The Relation Between Bullying and Cyberbullying, Emotional Intelligence, and Empathy in Portuguese Adolescents

Abstract

Bullying and cyberbullying share in their definition the pattern of aggressive and intentional conduct, characterized by a perceived or observed imbalance and repeated over time. Empathy and emotional intelligence are fundamental individual skills that can enable adolescents to develop less aggressive and more empathetic behaviors. The aim of this study is to explore the frequency of bullying and cyberbullying (victimization and/or aggression), the association of bullying, cyberbullying, empathy, and emotional intelligence, and to explore whether sex, age, educational level, and online recreational time are related to bullying and cyberbullying behaviors. A sample of 599 adolescents, aged between 13 and 21 years old, from basic and secondary schools located in the north of Portugal, participated in the study. The instruments used in the study were the Sociodemographic Questionnaire; the Bullying and Cyberbullying Behavior Questionnaire; the Basic Empathy Scale; and the Self-Perception of Emotional Intelligence Questionnaire. The main results indicated that victims demonstrate greater empathy and emotional intelligence. The results also showed that male adolescents take on more of the role of aggressors in bullying and cyberbullying, while female adolescents take on more of the role of victims. The study underscores the importance of developing and implementing prevention and intervention programs for bullying and cyberbullying, with a focus on promoting empathy and emotional intelligence among adolescents.

1. Introduction

The school environment should provide for the full development of the student, meaningful learning, and social interactions, but over the years, this context has also become a place where violent behavior between peers manifests itself, namely bullying and cyberbullying [1]. Since the end of the 90s, bullying has been extensively studied, and more recently, with the frequent use of the internet and social networks and with the proliferation of electronic devices, such as mobile phones, a new form of bullying has emerged, cyberbullying [2].

There is currently a consensus that schools are institutions where a significant number of students feel threatened (or harassed) by their peers, which justifies a progressive increase in concern about the problems of peer aggression, both in the school environment and in the virtual environment [3,4,5,6]. This concern is evidenced in the investigation, which specifically identifies bullying as one of the main health problems in adolescence, affirming the need to protect adolescents, especially in the online environment [6].

Bullying is a phenomenon that can be defined as aggressive and intentional behavior, carried out by peers, individually or in groups, with the aim of causing harm or discomfort at a physical, psychological, and social level [5,7,8,9]. This involves an imbalance of power, observed or perceived, and is repeated over time [5,10]. It is based on a relationship of domination–submission, considering as victims adolescents who are repeatedly abused and aggressors those who perpetrate aggressive behaviors towards the victims [11].

There are numerous typologies of traditional forms of bullying, such as physical bullying (e.g., punches, kicks, and physical threats), relational bullying (e.g., spreading rumors and social exclusion), verbal bullying (e.g., insulting, offending) [12], and cyberbullying (e.g., sending offensive messages or images through the use of information and communication technologies) [13,14,15].

Cyberbullying is a form of aggression in an online context, defined by a dynamic interaction characterized by the repetition of offensive and abusive behaviors over time using electronic devices [13,14], becoming problematic, especially among adolescents [16].

Both bullying and cyberbullying are aggressive behaviors whose purpose is to harm another person, which certainly refers to violent social behavior [5]. However, research shows that when we compare the two phenomena, the negative impacts of cyberbullying tend to be more severe, translating into higher absenteeism and poor school performance, proving to be a strong predictor of negative effects on adolescent health and well-being [2,10,14,17]. Several studies [3,7,12,18] investigated the factors that increase the risk of bullying and cyberbullying, as well as those that offer protection and prevent these behaviors. For example, in the systematic review carried out by Zych et al. [12], the authors observed that the perpetration of cyberbullying was associated with low levels of empathy. In addition, with regard to protective factors against bullying and cyberbullying, these were related to self-oriented personal skills, moderate use of technology, good academic performance and other-oriented social skills [12].

The General Model of Aggression can help us to understand the occurrence of bullying and cyberbullying since, according to Kokkinos and Antoniadou [19], it “proposes the contribution of person factors such as gender, personality traits, psychological states, technology use, and situational factors such as provocation (e.g., cyber-victimization) and perceived opportunities to act aggressively (e.g., online disinhibition effect)” (p. 60).

Empathy, as a personal factor, has been highlighted as one of the most important elements in explaining bullying and cyberbullying [12,18], along with the school environment, the level of control over personal information online, and abusive use of the internet [3,20]. Thus, empathy gains particular importance in this theme, especially because if it is adequate or high, it constitutes an important skill for social interactions, as it helps adolescents to understand and regulate their social behavior, to express their feelings, and to develop and maintain positive relationships [21,22]

Empathy is a complex, multidimensional concept defined as the ability to understand and share the emotional state of another person [23]). Most authors agree on the definition of empathy in two dimensions: cognitive empathy, which is the ability to recognize and understand the emotions of others, and affective empathy, which refers to the ability to feel and share the emotional experience of others [12,21,24,25]. It is important to mention that the ability to react empathically is positively related to prosocial behaviors and negatively related to bullying [24], that is, lower levels of empathy increase the probability of developing violent or aggressive behaviors [12,21,26], while higher levels of empathy are associated with defending victimized peers [26,27,28]. Empathy is considered one of the personality traits that most influence the prevention of involvement in bullying in the role of aggressor [29]. The ability to perceive the other’s state of mind, to understand the victim’s feelings, and to understand the potential impact that abusive and unjustified behavior may have on the victim are equally important elements [7].

