ABSTRACT
Aim
To examine longitudinal changes in bullying involvement, self‐esteem, and perceived emotional intelligence following participation in a mental health nursing‐led emotional education programme among primary school children.
Design
Multicentre, longitudinal, non‐randomized, uncontrolled quasi‐experimental pretest–posttest study.
Methods
The study was conducted in 5 public primary schools in Madrid, Spain. A total of 507 students in fourth to sixth grade completed baseline assessment and 345 completed follow‐up. The programme included five student sessions, 2 parent sessions, and one teacher session. Bullying involvement, self‐esteem, and perceived emotional intelligence were assessed using self‐report instruments. Pretest–posttest analyses used valid matched observations for each outcome, complemented by exploratory posttest association analyses.
Results
Bullying involvement decreased from pretest to posttest. No clear pretest–posttest changes were observed in self‐esteem, emotional attention, emotional clarity, or emotional repair. At posttest, higher self‐esteem, emotional clarity, and emotional repair were inversely associated with bullying involvement. These concurrent associations were interpreted as non‐causal.
Conclusion
Participation in the programme was followed by reduced bullying involvement, while self‐esteem and perceived emotional intelligence remained broadly stable. The uncontrolled design precludes causal attribution.
Implications for the Profession and/or Patient Care
School nurses and mental health nurses may contribute to bullying prevention through health education, early identification of psychosocial vulnerability, emotional support, and coordination with families and teachers.
Impact
The study addresses nursing involvement in school‐based bullying prevention. Bullying involvement decreased, while other psychosocial outcomes remained stable. The findings may inform nursing practice and future controlled evaluations in primary schools.
Reporting Method
Reporting was guided by the Transparent Reporting of Evaluations with Nonrandomized Designs statement.
No Patient or Public Contribution
Students, parents, and teachers participated in the intervention and data collection but were not involved in study design, analysis, interpretation, or manuscript preparation.
Keywords: bullying prevention, child mental health, emotional education, emotional intelligence, emotional regulation, mental health nursing, primary education, school bullying, school health, self‐esteem
1. Introduction
School bullying is one of the most prevalent psychosocial and public health problems affecting children and adolescents worldwide. Defined as intentional and repetitive aggressive behaviour occurring within a context of power imbalance between peers, bullying may involve physical aggression, verbal harassment, social exclusion, intimidation, relational violence, and cyberbullying (Olweus 1993; Smith et al. 2019). Exposure to bullying has been consistently associated with adverse consequences for mental health, emotional development, academic achievement, and social adjustment, significantly affecting children's quality of life and long‐term well‐being (Abregú‐Crespo et al. 2024; Arseneault 2018).
The negative impact of bullying extends beyond the immediate school environment. Previous research has demonstrated that children involved in bullying dynamics frequently report higher levels of anxiety, depressive symptoms, emotional distress, social isolation, and school maladjustment (Abregú‐Crespo et al. 2024; Moore et al. 2017). Moreover, victimization during childhood has been associated with an increased risk of psychological difficulties persisting in adolescence and adulthood, highlighting bullying as a major public health concern requiring early preventive action (Olweus 1993; Arseneault 2018).
Current evidence suggests that effective bullying prevention strategies should not focus exclusively on reducing aggressive behaviours but also on strengthening protective psychosocial factors that promote resilience and healthy interpersonal relationships (Durlak et al. 2011). Among these factors, emotional intelligence has emerged as a particularly relevant construct. According to the ability model proposed by Mayer and Salovey, emotional intelligence refers to the capacity to perceive, understand, and regulate emotions in oneself and others (Mayer et al. 2004). Higher levels of emotional intelligence have been associated with better psychological adjustment, improved social competence, healthier peer relationships, and lower involvement in bullying‐related behaviours (Extremera and Rey 2018; Zych et al. 2019).
Within the dimensions of emotional intelligence, emotional clarity and emotional repair appear especially relevant during childhood and adolescence. Emotional clarity facilitates understanding emotional experiences, whereas emotional repair refers to the ability to regulate negative emotional states and maintain psychological well‐being (Mayer et al. 2004). Previous studies have reported that children with greater emotional regulation skills are less likely to engage in aggressive behaviours and more likely to use adaptive coping strategies when facing interpersonal conflicts (Extremera and Rey 2018; Zych et al. 2019; Brackett et al. 2011).
Self‐esteem constitutes another important protective factor in child development. Defined as an individual's overall evaluation of self‐worth, self‐esteem has been linked to emotional stability, resilience, social competence, and mental health (Orth and Robins 2014). Children with higher self‐esteem generally demonstrate better psychosocial adjustment and lower vulnerability to victimization, whereas low self‐esteem has been associated with emotional distress, social withdrawal, and increased involvement in bullying situations (Orth and Robins 2014; Tsaousis 2016). Furthermore, emotional intelligence and self‐esteem appear to be closely related constructs, jointly contributing to children's emotional well‐being and social adaptation (Extremera and Rey 2018; Orth and Robins 2014).
Gender differences have also been widely documented in relation to emotional competencies and bullying involvement. Previous research suggests that girls tend to report higher levels of emotional awareness, empathy, and emotional regulation, whereas boys are more frequently involved in direct forms of aggression and externalizing behaviours (Garaigordobil and Machimbarrena 2019; Chaplin and Aldao 2013). Understanding these differences is essential for designing and evaluating interventions that adequately address the emotional and psychosocial needs of both boys and girls during primary education.
From a socioecological perspective, bullying is a multifactorial phenomenon influenced by individual, family, school, and community factors (Bronfenbrenner 1979). Consequently, contemporary prevention models increasingly advocate for whole‐school approaches involving students, families, and educational professionals (UNESCO 2019). Emotional education programs implemented within school settings have shown promising effects in promoting emotional competencies, improving school climate, strengthening interpersonal relationships, and reducing aggressive behaviours (Durlak et al. 2011; UNESCO 2019).
Within this framework, mental health nurses are uniquely positioned to contribute to bullying prevention and emotional health promotion. Mental health nursing professionals play a key role in health education, early identification of psychosocial vulnerability, mental health promotion, and coordination between schools, families, and healthcare services (Yosep et al. 2023). Recent reviews have highlighted the growing contribution of nursing‐led interventions in addressing bullying and fostering emotionally supportive educational environments (Yosep et al. 2023; Celdrán‐Navarro et al. 2023). Nevertheless, evidence specifically examining the effectiveness of mental health nursing‐led emotional education programs in Primary Education remains limited, particularly in longitudinal studies evaluating both emotional competencies and bullying outcomes (Yosep et al. 2023; Celdrán‐Navarro et al. 2023).
Despite the growing evidence supporting social and emotional learning programs, there is still limited longitudinal evidence evaluating mental health nursing‐led interventions implemented through a whole‐school approach in Primary Education, particularly those simultaneously involving students, families, and teachers.
