis of statements or the interpretation of cases constitutes the concluding phase of analysis in a multi-case study (Creswell, 2007; Merriam, 2009). Table 1 shows sample coding, and Table 2 the derived theme.

3.6. Ensuring Research Quality

We addressed construct, external, and internal validity through methodological rigor, participant diversity, and member checking, per Yin (2017)’s criteria for multi-case study research.

3.7. Ethical Considerations

The research was conducted strictly in compliance with the ethical standards outlined in the Research Service Agreement (F-URD-4.1-RSA, dated 11/19/2019). Before their participation, all individuals were thoroughly briefed on the study’s objectives, assured of the confidentiality of their responses, and informed of their right to withdraw from the study at any time without any repercussions. Written informed consent was obtained from every participant, affirming their voluntary engagement and comprehensive understanding of the study’s procedures. To safeguard the participants’ anonymity and the sensitive nature of the information collected, extensive measures were implemented to anonymize and securely store data, consistent with the guidelines and ethical standards of the university.

4.1. Case Study Overview

The study involves participants from various academic backgrounds, assigned pseudonyms for privacy. It includes four students from public University X enrolled in chemistry, elementary education – language, and information technology. Additionally, two participants studied computer engineering at the private University Y, while another two pursued computer science and computer engineering at the public University Z. This mix of public and private urban university participants enriches the study with varied experiences and perspectives.

4.2. Key Themes Emerging from the Research

This section provides an overview of the main themes that have emerged from the data, shedding light on the intricate aspects of the research topic. Each theme is explored in detail, providing insights into the identified patterns and their ramifications.

4.3. Cyberbullying and Social Media Platforms

The survey results from the three universities reveal a concerning spectrum of cyberbullying behaviors (Fig. 3). Degrading comments or hate speech was most common at 23.8%, followed by threats (17.2%), and unwanted sexual messages (14.5%). Other types included sharing unwanted pictures (16.3%), discrimination (8.4%), and impersonation (7.9%). Group harassment and ‘outing’ someone’s private information were less common but still reported. This highlights the need for better online safety and respect across social media platforms.

Fig. 3

Various forms of cyberbullying experienced by college students.

Facebook and Messenger (73%) are the primary mediums used where cyberbullying occurs, while platform X accounts for 25% of cyberbullying cases (Fig. 4).

Fig. 4

Mediums of cyberbullying.

This study’s exploration of eight cases reveals the dual nature of social media platforms such as Facebook and its Messenger feature (Fig. 5), which have become conduits for cyberbullying despite their intent to foster connections. These platforms have been misused for anonymous harassment, leading to lasting emotional distress. Facebook’s design and group features exemplify how they can enable the dissemination of misinformation and malicious personal assaults, thus weakening the integrity of respectful and authentic communication. This situation necessitates a collaborative endeavor to advocate for digital empathy, accountable online conduct, and platform regulations that reduce cyberbullying to establish a safer and more inclusive digital atmosphere.

Fig. 5

Example image of Facebook page and group chat messenger.

The analysis of the cases reveals how IT platform features and dynamics on social media can either facilitate or impede speech acts characteristic of cyberbullying. The characteristics of these IT platforms can significantly influence the prevalence and nature of cyberbullying incidents:

Visibility and Virality: Due to social media’s quick information distribution, offensive speech activities may spread more quickly. Bullying’s effects can be amplified by information that spreads rapidly, as was the case in Andrew’s case when false rumors spread to a large audience. Bullies may target more people since it is so simple to share content, which worsens the damage they do.

Amplification of Power Imbalances: Platform structures and algorithms can magnify the influence of certain users over others. Features such as trending topics, like and share mechanisms, and visibility algorithms can disproportionately elevate the voices of those with large followings or those who manipulate content for broader dissemination. This dynamic can exacerbate cyberbullying, misinformation, and other forms of online harm, as it allows dominant voices to overshadow or silence less powerful individuals.

Anonymity and Pseudonymity: Cyberbullying may be made more accessible by the option to post anonymously or use a pseudonym, since it gives offenders a sense of anonymity and allows them to use hurtful comments without worrying about being identified. This anonymity may give bullies the confidence to propagate untrue information, launch personal assaults, and use disparaging language. On the other hand, platforms that demand genuine names may prevent such speech actions by enhancing responsibility and lowering the possibility of bullying motivated by anonymity.

