School Bullying and Sports Participation Among Adolescents in the Andorran Educational System 

David Escofet Martín

Agnès Ros Morente

Cristina Torrelles-Nadal

Abstract

The present study aimed to examine school bullying (SB) among lower secondary education students in the Andorran educational system, considering sociodemographic variables, the perception of being bullied, and sports participation. A descriptive, comparative, cross-sectional research design was used. The sample consisted of 353 lower secondary education students from the Andorran educational system (49.01% female; M_age = 13.76; SD = 1.22). The Cisneros Self-test questionnaire was used to assess the SB index and its dimensions. Questions were also included to determine gender, educational cycle, perception of being bullied, whether participants practiced sports, and sport type. Descriptive analyses, nonparametric Mann-Whitney U tests, and binary logistic regression were performed. The results showed an SB prevalence of 29.18% and a substantial discrepancy between the perception of being bullied and the score obtained on the Cisneros Self-test, with a high proportion of inaccurate SB identification (26.06%). No significant differences were observed between sports participants and nonparticipants or between sport types, although gender differences were found among sports participants (p = .049; r = .138). The dimensions of contempt and ridicule (OR = 4.761; 95% CI [2.206, 10.275]; p < .001) and theft (OR = 2.356; 95% CI [1.154, 4.809]; p = .019) were also associated with a greater likelihood of inaccurate SB identification, whereas aggression (OR = 0.289; 95% CI [0.116, 0.720]; p = .008) facilitated more accurate identification. These findings underscore the importance of developing comprehensive policies aimed at improving the identification and prevention of SB among young people.

Introduction

School bullying (SB) is a problem that affects approximately 32% of young people worldwide (UNESCO, 2019). Being bullied can have short-, medium-, and long-term consequences (Arseneault, 2017; Sigurdson et al., 2015) in various areas, including psychological, occupational, academic, and physiological domains (Agley et al., 2021; Galán-Arroyo et al., 2023; González-Cabrera & Machimbarrena, 2023). 

SB is defined as repeated aggression over a prolonged period through negative behaviors carried out by one or more peers in a situation involving a power imbalance (Olweus, 1998). Consistent with this definition, UNESCO (2020) states that SB involves intentionality, causes social, emotional, and/or physical consequences for the victim, and occurs in contexts in which the victim has difficulty receiving effective support. 

Several authors have highlighted the limited literature on the relationship between SB and sports participation, as well as the need to further explore the variables that mediate this interaction (Carrasco-López et al., 2024; Escofet-Martín et al., 2025a; Martínez-Baena & Faus-Boscá, 2018). In this context, Tilindienė et al. (2010a) reported that athletes were less likely to be bullied, whereas other authors have described a higher prevalence among sports participants (Rusillo et al., 2024; Zapata-López et al., 2024).

Gender has also been shown to play an important role in this relationship. Several studies have emphasized the need to consider gender in analyses of SB (Martos-García et al., 2024; Rodkin et al., 2015), given the inconsistent findings in the existing empirical evidence. Marracho et al. (2021) and Vveinhardt and Fominiene (2022) reported higher prevalence rates among males than females; however, differences between genders have been reported to be smaller in terms of victimization (Díaz-Aguado et al., 2013). Other studies found no significant gender differences among sports participants (Benítez-Sillero et al., 2022; Escofet-Martín et al., 2025b; Tilindienė et al., 2010b). These contradictory findings justify the need for further research into the relationship between SB, sports participation, and gender.

The scientific literature has examined the incidence of SB in sports contexts by comparing individual and team sports. Marracho et al. (2021) found higher prevalence rates of SB in team sports (26.70%) than in individual sports (19.10%). However, they found no significant differences according to sport type. Vveinhardt and Fominiene (2022), in turn, identified a much higher prevalence of SB in team sports (36.60%) than in individual sports (16.60%), with significant differences between the two groups.