Baseline cross‐sectional findings from this research project, focusing on sex‐related differences in bullying, self‐esteem, and emotional intelligence, have been previously reported (Barba‐Muñoz et al. 2026). The present study extends that work by examining longitudinal pretest–posttest changes following the intervention.
Given the need for evidence‐based interventions that integrate emotional health promotion and bullying prevention, further research is required to examine changes in children's emotional development and psychosocial well‐being following participation in nursing‐led programmes. Therefore, the aim of the present study was to examine longitudinal changes in bullying, self‐esteem, and emotional intelligence following participation in a mental health nursing‐led emotional education programme involving students, families, and teachers among primary school children from five educational centres in Madrid, Spain. A secondary objective was to examine sex differences and identify emotional and psychosocial factors associated with lower levels of bullying involvement following the intervention.
Based on previous evidence, the following hypotheses were proposed: (H1) bullying scores would decrease following the intervention; (H2) self‐esteem levels would increase after participation in the programme; (H3) emotional intelligence dimensions would improve after the intervention; and (H4) higher emotional intelligence and self‐esteem would be associated with lower bullying involvement.
2. Materials and Methods
2.1. Study Design
A quantitative, longitudinal, multicentre, non‐randomized, uncontrolled quasi‐experimental study with a single‐group pretest–posttest design was conducted to examine changes in bullying, self‐esteem, and emotional intelligence following participation in a mental health nursing‐led emotional education programme among primary school children.
The study was conducted during the 2025–2026 academic year in 5 public primary schools located in Madrid, Spain. The intervention was implemented within the school setting and adopted a whole‐school approach involving students, families, and teachers. The educational programme focused on emotional education, emotional regulation, empathy, positive coexistence, conflict resolution, and bullying prevention.
A repeated‐measures approach was used to compare participants' emotional and behavioural outcomes before and after the intervention. Participants were not randomized, and no control or comparison group was included. Given the educational and ethical challenges associated with withholding a preventive emotional education programme from participating in schools, a single‐group quasi‐experimental design was considered appropriate for the study context. Accordingly, the observed pretest–posttest changes should not be interpreted as establishing a causal effect of the intervention. Rather, the findings represent within‐participant changes observed following participation in the programme, and alternative temporal, contextual, or developmental explanations cannot be completely excluded.
A cross‐sectional analysis of the baseline (pretest) data from this research project has previously been published, focusing on gender‐related differences in bullying, self‐esteem, and emotional intelligence among the initial sample of 507 students (Barba‐Muñoz et al. 2026). The present study represents the longitudinal intervention follow‐up of the same research project and focuses on within‐participant pretest–posttest changes following the mental health nursing‐led emotional education programme. Longitudinal analyses were conducted among participants with matched pretest and posttest assessments. The previously published article did not report the postintervention outcomes or the longitudinal pretest–posttest analyses presented in the current study.
The study was reported in accordance with the TREND statement for non‐randomized evaluations of behavioural and public health interventions. The intervention was described in accordance with the TIDieR checklist and guide to facilitate transparency and replication.
2.2. Participants and Setting
The study was conducted in five public primary schools located in Madrid, Spain. To preserve institutional confidentiality, participating schools were anonymized and identified throughout the manuscript as Centre 1, Centre 2, Centre 3, Centre 4, and Centre 5.
The initial pretest sample consisted of 507 students enrolled in the 4th, 5th, and 6th grades of Primary Education. Following implementation of the programme, 345 students completed the posttest assessment and constituted the follow‐up sample. The number of participants included in the paired pretest–posttest analyses varied by outcome according to the availability of valid matched observations for each instrument.
Participants were recruited through non‐probability convenience sampling based on school participation and accessibility. The inclusion criteria were: (1) enrolment in one of the participating schools; (2) attendance in the 4th, 5th, or 6th grade of Primary Education; (3) written informed consent provided by parents or legal guardians; and (4) voluntary student participation. For each outcome, questionnaires with incomplete data or more than 20% missing responses on the corresponding instrument were excluded from the relevant analysis.
No formal a priori sample‐size or statistical power calculation was used to determine recruitment. The intervention was conceived as a universal school‐based preventive programme and was offered to the entire eligible population of 4th‐, 5th‐, and 6th‐grade students in the participating schools. For ethical and educational reasons, participation in the programme was not restricted to a statistically selected subsample, as the potential benefits of the intervention were intended to be available to all eligible students. Consequently, the achieved sample was determined by the eligible school population, parental consent, student assent, and participation in the assessments rather than by a predefined statistical target.
The final follow‐up sample included 170 boys (49.3%) and 175 girls (50.7%), with a mean age of 10.04 years (SD = 0.93).
Descriptive characteristics of the follow‐up sample according to educational centre are presented in Table 1. The largest proportion of participants came from Centre 1 (n = 106), whereas Centre 2 contributed the smallest sample (n = 18). Overall, self‐esteem scores were within the moderate‐to‐high range across centres, although some variability was observed.
TABLE 1.
Follow‐up sample characteristics by educational centre.
| Educational center | n | Mean age (SD) | Boys n (%) | Girls n (%) | Mean self‐esteem (SD) | Range |
|---|---|---|---|---|---|---|
| Centre 1 | 106 | 10.54 (0.97) | 55 (51.9) | 51 (48.1) | 25.94 (5.05) | 8–34 |
| Centre 2 | 18 | 10.61 (0.50) | 7 (38.9) | 11 (61.1) | 26.56 (4.16) | 20–31 |
| Centre 3 | 86 | 9.80 (0.72) | 41 (47.7) | 45 (52.3) | 28.67 (3.96) | 16–34 |
| Centre 4 | 98 | 9.63 (0.71) | 50 (51.0) | 48 (49.0) | 26.72 (4.57) | 16–34 |
| Centre 5 | 37 | 10.00 (1.15) | 17 (45.9) | 20 (54.1) | 26.62 (5.32) | 14–34 |
| Total | 345 | 10.04 (0.93) | 170 (49.3) | 175 (50.7) | 26.95 (4.74) | 8–34 |
Source: Authors' own elaboration.
The follow‐up sample consisted of 345 primary school students with a balanced sex distribution (49.3% boys and 50.7% girls) and a mean age of 10.04 years (SD = 0.93). Mean self‐esteem scores ranged from 25.94 to 28.67 points across centres, indicating moderate‐to‐high levels of self‐esteem in the overall sample.
Following the descriptive characterization of the sample, the distribution of participants according to grade level, sex, and educational centre was examined (Table 2).
TABLE 2.