In recent years there have been studies using innovative methodologies that investigate the relationship between cyberbullying and empathy [26,30,31,32]. Studies highlight a pattern like that found in bullying, where higher levels of affective and cognitive empathy predict lower levels of cyberbullying [12]. In turn, teens who have difficulty adopting a perspective may be at greater risk of cyberbullying [12]. Additionally, high levels of empathy have been shown to be associated with less aggressive behaviors and more prosocial behaviors, most likely because they are associated with a greater ability to regulate one’s emotions [29]. Thus, empathy is fundamental to preventing cyberbullying and promoting prosocial behaviors [25].

Empathy is linked to emotional intelligence since empathy is an essential skill for understanding and using emotions, both your own and those of others. The literature has reported that emotional intelligence plays an important role in favoring coexistence in the school context [9,10]. Emotional intelligence has also received increasing attention in research, and it is also considered a protective variable in bullying and cyberbullying practices [9,33].

Mayer and Salovey [34], responsible for developing the first theoretical model about emotional intelligence, define the construct as the individual’s ability to perceive, understand, use, and manage emotions in an appropriate and adaptive way, in themselves and others, as well as the ability to use emotional resources to deal with challenging situations. Emotional intelligence involves a series of domains of intra- and interpersonal skills such as emotional stability, stress management, adaptability, decision-making, self-awareness, empathy, and conflict resolution [34,35].

The occurrence of bullying and cyberbullying seems to have a negative impact on the emotional intelligence of victims [33,36]. Among the consequences of bullying, the lack of self-esteem and the increase in aggressiveness caused by a low level of emotional intelligence have been highlighted, which perpetuates the cycle of violence and makes adolescents more likely to be victims [37]. Thus, adolescents have difficulties in relating to others due to their low ability to manage what they feel and their difficulty in asking for help [38].

A recent study of Rueda et al. [9] showed that there is a higher level of emotional perception in adolescents involved in bullying (both aggressor and victim) and a lower level of understanding and emotional regulation compared to adolescents who are not involved in these behaviors. Adolescents who demonstrate greater emotional intelligence are more prepared to face challenging situations, such as intimidation and threats. These adolescents are able to reduce these negative situations and deal with them successfully, resulting in less victimization and a lower incidence of psychological maladjustments [33,36].

The existing theoretical framework revealed that empathy and emotional intelligence are not the only factors associated with bullying and cyberbullying, and several authors evaluated the influence of other variables on these dynamics, namely age, sex, level of education [6,10,17,39], and online recreational time [6,40,41]. In addition, studies on the prevalence of bullying and cyberbullying have also evaluated the influence of sociodemographic variables between different age groups and education levels [4,11,17].

It is also important to mention that in bullying, male adolescents more often assume the role of victim/aggressor than female adolescents, which is justified by low self-esteem and little assertiveness [11], in almost all typologies except the relational bullying [4,6]. As regards the cyberbullying, in general, the existing literature suggests that it is strongly influenced by sex [4,39] but also by age, and that there are more cases of cyberbullying in female adolescents in early adolescence and males in late adolescence [40].

Finally, the influence of online recreational time also seems to be a relevant factor. Adolescents use new technologies to stay in constant contact with their peers [40]. However, with the recreational time spent online and the public exposure of information, they make us more vulnerable to interpersonal intrusion [3,42], which can culminate in certain forms of victimization [41].

Adolescence is characterized by a set of changes in cognition, emotions, and interpersonal relationships, which implies greater vulnerability and difficulty in regulating emotions [43]. It is a fundamental stage for the development of psychological resources that influence the way adolescents interact with others, and aggression is one of the most prevalent behavioral problems in adolescence, which has increased in recent years [6,44]. In adolescence, the search and construction of identity play a relevant role in the use of the internet, and adolescents seek immediate satisfaction through virtual communication, which often causes them to lose control over what they share on the internet, which can involve risks [41]. Some authors suggest that a lack of inhibition and social engagement contribute to online aggression, i.e., online bullies are more likely to engage due to anonymity, the likelihood of large audiences, rapid spread, and impact on victims provided by the online context, as they do not deal with the immediate emotional or psychological effects of their victims because they are not physically present but connected through technology, which facilitates cyberbullying [16,45,46]. In addition, the anonymity involved in virtual communication allows the dissemination and sharing of information to peers or strangers [40,41,42]. Therefore, there is an association between problematic use of the internet, adherence to social networks, and the perception of privacy on the internet [41,42].

Thus, cyberbullying is more likely among teens who share a technological culture, and many authors [40,41,42,45] have shown in their research that the problematic use of the internet has implications in the different manifestations of school violence (bullying and cyberbullying) both in the role of victim and aggressor.