Another variable to consider is the perception of being bullied. Several studies have highlighted that adolescents are familiar with SB situations but do not always identify them as an actual form of violence (Postigo et al., 2019; Rawlings, 2019). They tend to interpret these acts of aggression as mere jokes among peers and, when they do identify them as SB, may ultimately blame the victims for being bullied because of their behavior. 

Difficulty identifying SB behaviors may be related to theories concerning the normalization of violence (Bourdieu, 2000; Galtung, 1990), according to which certain forms of aggression eventually come to be perceived as natural or socially acceptable. Moreover, when adolescents identify these behaviors as SB, they may experience psychological distress arising from the contradiction between their perceptions and the reality they observe. This situation could be explained by the tendency to reduce such distress by reinterpreting or minimizing conflict situations, for example, by viewing SB as mere jokes, which could be interpreted through Cognitive Dissonance Theory (Festinger, 1957).

This phenomenon is amplified by the observation and reproduction of behaviors in the school environment, where students tend to imitate their peers’ behaviors, particularly when these are accepted by the group, in line with the principles of social learning theory (Bandura, 1977). This is compounded by the influence of the family, media, and social contexts, which play an important role in the normalization of violence (Valente et al., 2024). In this regard, educational and sports settings are not isolated spaces but rather function as a social microcosm in which the dynamics, values, and problems present in society are reproduced (Bronfenbrenner, 1979), significantly influencing how adolescents interpret and respond to situations involving violence.

In their study, Connell et al. (2018) found a 34.10% mismatch in the perception of SB. Specifically, this mismatch was observed among victims of SB who did not identify themselves as such. The authors also explained that traditional SB (aggression, insults, pushing, etc.) and relational SB (social exclusion and rumors) are easier for adolescents who experience them to recognize. They also suggested that SB becomes normalized and is viewed as acceptable in situations in which it is perceived that all students experience it.  

In a recent study, Larrañaga et al. (2024) reported an SB prevalenceof 27.90%, while 7% of students perceived themselves as being bullied. These results indicated that 20.90% of students inaccurately identified SB.  They also suggested that verbal and physical SB is more readily identified than relational SB. Another study reported an SB prevalence of 36.70%, while 24.20% of students perceived themselves as being bullied, with a discrepancy of 12.50%. In this case, victims who were able to identify SB were mostly those who experienced threats and social exclusion, whereas those who did not identify it accurately tended to experience direct verbal aggression (Sidera et al., 2020). These findings reveal contradictions regarding the identification of SB and the types of bullying that mediate this interaction. Furthermore, female adolescents have been reported to use more indirect forms of SB (social exclusion and rumor spreading), whereas male adolescents tend to use direct forms (aggression or insults) (Jiménez-Barbero et al., 2012).

Based on the literature reviewed, the overall objective of the study was to examine SB and differences in SB according to sociodemographic variables (educational cycle and gender), perception of being bullied, and sports participation (sports participants/nonparticipants; individual/team sport) among lower secondary education students in the Andorran educational system. The following specific objectives were derived from this overall objective: (1) to describe the overall prevalence of SB and its prevalence according to its dimensions (contempt and ridicule, coercion, restriction of communication, aggression, intimidation and threats, social exclusion and blocking, verbal harassment, and theft); (2) to analyze the degree of agreement between the perception of being bullied and scores on the Cisneros Self-test; (3) to compare whether there were significant differences in the SB index according to gender, sports participation, sport type, and the perception of being bullied; and (4) to identify the dimensions of SB that best predict inaccurate SB identification using a logistic regression model. 

With regard to the second objective, it was hypothesized that there would be a significant discrepancy between the perception of being bullied and the scores obtained on the Cisneros Self-test, with students underestimating SB. Regarding the third objective, it was hypothesized that adolescents in the Andorran educational system who participated in team sports would report a higher prevalence of SB than those who participated in individual sports. Finally, the fourth objective led to the hypothesis that individuals with higher scores on the more explicit dimensions of SB (aggression) would have a greater ability to identify SB. By contrast, the less explicit dimensions (social exclusion) would be less readily identified.