Distribution of participants according to grade level, sex, and educational centre at posttest.
| Educational centre | Sex | 4th Grade n (%) | 5th Grade n (%) | 6th Grade n (%) | Total | p a |
|---|---|---|---|---|---|---|
| Centre 1 | Boys | 15 (27.3) | 20 (36.4) | 20 (36.3) | 55 | 0.412 |
| Girls | 16 (31.4) | 19 (37.2) | 16 (31.4) | 51 | 0.412 | |
| Centre 2 | Boys | 1 (14.3) | 2 (28.6) | 4 (57.1) | 7 | 0.286 |
| Girls | 3 (27.3) | 4 (36.4) | 4 (36.3) | 11 | 0.286 | |
| Centre 3 | Boys | 14 (34.1) | 14 (34.1) | 13 (31.8) | 41 | 0.931 |
| Girls | 16 (35.6) | 15 (33.3) | 14 (31.1) | 45 | 0.931 | |
| Centre 4 | Boys | 18 (36.0) | 17 (34.0) | 15 (30.0) | 50 | 0.874 |
| Girls | 18 (37.5) | 16 (33.3) | 14 (29.2) | 48 | 0.874 | |
| Centre 5 | Boys | 5 (29.4) | 6 (35.3) | 6 (35.3) | 17 | 0.764 |
| Girls | 7 (35.0) | 7 (35.0) | 6 (30.0) | 20 | 0.764 | |
| Total | Boys | 53 (31.2) | 59 (34.7) | 58 (34.1) | 170 | 0.638 |
| Girls | 60 (34.3) | 61 (34.9) | 54 (30.8) | 175 | 0.638 |
Pearson's χ 2 test.
Source: Authors' own elaboration.
Overall, students were relatively evenly distributed across the 4th, 5th, and 6th grades of Primary Education. No statistically significant sex differences were observed in grade‐level distribution either within individual educational centres or in the overall sample (all p > 0.05), indicating a balanced academic composition between boys and girls.
This homogeneity is particularly relevant for the interpretation of subsequent analyses, as it reduces the potential confounding effect of educational level when examining differences in bullying involvement, self‐esteem, and emotional intelligence outcomes. Furthermore, the balanced representation across grades strengthens the internal validity of the longitudinal comparisons performed after the intervention.
Participant recruitment, intervention delivery, follow‐up, and inclusion in the outcome‐specific analyses are summarized in Figure 1.
FIGURE 1.
Flow diagram of participant recruitment, intervention delivery, follow‐up, and inclusion in the final analyses adapted from recommendations for reporting quasi‐experimental intervention studies. Source: Authors' own elaboration.
2.3. Intervention Program
The intervention was designed and delivered by a mental health nurse with specific training and professional experience in child and adolescent mental health, emotional education, and health promotion. To ensure consistency across schools, all sessions were conducted by the same professional.
The programme followed a whole‐school approach and included activities directed at students, families, and teachers. The intervention consisted of eight sessions delivered over an 8‐week period: five weekly classroom‐based sessions for students, two educational sessions for parents, and one training session for teachers.
Student sessions were conducted during regular school hours and lasted approximately 45 min each, corresponding to one standard class period. Sessions were delivered to intact classroom groups of approximately 25 students. Parent and teacher sessions were implemented separately and focused on reinforcing the emotional education contents addressed with students.
The student component addressed emotional identification and expression, emotional regulation, empathy, social skills, conflict resolution, positive coexistence, and bullying prevention. Parent sessions focused on emotional support, communication strategies, recognition of bullying indicators, and family involvement in children's emotional development. The teacher session addressed bullying prevention, emotional education, classroom coexistence, and the identification of students at psychosocial risk.
Participatory and experiential methodologies were employed throughout the intervention, including group discussions, role‐playing activities, cooperative learning strategies, reflective exercises, emotional awareness dynamics, and guided emotional regulation techniques. Digital presentations supported by interactive whiteboards and projectors were used in all schools to facilitate student engagement and standardize content delivery.
All participating schools received the same intervention structure, educational materials, and implementation procedures. The structure and contents of the intervention are summarized in Table 3.
TABLE 3.
Structure of the mental health nursing‐led emotional education programme.
| Target group | Sessions | Duration | Objectives | Main contents | Methodology |
|---|---|---|---|---|---|
| Students | 5 | 45 min/session | Develop emotional competencies and prevent bullying | Emotional identification, emotional regulation, empathy, social skills, conflict resolution, coexistence, bullying prevention | Role‐playing, cooperative learning, group discussions, reflective activities |
| Parents | 2 | 45 min/session | Strengthen family support and emotional education | Communication strategies, emotional support, bullying detection, family involvement | Workshops, discussion groups, practical exercises |
| Teachers | 1 | 45 min/session | Promote emotionally supportive classrooms | Emotional education, bullying prevention, coexistence, early detection of psychosocial vulnerability | Training seminar and case discussion |
Source: Authors' own elaboration.
To promote consistency in intervention delivery, all sessions followed a standardized protocol, used identical educational materials, and were based on predefined learning objectives. The same mental health nurse delivered all sessions across participating schools, thereby minimizing variability in implementation.
As shown in Table 3, the intervention adopted a whole‐school approach by simultaneously targeting students, families, and teachers. This structure was intended to reinforce emotional learning across the educational community and promote consistency between the different components of the programme.
2.4. Instruments
Three self‐report instruments were administered to assess school bullying, self‐esteem, and perceived emotional intelligence. All questionnaires were completed collectively during school hours under researcher supervision. The instruments were selected due to their previous use in Spanish child populations and their adequate psychometric properties.
2.4.1. Peer Harassment Questionnaire (PHQ)
School bullying was assessed using the Peer Harassment Questionnaire (PHQ), corresponding to the Spanish Cuestionario de Acoso entre Iguales (CAI) developed and validated by Magaz et al. Magaz et al. (2016). The instrument evaluates students' involvement in bullying situations within the school context and includes dimensions related to victimization, aggressive behaviours, and observer or bystander roles.
Higher scores indicate greater involvement in bullying dynamics. For the present analyses, the global PHQ score was used as the measure of bullying involvement. The PHQ demonstrated excellent internal consistency in the present sample (Cronbach's α = 0.91).
2.4.2. Self‐Esteem Scale for Primary Education (A‐EP)
Self‐esteem was assessed using the Self‐Esteem Questionnaire for Primary Education (A‐EP), developed by Ramos et al. Ramos et al. (2006). This instrument was specifically designed for children attending Primary Education and evaluates self‐perception, emotional self‐concept, personal value, and general self‐esteem.
A global self‐esteem score was calculated, with higher scores indicating more positive self‐evaluation. For descriptive purposes, self‐esteem levels were categorized as low, moderate, or high according to the instrument guidelines. The scale showed good internal consistency in the present study (Cronbach's α = 0.87).
2.4.3. Trait Meta‐Mood Scale (TMMS‐24)
Perceived emotional intelligence was measured using the Spanish version of the Trait Meta‐Mood Scale (TMMS‐24), adapted by Fernández‐Berrocal et al. Fernández‐Berrocal et al. (2004). The TMMS‐24 is a 24‐item self‐report instrument that evaluates three dimensions of emotional functioning: Emotional Attention, Emotional Clarity, and Emotional Repair.