The themes explored in this study are currently a cause of growing concern among the scientific community, highlighting the need for more studies that allow a greater knowledge and understanding of the variables that possibly explain bullying and cyberbullying. Although there are studies that suggest that the behavior of adolescents in bullying predicts similar behaviors in cyberbullying, that is, the factors associated with the phenomenon of bullying are potentially associated with cyberbullying, suggesting that the internet, computers, and smartphones are just new tools used to facilitate the perpetration of certain pre-existing behaviors [3,12,20]. However, there are few studies investigating not only these phenomena simultaneously but also emotional intelligence and empathy contributing to the increase in both the knowledge of aggressive behavior dynamics and effective strategies for identification, prevention, and intervention programs.

In this way, we highlight the importance of studying bullying and cyberbullying together, but also the importance of relating them to variables such as emotional intelligence and empathy and their dimensions. Thus, the present research aims to: (1) know the frequency of bullying and cyberbullying in a sample of adolescents; (2) analyze the relationship between bullying, cyberbullying, and empathy; (3) study the relationship between bullying, cyberbullying, and emotional intelligence; (4) understand if there are differences between bullying practices, cyberbullying, and the sociodemographic variables (sex, age, and level of education) of the participants; and (5) study the relationship between cyberbullying and online recreational time.

2. Materials and Methods

2.1. Study Design

The present research is quantitative and cross-sectional, considering that the objective is to quantify phenomena through statistical procedures and the variables were evaluated in a single moment. This is also a correlational study, since it is intended to understand the association between the variables studied [47].

2.2. Participants

A total of 599 Portuguese adolescents participated in the present study, of which 284 (47.4%) were females and 315 (52.6%) were males, aged between 13 and 21 years (M = 15.17, SD = 1.57). Of the 599 participants, 233 (38.9%) were in basic school and 366 (61.1%) were in secondary school, in schools in urban areas in the north of Portugal. Of the 599 participants, 577 (96.3%) said they used the internet, with smartphones being the most used type of equipment, 563 (94%), and 545 (91%) participants reported that the most frequent use of the internet is social networks. The participants and the four schools were selected using the convenience method. Only information from students who completed all instruments was included.

2.3. Instruments

Sociodemographic Questionnaire. The sociodemographic questionnaire was prepared by the authors, covering questions such as age, sex, and level of education, and also some questions related to internet usage habits and online recreational time, namely: “What type of equipment do you use the most to connect to the internet and/or social networks?”; “Which social network do you use the most?”; “How many hours a day do you spend on the internet and/or social networks?”.

Bullying and Cyberbullying Behaviors Questionnaire. The bullying behaviors and cyberbullying questionnaire (BCBQ) [4]), validated for the Portuguese population, is an adaptation of the victimization and bullying scales of the Revised Olweus Bully/Victim Questionnaire (OBVQ) [48]. The main objective of this questionnaire is to assess the prevalence of bullying and victimization behaviors in secondary school students and consists of a self-report measure, consisting of 36 items. The short version used in this study consists of 20 items, organized into two subscales: victimization (e.g., “They called me names, made fun of me, offended me, threatened me or provoked me”; “They posted videos or photos of me on the Internet without my authorization”) and aggression (e.g., “I hit, kicked or violently pushed another person”; “I have sent rude, threatening or distasteful messages by email, SMS or electronic messages to another person”) on a five-point scale, where 1 means “it never happened to me” and 5 means “several times a week” [5]. The questionnaire showed acceptable psychometric properties and revealed adequate internal consistency in both subscales, with a Cronbach α of 0.79 in the victimization subscale and 0.82 in the bullying subscale [5]. The Cronbach’s α values of the instrument in this study are 0.74 for the victimization subscale and 0.65 for the aggression subscale. Regarding the quality of fit, the factorial model revealed poor values (χ²/df = 5.085, CFI = 0.771, GFI = 0.901, NFI = 0.733; RMSEA = 0.071, P[rmsea ≤ 0.05] < 0.001) [47]. Almost all items have high factor weights (λ > 0.5) and high individual reliability (R² > 0.25).

The short version of the Basic Empathy Scale (BES-A). For empathy, the basic empathy scale, short version, was used, validated for the Portuguese population by Pechorro et al. [23]. The original version of this scale, the Basic Empathy Scale (Basic Empathy Scale—BES) [49], is a self-report measure consisting of 20 items, designed to measure two dimensions of empathy in adolescents: affective empathy and cognitive empathy. The short version (BES-A) used in this study consists of seven items in a five-point Likert-like format from 1 “strongly disagree” to 5 “strongly agree” in a two-dimensional structure: affective empathy (e.g., “After talking to a friend who is sad, I usually get sad too”) and cognitive empathy (e.g., “I usually notice how people feel even before they tell me”) [23]. The basic empathy scale, brief version, was originally validated in adolescents and has been shown to have good reliability and validity, revealing values from acceptable to good, always above 0.70 [23]. In the present study, Cronbach’s α values are 0.75 for the total scale, 0.83 for the affective empathy dimension, and 0.82 for cognitive empathy. With regard to the quality of adjustment, this factorial model presented adequate values (χ²/df = 3.975, CFI = 0.976, GFI = 0.976, NFI = 0.968; RMSEA = 0.071, P[rmsea ≤ 0.05] = 0.026) [47]). All items have high factor weights (λ > 0.65) and high individual reliability (R² > 0.44).