Because this study combines descriptive, comparative, and predictive analyses, some objectives do not have associated hypotheses because they are exploratory or descriptive in nature. 

Methodology

Research Design

This study adopted a quantitative methodological approach, with a descriptive, comparative, cross-sectional design, as data were collected at a single point in time without manipulating the variables. The comparative component involved comparisons among the different subgroups in the sample (gender, educational cycle, sports participation, and perception of being bullied). A predictive analysis was also conducted using binary logistic regression to explore the influence of the factors on the SB index.

Participants 

The sample consisted of 353 students from the Andorran educational system, with a mean age of 13.76 years (SD = 1.22), from the lower secondary schools of Encamp (N = 256; 72.52%), Santa Coloma (N = 59; 16.71%), and Ordino (N = 38; 10.76%). Students from all grades at the three lower secondary schools in the Andorran educational system participated, with their legal guardians authorizing their participation through informed consent. Of the total participants, 173 were female (49.01%), 179 were male (50.71%), and one person identified with another gender option (0.28%). 

The population of the second cycle of education in the Andorran educational system is approximately 1,200 students. With a sample size of 353 participants, the margin of error was 4.38%, with a 95% confidence level. These conditions provide an adequate basis for interpreting the study results.

Instruments

The instrument was divided into three sections: (a) Section 1. Sociodemographic variables, which included questions about the lower secondary school attended, grade, gender, age, and perception of being bullied (yes/no); (b) Section 2. Sports variables, which included questions to determine whether participants practiced sports (yes/no) and the type of sport they practiced (individual/team); (c) Section 3. Assessment of SB, consisting of items from the Cisneros Self-test (Piñuel & Oñate, 2012) to assess the SB index, intensity, and dimensions. 

The Cisneros Self-test (Piñuel & Oñate, 2012) consists of 50 items that assess the SB index, intensity, and eight dimensions of the different types of SB: intimidation and threats, theft, verbal harassment, restriction of communication, coercion, social exclusion and blocking, aggression, and contempt and ridicule. Scores range from 50 to 150 points, with scores of 59 or higher considered indicative of SB. The instrument takes between five and 10 minutes to complete. Responses are provided on a three-point Likert-type scale: 1 (Never), 2 (Rarely), and 3 (Often), and the questions do not specify a time frame for SB. The instrument has a reliability coefficient (Cronbach’s alpha) of 0.962.

Procedure

Participants were selected using nonprobability convenience sampling with institutional authorization. Data were collected between May and June 2025 using a Microsoft Forms questionnaire during school hours, in homeroom sessions, with informed consent from the families.

Once the study was completed, a detailed report on the prevalence rates found according to the variables analyzed was provided to the Andorran Ministry of Education and Higher Education. The possibility of implementing interventions and preventive actions aimed at case detection, awareness raising, and support for potential victims was also offered. 

The study received approval from the Research and Transfer Ethics Committee of the University of Lleida, certified on December 18, 2024, as documented in minutes 14/2024, and complied with ethical principles for research involving human participants. 

To ensure replicability and promote open science, the study data and results are openly available in the CORA data repository through the DOI: https://doi.org/10.34810/data2470.

Data Analysis

Data analysis was performed using IBM SPSS Statistics 30.0. After confirming the non-normality of the distribution using the Kolmogorov-Smirnov test, a descriptive analysis of the SB index and its dimensions, as well as of the different study groups, was conducted. Subsequently, inferential analyses were performed using the Mann-Whitney U test to examine differences according to educational cycle, gender, sports participation and sport type, as well as the perception and identification of SB. The SB identification variable was also created by cross-tabulating the perception of being bullied and SB according to the Cisneros Self-test, and binary logistic regression was performed to analyze the predictive power of the different SB dimensions for accurate and inaccurate identification of SB.