Emotional Attention refers to the degree to which individuals attend to and value their emotional states. Emotional Clarity assesses the ability to understand and differentiate emotions, whereas Emotional Repair evaluates the perceived capacity to regulate negative emotions and maintain emotional well‐being.
Although the original Spanish validation was conducted in university students, more recent psychometric evidence supports the use of the TMMS‐24 in child and adolescent populations. (Martínez‐Líbano et al. 2025) validated the instrument in a sample of 636 children and adolescents, reporting a 3‐factor structure corresponding to emotional attention, emotional clarity, and emotional regulation, together with high internal consistency and evidence of convergent validity.
Higher scores indicate greater perceived emotional competence. In the present study, internal consistency was good‐to‐excellent across all dimensions: Emotional Attention (α = 0.84), Emotional Clarity (α = 0.88), and Emotional Repair (α = 0.90). Nevertheless, validation evidence remains more limited for children at the youngest end of the age range included in the present study, and this should be considered when interpreting the findings.
The main characteristics of the instruments used in the study are summarized in Table 4.
TABLE 4.
Characteristics and internal consistency of the instruments used in the study.
| Instrument | Construct assessed | Main dimensions | Cronbach's α (current study) |
|---|---|---|---|
| PHQ (Peer harassment questionnaire) | School bullying | Victimization, aggression, bystander behaviours | 0.91 |
| A‐EP | Self‐esteem | Global self‐esteem | 0.87 |
| TMMS‐24 | Emotional intelligence | Emotional attention, emotional clarity, emotional repair | 0.84–0.90 |
Source: Authors' own elaboration.
2.5. Data Collection Procedure
Data collection was conducted within the participating schools during regular school hours using a standardized administration protocol. All assessments were supervised by the same mental health nurse who was responsible for implementing the intervention, thereby ensuring consistency in data collection procedures across all educational centres.
Baseline (pretest) assessments were completed prior to the implementation of the intervention programme. Follow‐up (posttest) assessments were administered approximately 1 week after the completion of the final intervention session to assess short‐term pretest–posttest changes following participation in the programme. Although minor variations in administration dates occurred due to school scheduling requirements, all posttest evaluations were completed within 2 weeks following the intervention.
Questionnaires were administered collectively in classroom settings under researcher supervision. Before completion, participants received standardized verbal instructions regarding the purpose of the questionnaires and the response procedure. Students were encouraged to seek clarification whenever necessary to ensure accurate understanding of questionnaire items. The average time required to complete the assessment battery was approximately 20–25 min.
To facilitate longitudinal analyses while maintaining participant anonymity, each student was assigned a unique identification code that enabled matching of pretest and posttest responses. No personal identifiers were collected at any stage of the study. All data were handled confidentially and stored in accordance with applicable data protection regulations.
Prior to data collection, school administrators, teaching staff, parents, and legal guardians received detailed information regarding the study objectives, procedures, voluntary nature of participation, and confidentiality safeguards. Written informed consent was obtained from parents or legal guardians, and verbal assent was obtained from participating students before enrolment in the study.
Following data collection, questionnaires were reviewed for completeness and data quality. Cases presenting more than 20% missing responses on any assessment instrument were excluded from the corresponding analyses, according to the predefined exclusion criteria. Data were subsequently coded, entered into a secure database, and checked for accuracy before statistical analysis.
2.6. Statistical Analysis
Primary statistical analyses were performed using IBM SPSS Statistics for Windows (IBM Corp., Armonk, NY, USA).
Descriptive statistics were calculated for all study variables. Categorical variables were expressed as frequencies and percentages, whereas continuous variables were summarised using means and standard deviations (SD).
Prior to inferential analyses, the distribution of quantitative variables was examined using the Shapiro–Wilk test, together with the assessment of skewness, kurtosis, and graphical inspection of histograms and Q–Q plots. Given the sample size and the approximate normal distribution of the main study variables, parametric analyses were conducted.
Differences between pretest and posttest scores for bullying, self‐esteem, and emotional intelligence dimensions were evaluated using paired‐samples Student's t‐tests. Mean differences and p‐values were reported. Effect sizes were calculated using Cohen's dz. for paired samples and interpreted according to conventional criteria as small (0.20), moderate (0.50), and large (0.80).
To account for the hierarchical structure of the data, an additional sensitivity analysis was conducted considering both the repeated‐measures nature of the study and classroom‐level clustering. Individual pretest–posttest change scores were calculated for participants with securely matched and complete observations for each outcome. These change scores were analysed using ordinary least squares regression models with classroom‐clustered robust standard errors. Classroom membership was defined according to educational centre, academic grade, and classroom section, resulting in 24 identifiable classroom clusters. Educational centre, sex, and academic grade were included as fixed adjustment covariates. Given the limited number of participating schools (n = 5), educational centre was modelled as a fixed effect rather than estimating a school‐level random effect. Because complete and securely matched pretest–posttest item‐level data were not available for all participants across all instruments, sample sizes varied according to outcome. These analyses were therefore considered sensitivity analyses rather than replacements for the primary paired comparisons.
An attrition analysis was conducted to compare participants who completed the follow‐up assessment (n = 345) with those who participated at baseline but did not complete the posttest (n = 162). Available baseline demographic characteristics and study outcomes were compared between groups using Pearson's chi‐square tests for categorical variables and independent‐samples Student's t‐tests for continuous variables, as appropriate.
Associations between bullying involvement, self‐esteem, and emotional intelligence variables at posttest were examined using Pearson's correlation coefficients. Correlation strength was interpreted as weak (r < 0.30), moderate (r = 0.30–0.49), or strong (r ≥ 0.50).
Covariate selection for the multivariable analyses was guided by a directed acyclic graph (DAG) developed a priori based on the literature reviewed in the Introduction (Figure 2). The DAG represented the hypothesised relationships among sociodemographic characteristics, educational context, emotional competencies, self‐esteem, and bullying involvement. Sex and academic grade were considered individual‐level potential confounding variables. Educational centre was additionally included as a contextual adjustment variable because school‐level differences could be associated with both psychosocial characteristics and bullying involvement. Accordingly, sex, academic grade, and educational centre were included as adjustment covariates in the multivariable models. Self‐esteem, emotional attention, emotional clarity, and emotional repair were considered the main psychosocial explanatory variables of interest.
FIGURE 2.
Directed acyclic graph (DAG) guiding covariate selection for the multiple regression analysis. Source: Authors' own elaboration.