Emotional Intelligence Self-Perception Questionnaire (QIE-AP) [50]. The QIE-AP consists of 18 items and four subscales consistent with the emotional intelligence model of Mayer and Salovey [34]: (a) Emotional perception, evaluation, and expression, which aims to assess a person’s ability to describe and distinguish his or her emotions from those of other people. The subscale consists of four items (e.g., “Through the tone of voice and gestures of others, I can tell if they are sad or angry”). (b) Emotional facilitation of thought aims to understand whether subjects support their information and decision process on the emotions felt. This subscale consists of five items (e.g., “I’m glad to think about the good things I have”). (c) Emotional understanding and analysis. This dimension is oriented towards the ease with which the person associates’ emotions with specific situations and people. The subscale is assessed by six items (e.g., “The loss of someone I care about makes me sad”). (d) Emotional regulation, related to the person’s ability to control their posture, expression, and behavioral decision in the face of positive or negative emotions, leading to the expansion of pleasurable situations and the avoidance of situations that trigger negative emotions, without harming personal and social success, always connoting learning and emotional growth favoring their well-being, as well as that of others. The emotion regulation subscale is assessed by three items (e.g., “I try to do what gives me the most pleasure”). All items are answered by the participants using a five-point Likert scale (1—“strongly disagree” to 5—“strongly agree”).

The Emotional Intelligence Self-Perception Questionnaire Portuguese version [50] revealed Cronbach’s α values between 0.70 for the subscale of perception, evaluation, and emotional expression and 0.77 for the subscale of emotional regulation.

In the present study, the Cronbach’s α value is 0.74 for the total scale. As for the dimensions, the α value varies between 0.68 (emotional understanding and analysis) and 0.50 (emotional perception, evaluation, and expression). The factorial model of this Emotional Intelligence Scale revealed an acceptable quality of adjustment (χ²/df = 3.683, CFI = 0.817, GFI = 0.922, NFI = 0.769; RMSEA = 0.067, P[rmsea ≤ 0.05] < 0.001) [47]. All items except two have high factor weights (λ > 0.5) and high individual reliability (R² > 0.25).

2.4. Procedures

The research project was subject to approval by the university’s Ethics Committee (Doc. 78-CE-UTAD-2022) and submitted for approval by the General Directorate of Education of the Ministry of Education of Portugal through the School Survey Monitoring System. Free and informed consent was also requested from the parents of all participants under 18 years of age. All the participants have given their informed consent to participate. In addition, the principles of voluntary participation and confidentiality of the responses collected were explained to the participants and their parents. All participants filled out the paper questionnaire in a classroom with the presence of a researcher. Before administering the questionnaire, the researcher made a brief approach to the theme under study and clarified any doubts. The researcher was present in the classroom until the filling out of the research questionnaire was completed. This monitoring in the classroom was carried out jointly with the head teacher of the subject and/or class director to ensure that the young participants understand the instruction to fill out the questionnaires.

2.5. Data Analytics Strategy

Initially, descriptive statistics, means (M), and standard deviations (SD), were performed in relation to the variables under study. To verify the assumption of the normality of the data, the skewness and kurtosis coefficients were computed. Through the analysis of these values, it was found that it was appropriate to perform non-parametric tests.

To evaluate the factorial structure of the instruments used in the investigation, the methodology of structural equation analysis (Structural Equation Modeling—SEM) was used. To estimate the unknown parameters of the different trajectories of the instruments, the maximum likelihood method was used. The overall quality of fit was made according to the following indices: ratio chi square statistics/degrees of freedom (χ²/df), comparative fit index (CFI), goodness of fit index (GFI), normed fit index (NFI), root mean square error of approximation (RMSEA), and P[rmsea ≤ 0.05] [47,51,52]. The quality of the local adjustment was estimated by the factor weights and the individual reliability of the items. Cronbach’s α was used to evaluate the reliability of each of the scales.

Subsequently, the frequency of bullying and cyberbullying was analyzed according to the sex of the participants. The association between victimization and aggression by bullying and cyberbullying, emotional intelligence, and empathy was evaluated using Spearman’s correlation coefficient, and according to Cohen [53], a correlation value of 0.1 indicates a weak association, a value of 0.3 a moderate association, and a correlation value equal to or greater than 0.5 is indicative of a strong association.

In order to verify whether sociodemographic variables (sex, age, and level of education) significantly influence victimization and aggression from bullying and cyberbullying, the Wilcoxon-Mann–Whitney test was used.

Then, to verify whether there were differences in cyberbullying in relation to the recreational time spent online, the Kruskal–Wallis test was used, followed whenever possible by post hoc tests.

The statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) (version 29.0) for Windows and Analysis of Moment Structures (AMOS) (version 29.0). In all statistical analyses, significance values of 5% were considered.

3. Results

This section begins with the presentation of the values of the means (M), standard deviation (SD), skewness, and kurtosis for the variables studied: victimization and the aggression of bullying and cyberbullying, total empathy (TE) and its dimensions: affective empathy (AE) and cognitive empathy (CE), and emotional intelligence (EI) and its dimensions: perception, evaluation, and emotional expression (PE), emotional understanding and analysis (EU), emotional regulation (ER), and emotional thought facilitation (EF) (

Table 1. Descriptive measures and univariate normality.

Variable	Min.	Max.	M	SD	Sk	Ku
BV	6	25	8.963	3.203	1.886	4.375
CV	2	8	2.349	0.747	3.021	12.208
BA	6	16	7.015	1.663	2.349	6.393
CA	2	7	2.118	0.471	5.671	40.638
TE	8	35	24.838	4.052	0.098	0.792
AE	3	15	8.736	2.870	0.039	−0.392
CE	4	20	16.102	2.381	−0.379	0.981
EI	42	90	75.240	6.748	−0.606	1.446
PE	8	20	17.212	1.931	−0.688	1.032
CE	10	30	25.509	2.913	−0.726	1.523
RE	3	15	12.447	1.876	−0.894	1.897
EF	11	25	21.072	2.499	−0.567	0.284

Note: BV (Bullying Victimization), CV (Cyberbullying Victimization), BA (Bullying Aggression), CA (Cyberbullying Aggression), TE (Total Empathy), AE (Affective Empathy), CE (Cognitive Empathy), EI (Emotional Intelligence), PE (Perception, Evaluation, and Emotional Expression), CE (Comprehension and Emotional Analysis), ER (Emotional Regulation), EF (Emotional Facilitation of Thought).

).

3.1. Frequency of Bullying and Cyberbullying

In what concerns the frequency of victimization in females, of the 284 participants, 243 reported having been victims of at least one bullying behavior (≈85.56%) and 76 (≈26.76%) of cyberbullying. Regarding the 315 male participants, 236 (≈74.92%) reported having been victims of at least one bullying behavior and 70 (≈22.22%) of cyberbullying.

The victims reported that the behavior they suffered the most was “They called me names, made fun of me, offended me, threatened me or provoked me” (item 1). Of the 599 participants, 323 (≈53.92%) reported having suffered from this behavior, of which 222 (≈68.7%) stated that they had been victims once or twice during the year, 42 (≈13%) two to three times a month, 22 (≈6.8%) once a week, and 37 (≈11.4%) several times a week.

Regarding aggressive behavior, the most scored item was: “I called names, threatened, made fun of or provoked another person” (item 9). Of the 599 participants, 183 (≈30.55%) reported having perpetrated this behavior, of which 139 (≈75.96%) stated that they had practiced this behavior once or twice during the year, 26 (≈14.21%) two to three times a month, 7 (≈3.8%) once a week, and 11 (≈6%) several times a week.

3.2. Correlational Analysis: Bullying, Cyberbullying, Empathy, and Emotional Intelligence

In order to explore the association between victimization and perpetration of bullying and cyberbullying, total empathy and its dimensions (affective empathy and cognitive empathy) and emotional intelligence and its dimensions (emotional perception, evaluation, and expression, emotional understanding and analysis, emotional regulation, and emotional thought facilitation) were used, and a correlation coefficient was used (

Table 2. Association between bullying, cyberbullying, empathy, and emotional intelligence.

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.
1. BV	1	0.447 **	0.424 **	0.200 **	0.151 **	0.150 **	0.067	0.075	0.172 **	0.126 **	−0.099 *	0.011
2. CV		1	0.329 **	0.416 **	0.081 *	0.069	0.043	−0.019	0.106 **	0.025	−0.104 *	−0.084 *
3. BA			1	0.383 **	−0.003	0.014	−0.033	−0.008	0.109 **	0.021	0.105 **	−0.058
4. CA				1	0.045	0.052	−0.003	−0.046	0.011	−0.050	−0.041	−0.060
5. TE					1	0.795 **	0.679 **	0.424 **	0.315 **	0.344 **	0.269 **	0.269 **
6. AE						1	0.150 **	0.232 **	0.094 *	0.261 **	0.105 **	0.123 **
7. CE							1	0.421 **	0.443 **	0.257 **	0.330 **	0.274 **
8. EI								1	0.616 **	0.811 **	0.711 **	0.683 **
9. PE									1	0.397 **	0.352 **	0.218 **
10. CE										1	0.477 **	0.362 **
11. ER											1	0.248 **
12. EF												1

Note: * p < 0.05; ** p < 0.001. BV (Bullying Victimization), CV (Cyberbullying Victimization), BA (Bullying Aggression), CA (Cyberbullying Aggression), TE (Total Empathy), AE (Affective Empathy), CE (Cognitive Empathy), EI (Emotional Intelligence), PE (Perception, Evaluation, and Emotional Expression), CE (Comprehension and Emotional Analysis), ER (Emotional Regulation), EF (Emotional Facilitation of Thought).