Privacy Controls: Cyberbullying may be stopped by features that let users manage their privacy settings and control who can view their material. The usage of private settings demonstrates how privacy measures may prevent bullies from getting access to sensitive material. By limiting visibility, bullies may have fewer opportunities to target their victims.

This study highlights how social media’s features can influence cyberbullying, aligning with Habermas’s conditions for rational discourse: truth, non-coercion, sincerity, and normative rightness. The quick spread of misinformation challenges the truth. Power imbalances and anonymity can create a coercive environment, limiting free speech. Misleading stories and tactics undermine sincerity. The absence of universal respectful norms hampers normative rightness. These findings emphasize the need for platforms to foster environments that discourage cyberbullying and promote respectful, truthful, and free communication, supporting a healthier digital space.

5.1. Strategic Approaches to Mitigating Cyberbullying

To address the issue of cyberbullying on social media, implementing a strategy that understands today’s online behavior and fundamental tenets of effective communication is necessary. This involves examining how people interact online today and combining that with fundamental communication principles. Drawing on Habermas’s principles, a pillar is proposed (Fig. 6) to foster a digital environment where genuine, rational, and enriching interactions prevail, contributing to more respectful and enriching interactions.

Fig. 6

Pillars to counter cyberbullying on social platforms.

5.2. Theoretical and Practical Implications of Study Findings

The implications of our findings extend into both theoretical and practical domains. The study underscores the critical need for governance models on social media that foster meaningful participation and dialogue (McCarthy et al., 2023), emphasizing the role of ethical communication in platform policies. Furthermore, the success of interventions in various contexts highlights the potential of integrated, theory-based approaches to mitigate cyberbullying effectively (Barlett, 2023; Özgür, 2020).

To address cyberbullying and foster a culture of respect and understanding online, it is crucial to prioritize ethical involvement and mutual respect, which are key principles of discourse ethics. Adapting discourse ethics for digital platforms emphasizes ethical conduct within online learning communities and provides a roadmap for creating digital environments where constructive and respectful discourse can flourish (Luppicini, 2009).

Our research uncovers insights into the dynamics of digital communication, primarily focusing on the challenges and potential pathways forward in addressing cyberbullying. It illustrates how online communication can differ from face-to-face interaction, with anonymity and the digital space presenting new challenges. It also highlights a growing online aggression culture, reflecting broader societal issues. The blending of online and offline lives, as seen in personal experiences, suggests a close connection between the digital and physical worlds. While our study has primarily focused on the disruptive impact of cyberbullying within digital communications, it is essential to acknowledge the existence of online communities and interactions where rational discourse thrives. These positive examples serve as guiding lights, showcasing the capacity of online platforms to foster environments conducive to informed, respectful, and productive conversations. Platforms that incorporate design features encouraging transparency, accountability, and mutual respect among users significantly contribute to the prevalence of such rational discourse.

The theoretical implications reaffirm Habermas’s theory’s applicability in analyzing the complexities of digital communication, as shown in Table 5.

Practically, there is a call for digital etiquette education and targeted support for cyberbullying victims. Policy implications suggest stricter cyberbullying laws, mandated platform regulation, and digital literacy in educational curricula. Our study emphasizes the importance of using diverse and well-informed strategies to address cyberbullying, which aligns with the broader discussion on digital ethics and platform governance. This entails dealing with the technological components of cyberbullying and actively involving oneself in the socio-cultural aspects of digital engagement.

5.3. Study Limitations and Future Directions

This study’s applicability to all cyberbullying cases may be limited due to the unique contexts affecting individuals’ experiences. While focused on cyberbullying among college students on social media through the Communicative Action Theory framework, future research should further examine strategies for creating a secure online environment and the roles of educational institutions and social media platforms in mitigating cyberbullying. Additionally, exploring the influence of bystanders and the broader online community could enhance our understanding of cyberbullying dynamics.

This groundwork paves the way for future inquiries into the nuances of cyberbullying within the digital landscape, emphasizing the importance of digital etiquette education, victim support, and legislative measures for a safer online environment. Harnessing the potential of online platforms for constructive discourse is essential for fostering inclusive and healthy digital communities.