The DAG represents the hypothesised relationships between the main psychosocial explanatory variables (self‐esteem, emotional attention, emotional clarity, and emotional repair), individual‐level potential confounding variables (sex and academic grade), the contextual adjustment variable (educational centre), and bullying involvement at posttest. Arrows represent hypothesised directional relationships based on the literature reviewed. Sex, academic grade, and educational centre were included as adjustment covariates in the multiple regression model. Bullying involvement at posttest, assessed using the Peer Harassment Questionnaire (PHQ), was considered the outcome variable.
An exploratory multiple linear regression analysis was conducted using posttest bullying involvement as the dependent variable. Self‐esteem, emotional attention, emotional clarity, and emotional repair measured at posttest were included as psychosocial explanatory variables, while sex, academic grade, and educational centre were included as adjustment covariates. Because the psychosocial variables and bullying involvement were assessed concurrently at posttest, the resulting coefficients were interpreted as cross‐sectional associations rather than predictive or causal effects. Baseline bullying involvement was not included in this exploratory posttest association model because its purpose was to characterize contemporaneous relationships among psychosocial variables and bullying at the postintervention assessment. Standardized beta coefficients (β), 95% confidence intervals (95% CI), and adjusted coefficients of determination (adjusted R 2) were calculated. Multicollinearity was assessed using tolerance and variance inflation factor (VIF) values. All VIF values were below 2.0, indicating the absence of relevant multicollinearity.
As an exploratory posttest analysis, Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) was performed to examine heterogeneity across multicategorical profiles defined by educational centre, sex, and academic grade. These analyses used participants with valid posttest data for each corresponding outcome and did not require availability of a matched pretest observation; consequently, sample sizes differed from those used in the paired pretest–posttest analyses. Null models were first estimated to quantify the proportion of variance attributable to differences between profiles using the Variance Partition Coefficient (VPC). Additive models including educational centre, sex, and academic grade as fixed effects were subsequently fitted to assess the extent to which between‐profile heterogeneity was explained by the independent contribution of these characteristics. Full MAIHDA results are presented in Table S1. Statistical significance was established at p < 0.05 for all analyses.
2.7. Ethical Considerations
The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and applicable national and international regulations governing research involving minors. The study protocol was reviewed and approved prior to data collection by the Research Ethics Committee of Universitat Jaume I (CEISH/08/2025; 12 May 2025). Written informed consent was obtained from the parents or legal guardians of all participating children, and verbal assent was obtained from the children before their participation. Participation was voluntary, anonymous, and confidential. Authorization was also obtained from the participating schools.
2.7.1. Trial Registration
This study was not prospectively registered in a clinical trial registry. At the time of study conception, it was designed and ethically approved as a school‐based, non‐randomized, uncontrolled quasi‐experimental educational intervention. Ethical approval was obtained prior to data collection (CEISH/08/2025).
3. Results
3.1. Participant Characteristics
Participant characteristics are presented in Table 1.
Of the 507 students who completed the baseline assessment, 345 completed the posttest assessment and constituted the follow‐up sample, whereas 162 participants did not complete the follow‐up assessment. The follow‐up sample consisted of 170 boys (49.3%) and 175 girls (50.7%), with a mean age of 10.04 years (SD = 0.93).
Participants were recruited from five public primary schools, anonymized as Centre 1, Centre 2, Centre 3, Centre 4, and Centre 5. Students were enrolled in the 4th, 5th, and 6th grades of Primary Education. The distribution across educational levels was balanced, with 32.8% of students attending 4th grade, 34.8% attending 5th grade, and 32.4% attending 6th grade.
An attrition analysis comparing participants who completed follow‐up (n = 345) with those who did not (n = 162) showed no statistically significant differences in sex distribution (χ 2(1) = 1.01, p = 0.315; Cramer's V = 0.05) or academic grade (χ 2(2) = 0.08, p = 0.960; Cramer's V = 0.01). However, attrition differed significantly across educational centres (χ 2(4) = 96.80, p < 0.001; Cramer's V = 0.44), with loss‐to‐follow‐up rates ranging from 4.4% to 64.1% across participating schools.
Within the follow‐up sample, Centre 3 presented the highest mean self‐esteem scores, whereas Centre 1 showed the lowest mean self‐esteem and the greatest variability in scores.
3.2. Pretest–Posttest Changes in Bullying, Self‐Esteem, and Emotional Intelligence
In the primary paired analyses, pretest and posttest scores were compared using participants with valid matched observations for each outcome. Because the availability of complete matched data differed across instruments, the paired analytic sample varied by outcome (PHQ, n = 201; A‐EP, n = 241; TMMS‐24, n = 238). Table 5 summarizes the within‐participant changes observed between the two assessment time points.
TABLE 5.
Pretest–posttest comparison of bullying, self‐esteem, and emotional intelligence outcomes using matched observations.
| Variable | n paired | Pretest mean (SD) | Posttest mean (SD) | Mean difference | p | Cohen's dz |
|---|---|---|---|---|---|---|
| Bullying involvement (PHQ total score) | 201 | 9.14 (10.14) | 7.00 (8.60) | −2.15 | < 0.001 | −0.27 |
| Self‐esteem (A‐EP) | 241 | 27.36 (4.21) | 27.41 (4.47) | +0.05 | 0.809 | 0.02 |
| Emotional attention (TMMS‐24) | 238 | 22.95 (7.48) | 23.15 (8.63) | +0.20 | 0.666 | 0.03 |
| Emotional clarity (TMMS‐24) | 238 | 26.42 (7.50) | 26.74 (8.56) | +0.33 | 0.528 | 0.04 |
| Emotional repair (TMMS‐24) | 238 | 28.52 (7.48) | 28.66 (8.62) | +0.13 | 0.798 | 0.02 |
Note: Mean difference was calculated as posttest minus pretest. Cohen's dz represents the standardized within‐participant change; negative values indicate lower scores at posttest.
Source: Authors' own elaboration.
A statistically significant reduction in bullying involvement was observed between the pretest and posttest assessments (mean change = −2.15, p < 0.001), with a small within‐participant effect size (Cohen's dz. = −0.27). In contrast, no statistically significant pretest–posttest changes were observed in self‐esteem (mean change = +0.05, p = 0.809), Emotional Attention (mean change = +0.20, p = 0.666), Emotional Clarity (mean change = +0.33, p = 0.528), or Emotional Repair (mean change = +0.13, p = 0.798). Effect sizes for these outcomes were negligible.
To evaluate the robustness of the primary findings while accounting for the hierarchical structure of the data, an additional sensitivity analysis was performed using securely matched pretest–posttest observations with identifiable classroom membership and classroom‐clustered robust standard errors. Educational centre, sex, and academic grade were included as adjustment covariates.
The reduction in bullying involvement remained statistically significant after accounting for classroom clustering (n = 197; mean change = −2.14; 95% CI: −3.27 to −1.01; p < 0.001). Consistent with the primary paired analyses, no statistically significant changes were observed for self‐esteem (n = 235; mean change = 0.05; 95% CI: −0.35 to 0.45; p = 0.794), Emotional Attention (n = 232; mean change = 0.25; 95% CI: −0.59 to 1.10; p = 0.538), Emotional Clarity (n = 232; mean change = 0.29; 95% CI: −0.59 to 1.17; p = 0.503), or Emotional Repair (n = 232; mean change = 0.13; 95% CI: −0.64 to 0.91; p = 0.725).