).

The results obtained in Table 2 indicate positive and significant correlations between bullying victimization and total empathy (r = 0.151, p < 0.001) and affective empathy (r = 0.150, p < 0.001), which indicates that adolescents who report greater victimization in bullying tend to have more total and affective empathy. In victimization in cyberbullying, there is a positive, although weak, correlation with total empathy (r = 0.081, p < 0.001).

Regarding the emotional intelligence variable and its dimensions, the results show positive and significant correlations between bullying victimization and emotional perception, evaluation, and expression (r = 0.172, p < 0.001) and emotional understanding and analysis (r = 0.126, p < 0.001). The results also indicate a negative and significant correlation with emotional regulation (r = −0.099, p < 0.001), i.e., the greater the victimization in bullying, the lower the ability to regulate oneself emotionally.

The results point to significant correlations between victimization in cyberbullying and the various dimensions of emotional intelligence, namely a positive and significant correlation with emotional perception, evaluation, and expression (r = 0.106, p < 0.001), and negative and significant correlations with emotional regulation (r = −0.104, p < 0.001) and with emotional facilitation of thought (r = −0.084, p < 0.001).

In bullying aggression, the results obtained demonstrate positive and significant correlations with emotional perception, evaluation, and expression (r = 0.109, p < 0.001) and with emotional regulation (r = 0.105, p < 0.001).

3.3. Comparative Differential Analysis: Bullying, Cyberbullying, and Sex

To verify whether sex significantly influences bullying victimization and aggression (BV; BA) and cyberbullying (CV; CA), the non-parametric Wilcoxon-Mann–Whitney test was used (

Table 3. Comparative differential analysis of factors of victimization and aggression relative to sex.

Variable	Male (N = 315) Mean Rank	Female (N = 284) Mean Rank	Z	p
BV	280.52	321.61	−2.938	0.003
CV	294.43	306.18	−1.107	0.268
BA	329.61	267.16	−4.889	0.000
CA	306.63	292.65	−2.059	0.039

Note: BV (Bullying Victimization), CV (Cyberbullying Victimization), BA (Bullying Aggression), CA (Cyberbullying Aggression).

). According to the results obtained, it is verified that there are significant differences between sexes with regard to bullying in victimization. Furthermore, by analyzing the means of the orders, we can affirm that females have higher scores than males. Regarding aggression, significant differences are found in bullying and cyberbullying practices, with male subjects having higher scores.

3.4. Comparative Differential Analysis: Bullying, Cyberbullying, and Age

To verify whether age, grouped into two classes [13, 15] and [16, 21], has a statistically significant effect on bullying and cyberbullying, the non-parametric Wilcoxon-Mann–Whitney test was used, since the conditions for the applicability of the parametric test are not met (

Table 4. Comparative differential analysis of victimization and aggression in relation to age.

Variable	[13, 15] Mean Rank	[16, 21] Mean Rank	Z	p
BV	298.98	300.65	−0.116	0.907
CV	287.22	308.14	−1.924	0.054
BA	292.99	304.46	−0.876	0.381
CA	296.94	301.95	−0.719	0.472

Note: BV (Bullying Victimization), CV (Cyberbullying Victimization), BA (Bullying Aggression), CA (Cyberbullying Aggression).

). The results obtained demonstrate the existence of statistically significant differences in victimization in cyberbullying regarding age, with adolescents aged between 16 and 21 being the ones with the highest values in all dimensions under study.

3.5. Comparative Differential Analysis: Bullying, Cyberbullying, and Education Level

To verify whether the educational level (basic or secondary) significantly influences victimization and aggression in bullying and cyberbullying, the non-parametric Wilcoxon-Mann–Whitney test was used (

Table 5. Comparative differential analysis of victimization and aggression in relation to educational level.

Variable	Basic School Mean Rank	Secondary School Mean Rank	Z	p
BV	288.29	317.51	−2.050	0.040
CV	289.45	315.79	−2.435	0.015
BA	288.94	316.54	−2.121	0.034
CA	295.91	306.12	−1.476	0.140

Note: BV (Bullying Victimization), CV (Cyberbullying Victimization), BA (Bullying Aggression), CA (Cyberbullying Aggression).

).

The results showed that there are statistically significant differences in bullying and cyberbullying victimization regarding the educational level, with adolescents at the secondary school being those with considerably higher values. Regarding aggression, it is verified that both bullying and cyberbullying were higher in adolescents in secondary school; however, it was in bullying aggression that the differences were more significant.

3.6. Comparative Differential Analysis: Cyberbullying and Online Recreation Time

In order to verify whether there are statistically significant differences in cyberbullying in relation to the online recreational time per day in the use of the internet, grouped into the following categories: up to one hour, more than one and up to three hours, between four and six hours, and seven or more hours, the non-parametric Kruskal–Wallis test was used, followed by the post hoc test (

Table 6. Comparative differential analysis of victimization and perpetration in cyberbullying regarding online recreational time.