Results

The SB index had a mean score of 58.50 points (SD = 12.30), with 29.18% of the sample scoring at or above the cutoff established for SB (59 points).

As shown in Table 1, most of the groupings showed similar values for the SB index, with more pronounced descriptive differences in the variables related to the perception of being bullied and its identification.

Table 1

Descriptive analysis by grouping and SB index

See Table

As shown in Table 2, most participants were concentrated in the very low or medium levels across all SB dimensions. However, dimensions such as restriction of communication, contempt and ridicule, and social exclusion and blocking showed a greater relative presence in the high and very high categories.

Table 2

Distribution of school bullying dimensions by standardized categories (n, %)

See Table

The comparison between the perception of being bullied and the classification obtained based on the SB index cutoff showed a high prevalence of errors in SB identification. Cross-tabulating the two variables identified four groups: those who did not perceive themselves as being bullied and did not present SB according to the Cisneros Self-test (N = 248; 70.25%), those who did not perceive themselves as being bullied but did present SB according to the index (N = 90; 25.50%), those who perceived themselves as being bullied without reaching the established cutoff (N = 2; 0.57%), and those who perceived themselves as being bullied and did present SB according to the Cisneros Self-test (N = 13; 3.68%).

Based on this classification, two types of identification were distinguished: accurate, when there was agreement between the perception of being bullied and the results obtained on the Cisneros Self-test, and inaccurate, when there was a discrepancy between the two. This distribution is presented in Table 3.

Table 3

Cross-tabulation of perception, Cisneros Self-test results, and SB identification

See Table

Table 4 presents the results of the Mann-Whitney U test for the SB index according to the grouped variables. Statistically significant differences were observed only among sports participants according to gender, as well as for the variables perception of being bullied and SB identification. 

Table 4

Results of the Mann-Whitney U test for the SB index according to grouped variables

See Table

Among adolescents who participated in sports, significant differences in the SB index were identified according to gender (p = .049), with higher mean ranks in the female group, although the effect size was small (r = .138).

With regard to the perception of being bullied, participants who reported being bullied obtained significantly higher scores on the SB index (p < .001), with a small-to-moderate effect size (r = .248). 

Finally, significant differences were observed according to SB identification, such that adolescents with inaccurate identification had a considerably higher SB index (p < .001), with a large effect size (r = .694).

To analyze the SB dimensions associated with its identification, a binary logistic regression model was estimated using the standardized categories of the Cisneros Self-test to ensure model adequacy. 

Table 5 revealed that three SB dimensions were significantly associated with inaccurate SB identification. First, the contempt and ridicule dimension was the most relevant predictor (β = 1.560; p < .001), with an odds ratio (OR) of 4.761 (95% CI [2.206, 10.275]), indicating that adolescents with high scores on this dimension were more than four times as likely to identify SB inaccurately. The theft dimension also showed a significant association (β = 0.857; p = .019), approximately doubling the risk of inaccurate identification (OR = 2.356; 95% CI [1.154, 4.809]). By contrast, the aggression dimension showed an inverse effect (β = –1.240; p = .008), indicating a lower likelihood of inaccurately identifying SB (OR = 0.289; 95% CI [0.116, 0.720]). The remaining dimensions did not reach statistical significance, so no association with identification error could be established in this model. 

Table 5

Summary of binary logistic regression – dependent variable, inaccurate SB identification

See Table

Discussion

The present study examined SB among lower secondary education students in the Andorran educational system, considering sociodemographic variables and sports participation. The results showed an SB prevalence of 29.18%, with a higher concentration in the high and very high categories for the dimensions of contempt and ridicule, social exclusion and blocking, and restriction of communication. A substantial discrepancy was also identified between the perception of being bullied and the results of the Cisneros Self-test, with adolescents who inaccurately identified SB obtaining significantly higher scores. No significant differences in SB were observed between genders, between sports participants and nonparticipants, or between sport types. However, differences were found among sports participants according to gender, as well as according to the perception of being bullied and SB identification. Finally, SB dimensions with predictive capacity for inaccurate SB identification were identified, thereby addressing the objectives of the study.