The sensitivity analysis therefore yielded the same substantive pattern as the primary paired analysis: bullying involvement decreased significantly between assessments, whereas changes in self‐esteem and perceived emotional intelligence were small and not statistically significant. Detailed results of the clustering‐sensitive analyses are provided in Table S2.
3.3. Postintervention Differences According to Sex and Educational Centre
Given the previously reported sex differences in emotional competencies and bullying involvement, postintervention outcomes were further examined according to sex and educational centre. Tables 6 and 7 summarize these comparisons.
TABLE 6.
Distribution of bullying and self‐esteem levels according to sex in the posttest assessment.
| Variable | Boys n (%) | Girls n (%) | p | Effect size |
|---|---|---|---|---|
| Low bullying | 80 (47.1) | 107 (61.1) | 0.011 | V = 0.20 |
| Moderate bullying | 75 (44.1) | 62 (35.4) | ||
| High bullying | 15 (8.8) | 6 (3.5) | ||
| Low self‐esteem | 22 (12.9) | 11 (6.3) | 0.008 | V = 0.22 |
| Moderate self‐esteem | 100 (58.8) | 85 (48.6) | ||
| High self‐esteem | 48 (28.3) | 79 (45.1) |
Source: Authors' own elaboration.
TABLE 7.
Emotional intelligence scores according to sex in the posttest assessment.
| Variable | Boys mean (SD) | Girls mean (SD) | p | Cohen's d |
|---|---|---|---|---|
| Emotional attention | 25.9 (4.6) | 28.8 (4.1) | 0.006 | 0.51 |
| Emotional clarity | 24.3 (4.2) | 27.1 (3.8) | 0.006 | 0.51 |
| Emotional repair | 23.8 (4.4) | 26.4 (3.9) | 0.006 | 0.51 |
Source: Authors' own elaboration.
Girls showed significantly lower levels of bullying involvement and higher levels of self‐esteem compared with boys. Similar trends were observed across all educational centres, although statistically significant sex differences were particularly evident in Centres 1, 3, and 4.
Female students obtained significantly higher scores across all dimensions of emotional intelligence. The largest centre‐specific differences were observed in Centre 3, Female students showed higher posttest scores across the three dimensions of perceived emotional intelligence. These findings represent cross‐sectional posttest differences and should not be interpreted as differences in response to the programme.
3.4. Correlations Between Bullying, Self‐Esteem, and Emotional Intelligence
Posttest correlations were examined between bullying involvement, self‐esteem, and perceived emotional intelligence. The strongest association was observed between bullying and emotional repair (r = −0.51), indicating that higher emotional repair scores were associated with lower levels of bullying involvement. Positive correlations were observed between self‐esteem and emotional competencies, particularly emotional repair and emotional clarity (Table 8).
TABLE 8.
Posttest correlations among bullying involvement, self‐esteem, and perceived emotional intelligence.
| Variables | r | 95% CI | p |
|---|---|---|---|
| Bullying—Self‐esteem | −0.48 | −0.56 to −0.39 | < 0.001 |
| Bullying—Emotional Attention | −0.24 | −0.33 to −0.14 | < 0.001 |
| Bullying—Emotional Clarity | −0.45 | −0.53 to −0.36 | < 0.001 |
| Bullying—Emotional Repair | −0.51 | −0.59 to −0.43 | < 0.001 |
| Self‐esteem—Emotional Attention | 0.38 | 0.29 to 0.46 | < 0.001 |
| Self‐esteem—Emotional Clarity | 0.57 | 0.50 to 0.64 | < 0.001 |
| Self‐esteem—Emotional Repair | 0.62 | 0.55 to 0.68 | < 0.001 |
| Emotional Clarity—Emotional Repair | 0.69 | 0.63 to 0.74 | < 0.001 |
Source: Authors' own elaboration.
Posttest correlations among bullying involvement, self‐esteem, and perceived emotional intelligence.
3.5. Multivariable Associations With Bullying Involvement at Posttest
The exploratory posttest multivariable model explained approximately 48% of the variability in bullying involvement at posttest (adjusted R 2 = 0.48). Higher emotional repair, self‐esteem, and emotional clarity scores showed the strongest inverse associations with bullying involvement. Male sex was associated with higher bullying scores, whereas academic grade showed a small inverse association with bullying involvement (Table 9). Because the psychosocial variables and bullying involvement were assessed concurrently at posttest, these associations should not be interpreted as predictive or causal relationships.
TABLE 9.
Exploratory multiple linear regression model of variables associated with bullying involvement at posttest.
| Variable | β | 95% CI | p |
|---|---|---|---|
| Self‐esteem | −0.31 | −0.42 to −0.20 | < 0.001 |
| Emotional attention | −0.12 | −0.21 to −0.03 | 0.014 |
| Emotional clarity | −0.24 | −0.35 to −0.13 | < 0.001 |
| Emotional repair | −0.37 | −0.49 to −0.25 | < 0.001 |
| Male sex | 0.19 | 0.08 to 0.30 | 0.002 |
| Academic grade | −0.09 | −0.17 to −0.01 | 0.038 |
Note: Model statistics: Adjusted R 2 = 0.48; F = 32.71; p < 0.001. The model was additionally adjusted for educational centre. Educational centre coefficients are not displayed.
Source: Authors' own elaboration.
Exploratory posttest MAIHDA, conducted using participants with valid posttest data for each outcome, showed generally low‐to‐moderate heterogeneity across profiles defined by educational centre, sex, and academic grade. Bullying involvement showed the highest initial between‐profile variability (VPC = 6.39%), followed by self‐esteem (VPC = 4.78%), emotional repair (VPC = 3.60%), and emotional clarity (VPC = 2.22%). After inclusion of the additive effects of educational centre, sex, and academic grade, the residual between‐profile variance approached zero across outcomes, indicating that most of the observed heterogeneity was explained by the independent contribution of these characteristics rather than by additional profile‐specific effects. Detailed results are provided in Table S1.
Overall, the posttest analyses indicated inverse associations between emotional competencies, self‐esteem, and bullying involvement. These findings provide additional information on the psychosocial characteristics associated with bullying at the postintervention assessment but do not establish temporal, predictive, or causal relationships.
4. Discussion
The present study examined longitudinal changes in bullying involvement, self‐esteem, and emotional intelligence following participation in a mental health nursing‐led emotional education program among primary school children from five educational centres in Madrid, Spain. Overall, bullying involvement decreased significantly between the pretest and posttest assessments, whereas no statistically significant changes were observed in self‐esteem or perceived emotional intelligence. Given the uncontrolled single‐group design, these findings should be interpreted as changes occurring following participation in the program rather than as established causal effects of the intervention. Furthermore, higher emotional intelligence and self‐esteem were associated with lower levels of bullying involvement.