Variable	Up to 1 h Mean Rank	1 to 3 h Mean Rank	4 to 6 h Mean Rank	7 or More Hours Mean Rank	${\mathit{\chi }}_{\mathit{k}\mathit{w}}^2$	p
CV	280.71	279.09	314.60	353.21	20.970	<0.001
CA	292.93	295.44	304.02	310.10	4.551	0.443

Note: CV (Cyberbullying Victimization), CA (Cyberbullying Aggression).

).

The results indicate that online recreational time has a statistically significant effect on cyberbullying victimization. According to the results of pairwise comparisons tests, the class of 1 to 3 h is significantly different from the classes 4 to 6 h (p = 0.013) and 7 or more hours (p < 0.001). Moreover, individuals who spend more recreational time online are more often victims of cyberbullying.

4. Discussion

The present research aimed to know the frequency of bullying and cyberbullying in adolescents, to analyze the relationship between bullying, cyberbullying, and empathy, to study the relationship between bullying, cyberbullying, and emotional intelligence, to understand if there are differences between bullying practices, cyberbullying, and emotional intelligence and the sociodemographic variables (sex, age, and level of education) of the participants, and finally to study the relationship between cyberbullying and online recreational time.

The first objective was to explore the frequency of bullying and cyberbullying in adolescents. The results showed that most participants reported having been victims of at least one behavior during the year in both phenomena. However, the prevalence of victimization in bullying is significantly higher than victimization in cyberbullying, for both females and males. With regard to aggressive behaviors, approximately one third of the participants reported having practiced this type of behavior and stated that they did so between once or twice during the year. These results are in line with those obtained by other authors, where it is evidenced that there are more adolescents involved in bullying than in cyberbullying [4] and that in both phenomena, the most common role is that of the victim [11,39]. We can conjecture that these data come from social desirability, but also because it is easier to assume that one has been a victim of bullying or cyberbullying than to recognize that the role of aggressor is perpetuated. Concerning the types of bullying, the most frequent behavior was verbal bullying, manifested through offenses, provocations, and threats, such as “They called me names, made fun of me, offended me, threatened me or provoked me” (item 1). The data obtained corroborate the conclusions of a study carried out by Coelho et al. [4] in a Portuguese context, which also highlighted that verbal bullying behaviors were the most frequently reported.

Concerning the relationship between bullying, cyberbullying, and empathy and its dimensions, we observed that empathy has a positive relationship with victimization in both phenomena, possibly as a protective factor, although this relationship is stronger in victimization and bullying. Our results showed that adolescents who report greater victimization in bullying tend to demonstrate more total and affective empathy, while in cyberbullying only total empathy has some relevance, although it is less significant. Our results are corroborated in the literature by several authors [7,12,22], which suggest that empathy can be an adaptive response to negative experiences, promoting a deeper understanding of the emotions and needs of others.

With reference to the relationship between bullying, cyberbullying, and emotional intelligence and its dimensions, the results showed that victimization in both phenomena is related to different dimensions of emotional intelligence, suggesting that the greater the experience of victimization, the greater the ability to perceive, evaluate, and express emotions, and that victims may develop a better ability to understand their own emotions and those of others as a way to cope with challenging social interactions [9]). The results are in line with other authors [33,36,38,41], who add that when adolescents manage to develop emotional self-regulation skills, they tend to have fewer negative emotions related to the expression of aggression or anger, and these harmful behaviors tend to decrease.

Regarding victimization in cyberbullying, and specifically the dimensions of emotional regulation and emotional facilitation of thought, the results indicate negative and significant associations. This result can be explained by the fact that victims may face difficulties in regulating and guiding their thinking when exposed to online victimization, which can hinder their ability to deal with cyberbullying in a constructive way. The results found are in line with the studies carried out by Peña-Casares and Aguaded-Ramirez [37] and Arévalo et al. [10], who add that a low level of emotional intelligence makes adolescents more likely to be victims, contributing to feelings of isolation, low self-esteem, and emotional dysregulation.

As for aggression in bullying, the results indicated positive associations with the dimensions of emotional perception, evaluation, and expression and with emotional regulation. These results were not expected; however, we can conjecture that the aggressors may have a developed emotional sensitivity [33]. This allows them, on the one hand, to understand the emotions of others, namely adolescents who appear to be more vulnerable, and on the other hand, they can be more competent in regulating and controlling their thoughts related to emotions, which can influence their ability to act aggressively, intentionally, or as planned.

The results showed that female adolescents have higher scores regarding victimization. Taking into account aggression, both in bullying and cyberbullying, we found that male adolescents have higher scores. However, it is in aggression bullying that the difference is more significant, demonstrating that male adolescents assume more of the role of aggressors. These results are corroborated by Coelho and Sousa [5], while other studies find no differences in sex in victimization [4,6]. With regard to cyberbullying, other studies show that male adolescents take on the role of cyberaggressors more and that female adolescents are more cybervictims [4,10,39], which is in line with our results. Although the results of Calmastra et al. [39], demonstrate that there are differences between male and female adolescents in involvement in cyberbullying, they report a low association between the variables. In turn, Feijóo et al. [6]) show that sex does not seem to have an influence on cyberbullying behaviors.