The prevalence observed in the present study is consistent with previous literature (Ariani et al., 2025; Biswas et al., 2020; Modecki et al., 2014). However, the Andorran context has distinctive characteristics, such as population size, the educational system, and culture. Johansson et al. (2022) noted that the differences observed in SB may be attributable to social, cultural, and organizational factors specific to each educational system. They also indicated that schools are not isolated spaces but rather social contexts influenced by the sociocultural environment (Bronfenbrenner, 1979), which may affect both interaction dynamics and responses to SB situations.

With regard to sports participation, a significant gender difference was observed among sports participants. This result contradicts previous studies that reported higher prevalence rates among males (Marracho et al., 2021; Vveinhardt & Fominiene, 2022) or found no significant gender differences in SB (Benítez-Sillero et al., 2022; Escofet-Martín et al., 2025b; Tilindienė et al., 2010b), as females in the present study obtained significantly higher scores on the SB index. 

This finding reinforces the need to continue investigating the role of gender in SB (Martos-García et al., 2024; Rodkin et al., 2015) and suggests the need to develop gender-sensitive policies for the prevention, detection, and intervention of SB to contribute to its eradication (Díaz-Aguado et al., 2013).

Regarding sport type, no significant differences in SB were observed between participants in team and individual sports, consistent with the results of Marracho et al. (2021), but in contrast to those of Vveinhardt and Fominiene (2022). Similarly, no differences were identified between sports participants and nonparticipants. These findings do not consistently support any of the previous studies suggesting that sports participants are less likely to be bullied (Tilindienė et al., 2010a) or, conversely, more likely to be bullied (Rusillo et al., 2024; Zapata-López et al., 2024). It is important to emphasize that the quality of the sports experience is a relevant factor. Factors such as the influence of role models for athletes, the team environment, the competitive setting, or the role of coaches may foster certain relational dynamics that increase the risk of SB, particularly when aggressive behaviors are normalized or supervision is insufficient (Evans et al., 2016; Kalina et al., 2024; Nery et al., 2018). Therefore, further research is needed into the different variables that mediate this relationship (Carrasco-López et al., 2024; Escofet-Martín et al., 2025a; Martínez-Baena & Faus-Boscá, 2018).

Another important finding in efforts to address SB is the high rate of inaccurate identification found in the sample. The data indicate a high proportion of inaccurate SB identification, suggesting a concerning normalization of violence among adolescents, as they regard SB as mere jokes or blame the victim (Postigo et al., 2019). The study results are close to the discrepancy reported in previous literature. Specifically, the discrepancy observed was 8.04% lower than that reported by Connell et al. (2018) and 13.56% higher than that described by Sidera et al. (2020), who reported an inaccurate identification rate of 12.50%. The study whose findings were closest to those obtained in the present study was that of Larrañaga et al. (2024), which reported a rate 5.16% lower. Taken together, these studies and the present research suggest inaccurate SB identification rates ranging from 34.10% to 12.50%, reinforcing the concern expressed above. 

Furthermore, if these behaviors are observed and normalized by the group, they will be accepted as valid, a code of silence will emerge, and a culture of bullying will become established in schools. In this context, the lack of specific training among teachers, particularly Physical Education teachers, may hinder the detection of and intervention in SB situations. Therefore, training initiatives should be promoted for both practicing and preservice teachers, and the interventions they implement should also be reviewed (García-López & González, 2023; Ríos-Valdés et al., 2025). It is also essential to promote social and administrative policies aimed at curbing the normalization of violence in other contexts, including family, media, and social settings (Valente et al., 2024). Parenting styles also play a crucial role, as avoiding authoritarian parenting practices may help reduce SB behaviors. In this regard, it is essential to implement family guidance programs that promote positive parenting (Charalampous et al., 2018). Acting only within educational and sports contexts does not ensure effective intervention, as schools, secondary schools, and sports clubs or centers are settings in which the behaviors present in the societies to which they belong are reproduced (Bronfenbrenner, 1979).