One of the principal findings was the significant reduction in overall bullying involvement between the pretest and posttest assessments. The magnitude of this within‐participant change was small (Cohen's dz. = −0.27). These findings are consistent with previous evidence showing that social and emotional learning programmes may be associated with reductions in bullying‐related behaviours, stronger prosocial interactions, and more supportive school environments. However, because the present study did not include a control group, alternative temporal, contextual, or developmental explanations for the observed change cannot be excluded. Bullying is a complex phenomenon influenced by individual, relational, family, and school‐level factors; therefore, the observed reduction should not be attributed exclusively to participation in the programme.
The magnitude of the observed pretest–posttest difference in overall bullying involvement (d = 0.68) was moderate‐to‐large, suggesting that the change may be meaningful from an educational and public health perspective. Bullying is a complex phenomenon influenced by individual, relational, family, and school‐level factors; therefore, the observed differences should not be attributed exclusively to participation in the program. Nevertheless, the findings highlight the potential relevance of structured emotional education programs implemented within school settings.
In contrast, no statistically significant pretest–posttest changes were observed in self‐esteem or in the three dimensions of perceived emotional intelligence assessed by the TMMS‐24: Emotional Attention, Emotional Clarity, and Emotional Repair. Effect sizes for these outcomes were negligible. These findings suggest that the short‐term changes observed following the programme were not accompanied by measurable changes in these psychosocial outcomes. Self‐esteem and emotional competencies may require longer periods of development or more intensive intervention to produce detectable changes, although this interpretation should be examined in future controlled studies with longer follow‐up periods.
Although Emotional Repair did not show a statistically significant pretest–posttest change, it demonstrated the strongest inverse association with bullying involvement in the posttest multivariable analysis. Students reporting higher emotional repair scores also reported lower levels of bullying involvement. However, because emotional repair and bullying were measured concurrently at posttest, this association should not be interpreted as demonstrating a protective or causal effect. Rather, the findings indicate a relationship between perceived emotional regulation and bullying involvement that warrants further longitudinal investigation.
The correlational analyses further showed moderate‐to‐strong negative associations between bullying involvement and self‐esteem, emotional clarity, and emotional repair. The strongest association was observed between bullying and emotional repair. Positive correlations were also observed between self‐esteem and emotional intelligence dimensions, particularly emotional clarity and emotional repair. These findings suggest that emotional competencies and self‐esteem are closely related to bullying involvement and psychosocial adjustment, although the direction of these relationships cannot be established from the present analyses.
The multivariable model explained approximately 48% of the variance in bullying scores at posttest. Emotional repair, self‐esteem, and emotional clarity showed the strongest inverse associations with bullying involvement. These findings highlight the relevance of emotional regulation processes and self‐concept in understanding bullying‐related behaviours. Nevertheless, because the explanatory variables and bullying outcome were assessed concurrently at posttest, the regression coefficients should be interpreted as associations rather than causal or predictive effects.
Another relevant aspect of the program was its whole‐school approach. Unlike interventions directed exclusively at students, the program simultaneously involved students, families, and teachers. Contemporary bullying prevention frameworks emphasize coordinated action across the educational community. Family participation may reinforce emotional learning and communication beyond the classroom, while teacher involvement may contribute to emotionally supportive learning environments. However, the present design does not allow the independent contribution of the student, family, and teacher components to be determined.
From a mental health nursing perspective, the findings support further consideration of nurses within school‐based mental health promotion and bullying prevention initiatives. Mental health nurses possess competencies in emotional education, psychosocial assessment, health promotion, and early intervention that may contribute to multidisciplinary educational initiatives. The reduction in bullying involvement observed following participation in the programme suggests that nursing‐led emotional education warrants further evaluation as a school‐based approach to bullying prevention and positive school coexistence; although controlled studies are required to establish its effectiveness.
Several strengths of the present study should be highlighted. These include its multicentre design, relatively large initial sample size, longitudinal pretest–posttest assessment, inclusion of multiple educational stakeholders, and simultaneous evaluation of bullying, self‐esteem, and emotional intelligence. Furthermore, the incorporation of correlational and multivariable analyses provided additional information on the emotional and psychosocial factors associated with bullying involvement.
Nevertheless, several limitations should be considered when interpreting the findings. First, the quasi‐experimental design without a control group limits the ability to establish causal relationships between participation in the program and the observed pretest–posttest changes. External factors, contextual influences, maturation, or natural developmental processes cannot be excluded as alternative explanations. Second, the use of self‐report measures may have introduced response bias, including social desirability effects and subjective interpretations of emotional experiences.
Although recent psychometric evidence supports the use of the TMMS‐24 in child and adolescent populations, validation evidence remains more limited for children at the youngest end of the age range included in the present study. Therefore, the construct validity of the instrument in these younger participants should be interpreted with caution.
Third, attrition between baseline and follow‐up assessments resulted in a reduction of the sample from 507 to 345 participants, representing a loss of approximately 32% of the initial sample. This loss was primarily attributable to student absenteeism on the days of questionnaire administration and the exclusion of incomplete or incorrectly completed questionnaires that did not meet the predefined quality criteria. Although no significant differences were observed according to sex or academic grade, attrition differed substantially across educational centres, which may have introduced selection bias. The possibility of attrition bias therefore cannot be excluded.
Finally, posttest assessments were conducted shortly after completion of the program, limiting conclusions regarding the long‐term sustainability of the observed changes. Future studies should incorporate controlled or randomized designs, longer follow‐up periods, and multi‐informant assessment strategies involving teachers, parents, and peers. Further research should also examine the mechanisms underlying changes in emotional competencies and bullying involvement and evaluate the longer‐term effectiveness and scalability of nursing‐led interventions across diverse educational settings.
Despite these limitations, a reduction in bullying involvement was observed following participation in the mental health nursing‐led emotional education programme, whereas self‐esteem and perceived emotional intelligence remained broadly stable. Strengthening emotional competencies and self‐esteem may represent a relevant component of school‐based bullying prevention and mental health promotion initiatives. Integrating emotional education into school health initiatives and fostering collaboration between healthcare and educational systems may therefore represent a promising approach to promoting safer and more supportive school environments. The study was conducted exclusively in public schools from one Spanish region, which should also be considered when assessing the generalizability of the findings.
4.1. Implications for Nursing Practice
The findings support the potential contribution of school nurses and mental health nurses to school‐based bullying prevention and emotional health promotion. Nursing professionals are well positioned to contribute to health education, early identification of psychosocial vulnerability, emotional support, and coordination between students, families, teachers, primary care services, and other community resources (Barba Muñoz et al. 2026; Vargas‐Martínez et al. 2025).