Taking into account the differences between bullying and cyberbullying practices in terms of age, our results indicate that older adolescents are the ones who assume more of the role of victim, both in bullying and cyberbullying. In the empirical literature, there is no unanimity, since many authors report that cyberbullying, in general, is perpetrated by older adolescents [4,45]) or in late adolescence [40], while others report not having found differences in cyberbullying with regard to age [17].

Regarding the objective of understanding whether the level of education influences bullying and cyberbullying behaviors, we found that adolescents at the secondary level are those who present considerably higher values in both phenomena, both in aggression and victimization; however, it is in aggression and bullying that the differences are more evident. Our results are in line with studies of Calmaestra et al. [39], which indicate that adolescents in the higher school years are more involved in cyberbullying and that the longer the school year, the greater the involvement in cyberbullying, based on the mastery of new technologies and the need to affirm and build identity. It is also important to mention that the results found regarding bullying are in line with those of Carvalho et al. [17], who studied a sample of Portuguese students, but with regard to cyberbullying, the authors did not find significant associations with schooling.

Finally, with regard to the objective of studying whether cyberbullying is related to online recreational time, it was evidenced that there is a relationship between victimization in cyberbullying and recreational time spent online, even more so when adolescents spend more hours online. Other studies confirm that problematic use of technology has been associated with greater involvement in the different manifestations of school violence [41], and that teens who are more addicted to using the internet are more likely to become online victims [40], which corroborates our results. Other authors [42,45]; also mentioned that young people involved in cyberbullying tend to spend more time online, have greater use of social networks, have various online risk behaviors, and have greater problematic use of the internet.

Bullying and cyberbullying share in their definition the pattern of aggressive and intentional behavior, perpetrated by peers, individually or in groups, characterized by a perceived or observed imbalance, and repeated over time with the intention of causing harm or discomfort at a physical, psychological, and social level [5,13,14]. However, bullying and cyberbullying are expressed differently, and this can have an influence in terms of signaling and impact. Cyberbullying can reach a much larger audience and be harder to control; therefore, it can have a devastating psychological impact on the victim, as it is more difficult to deal with and resolve, sometimes due to the anonymity of the aggressor [14]. Traditional bullying is usually more linked to a school environment, and the fact that our study was conducted in schools may have influenced a more conscious awareness of the bullying problem compared to cyberbullying. However, cyberbullying presents new challenges and nuances in relation to traditional bullying, requiring a more complex and comprehensive approach from institutions, governments, and society in general. In this sense, in future studies, it is important to reflect and discuss the differences and implications of these types of violence in order to find effective ways to prevent and combat this type of violence.

5. Contributions, Limitations, and Future Research

The importance of this study is related to the scarcity of scientific research that jointly addresses the phenomena of bullying and cyberbullying and that relates them to empathy and emotional intelligence, namely in the Portuguese context. Promoting empathy is an essential value in relationships with others, and schools have a role in the development of these individual variables. Regarding this, developing programs focused on empathy and emotional intelligence in schools, particularly at an early age, is essential, not only aimed at students but also at educational agents. Such initiatives can play a crucial role in fostering socio-emotional skills, which may contribute to reducing bullying and cyberbullying. By teaching students how to understand and manage their own emotions, as well as empathize with others, these programs can encourage a more compassionate and supportive school and family environment. Notwithstanding the relevance of the results found in this research, it is important to mention the limitations of this research, namely, to take them into account in the interpretation of the results.

The fact that the study is cross-sectional, and, therefore, it is not possible to establish a causal relationship between the variables is recognized as a limitation, and future studies, of a longitudinal nature, with a larger and more representative sample, may be more appropriate to verify this causal relationship and add solidity to the results. Another limitation is related to the instruments used in this study; on the one hand, the use of self-report instruments when collecting data may have increased the risk of random, hasty responses or of the content being perceived subjectively, as well as the existence of responses due to social desirability. On the other hand, the instruments that evaluated cyberbullying were composed of a reduced number of items. Finally, it was used for a convenience sample that may not have statistical value for inferring population. For future research, it could be beneficial to use multiple sources and/or other instruments that analyze in depth the specificities and typologies of cyberbullying with regard to the involvement of adolescents.

In the future, it would be pertinent to study both phenomena together and their interaction, not only considering individual characteristics (e.g., assertiveness, personal identity) but also social (e.g., peer dynamics), family (e.g., internet monitoring by parents), and contextual (e.g., school climate) factors.

It should also be noted that the existing literature explains the dimension of victimization in both phenomena; in the case of aggression, and especially with regard to cyberbullying, the studies are scarcer; however, the evidence we found shows that both phenomena are related and that they share similar behaviors [3,12,20]. This evidence leads us to suppose that educational programs aimed at preventing bullying can also play an important role in the prevention of cyberbullying, since the different factors involved, namely individual factors, the context, and the roles (victimization/aggression), are interconnected in both phenomena. Thus, we hope that the data from this study will represent a contribution to the development of prevention programs that contemplate these problems.