To deepen our understanding of SB, it is essential to analyze how the different dimensions influence its identification by adolescents. The data from the present study indicate a statistically significant association between SB identification and three of its dimensions. First, the contempt and ridicule dimension showed a particularly relevant association, as high levels of this dimension increased the likelihood of inaccurate SB identification more than fourfold. The theft dimension was also associated with a greater risk of identification error, approximately doubling the likelihood of inaccurate SB identification. 

By contrast, the aggression dimension showed an inverse association, as high levels were associated with a greater likelihood of accurately identifying SB and reduced the likelihood of inaccurate identification by more than 70%. This finding differs from that reported by Sidera et al. (2020), who observed no significant relationships between SB identification and the dimensions of aggression, intimidation and threats, or verbal harassment. However, the results partially agree with those of Connell et al. (2018) and Larrañaga et al. (2024), who indicated that physical aggression facilitates SB identification.

This difference could be explained by the explicit and overt nature of physical aggression, in contrast to more subtle forms of SB, such as contempt and ridicule and theft, which tend to become normalized in interpersonal interactions. Aggression occurs more explicitly and causes physical and psychological harm to victims, facilitating its recognition. By contrast, failure to identify covert forms of violence may contribute to their normalization, with important long-term implications, as being a victim or perpetrator during adolescence has been associated with the continuation of violent behaviors into adulthood and even in intimate relationships (Adhia et al., 2019; Ruiz-Narezo et al., 2020), within what is known as the cycle of violence (Widom, 1989). 

Therefore, there is a clear need to address the accurate identification of all dimensions of SB through the development of comprehensive educational policies. These policies should be accompanied by programs that foster self-esteem, emotional self-efficacy, the development of secure bonds, and resilience, recognizing that accurate identification of SB is a first step toward breaking the cycle of violence.

Conclusions

The results of the present study make assistance contribution to the understanding of SB among lower secondary education students in the Andorran educational system, providing data that should be taken into account in future educational and social policies. It should also be considered that the Andorran context has a limited research tradition in the study of SB, particularly in relation to sports.

One of the main limitations of the study is its cross-sectional design, which prevents analysis of how this issue evolves and how it interacts with the variables examined. Another relevant limitation is the exclusion of variables related to the family, social, or media contexts, as these may have an important influence on SB. Although the sample is representative of the Andorran context, the results have limitations in terms of external validity and cannot be directly generalized to other contexts. In addition, the use of convenience sampling also limits the external validity of the results. 

One future line of research should involve longitudinal studies assessing how SB evolves and allowing potential causal factors and long-term effects to be identified. Further research integrating the family and social contexts is also essential to improve understanding of the impact of the environment on SB dynamics. Qualitative studies should also be conducted to interpret the processes involved in adolescents’ identification of SB, exploring aspects related to peer pressure, shame, cognitive dissonance, and different thresholds for perceiving violence and identifying it as SB. Similarly, further research into gender differences in SB in sports would be valuable, with particular attention to indirect dimensions such as social exclusion and blocking.

In conclusion, the findings highlight the importance of developing comprehensive educational policies that address the multifaceted nature of SB, considering both its explicit and covert aspects. These efforts should extend beyond the school context to include family, media, and social settings in order to improve prevention, detection, and intervention. 

Acknowledgments

The authors would like to thank the Ministry of Institutional Relations, Education and Universities of the Government of Andorra for its institutional support and the assistance provided for conducting the study, particularly in facilitating access to and coordination with the participating schools. 

Conflict of interest

No conflict of interest was reported by the authors.