Previous evidence indicates that school‐based psychoeducational programmes can contribute to reducing bullying‐related behaviours, although their effectiveness may vary according to educational stage, school climate, cultural context, family involvement, and programme adaptation (Zafra‐Agea et al. 2025). In this context, school and mental health nurses may contribute not only to intervention delivery but also to the assessment of needs, adaptation of educational content, identification of students requiring additional support, and coordination of multidisciplinary responses.
The present findings should nevertheless be interpreted cautiously. Participation in the nursing‐led programme was followed by a reduction in bullying involvement, whereas no statistically significant pretest–posttest changes were observed in self‐esteem or perceived emotional intelligence. Therefore, the results do not establish the effectiveness of nursing‐led interventions or the independent contribution of the nursing component. Controlled studies are needed to determine their effectiveness, sustainability, and implementation requirements in different school settings.
5. Conclusions
The present study provides preliminary evidence of a reduction in overall bullying involvement following participation in a mental health nursing‐led emotional education programme among primary school students. In contrast, no statistically significant pretest–posttest changes were observed in self‐esteem, Emotional Attention, Emotional Clarity, or Emotional Repair.
Emotional repair, emotional clarity, and self‐esteem showed the strongest inverse associations with bullying involvement at posttest, highlighting the relevance of emotional competencies and self‐concept in understanding bullying‐related behaviours. These associations were cross‐sectional and should not be interpreted as predictive or causal. These findings support further consideration of emotional education within school health promotion initiatives and reinforce the potential role of mental health nurses in the development and implementation of whole‐school preventive interventions.
Overall, the results suggest that emotional education programmes involving students, families, and teachers may represent a promising strategy for supporting bullying prevention and school‐based psychosocial well‐being during primary education. However, given the uncontrolled single‐group design, the observed changes should not be interpreted as establishing a causal effect of the intervention. Controlled studies with longer follow‐up periods are needed to confirm these findings.
Author Contributions
Malena Barba‐Muñoz: conceptualization, investigation, methodology, writing – original draft, writing – review and editing. Maria Pilar Molés‐Julio: validation, methodology, supervision, writing – review and editing. José A. Zafra‐Agea: writing – review and editing, investigation, methodology, validation, visualization, formal analysis, supervision, data curation.
Funding
The authors have nothing to report.
Ethics Statement
The study was conducted in accordance with the Declaration of Helsinki and was approved prior to data collection by the Research Ethics Committee of Universitat Jaume I (CEISH/08/2025; 12 May 2025). Written informed consent was obtained from the parents or legal guardians of all participating children prior to data collection.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Exploratory posttest multilevel analysis of individual heterogeneity and discriminatory accuracy (maihda) across bullying, self‐esteem, and emotional intelligence outcomes.
Table S2: Sensitivity analysis of pretest–posttest changes accounting for classroom clustering.
Acknowledgements
The authors would like to thank the participating schools, students, families, and educational staff for their collaboration and commitment throughout this study. Their participation made this research possible and contributed to advancing knowledge on bullying prevention and school nursing practice.
Contributor Information
Maria Pilar Molés‐Julio, Email: mjulio@uji.es.
José A. Zafra‐Agea, Email: joseantonio.zafra@uab.cat.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- Abregú‐Crespo, R. , Garriz‐Luis A., Ayora M., et al. 2024. “School Bullying in Children and Adolescents With Neurodevelopmental and Psychiatric Conditions: A Systematic Review and Meta‐Analysis.” Lancet Child Adolesc Health 8, no. 2: 122–134. [DOI] [PubMed] [Google Scholar]
- Arseneault, L. 2018. “Annual Research Review: The Persistent and Pervasive Impact of Being Bullied in Childhood and Adolescence.” Journal of Child Psychology and Psychiatry, and Allied Disciplines 59, no. 4: 405–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Barba Muñoz, M. , Zafra‐Agea J. A., Martí Marco E., et al. 2026. “Effectiveness of School‐Based Psychoeducational Program in Reducing Bullying and Improving Self‐Esteem: A Systematic Review.” Health 14, no. 3: 330. 10.3390/healthcare14030330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Barba‐Muñoz, M. , Molés‐Julio M. P., and Zafra‐Agea J. A.. 2026. “School Bullying, Self‐Esteem, and Emotional Intelligence Among Primary Education Students: A Gender‐Based Pretest Analysis From a School Nursing Perspective.” Health 14, no. 16: 2571. 10.3390/healthcare14162571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Brackett, M. A. , Rivers S. E., and Salovey P.. 2011. “Emotional Intelligence: Implications for Personal, Social, Academic, and Workplace Success.” Social and Personality Psychology Compass 5, no. 1: 88–103. [Google Scholar]
- Bronfenbrenner, U. 1979. The Ecology of Human Development. Harvard University Press. [Google Scholar]
- Celdrán‐Navarro, M. C. , Leal‐Costa C., Suárez‐Cortés M., Molina‐Rodríguez A., and Jiménez‐Ruiz I.. 2023. “Nursing Interventions Against Bullying: A Systematic Review.” International Journal of Environmental Research and Public Health 20, no. 4: 2914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chaplin, T. M. , and Aldao A.. 2013. “Gender Differences in Emotion Expression in Children: A Meta‐Analytic Review.” Psychological Bulletin 139, no. 4: 735–765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Durlak, J. A. , Weissberg R. P., Dymnicki A. B., Taylor R. D., and Schellinger K. B.. 2011. “The Impact of Enhancing Students' Social and Emotional Learning: A Meta‐Analysis of School‐Based Universal Interventions.” Child Development 82, no. 1: 405–432. [DOI] [PubMed] [Google Scholar]
- Extremera, N. , and Rey L.. 2018. “Emotional Intelligence and Peer Victimization in Youth: A Meta‐Analytic Review.” Aggression and Violent Behavior 43: 1–8. [Google Scholar]
- Fernández‐Berrocal, P. , Extremera N., and Ramos N.. 2004. “Validity and Reliability of the Spanish Modified Version of the Trait Meta‐Mood Scale.” Psychological Reports 94, no. 3: 751–755. [DOI] [PubMed] [Google Scholar]
- Garaigordobil, M. , and Machimbarrena J. M.. 2019. “Gender Differences in Bullying and Cyberbullying Among Children and Adolescents.” Psychosocial Intervention 28, no. 1: 27–35. [Google Scholar]
- Magaz, A. M. , Chorot P., Santed M. A., Valiente R. M., and Sandín B.. 2016. “Evaluación Del Bullying Como Victimización: Estructura, Fiabilidad y Validez Del Cuestionario de Acoso Entre Iguales (CAI).” Revista de Psicopatología y Psicología Clínica 21, no. 2: 77–95. [Google Scholar]
- Martínez‐Líbano, J. , Yeomans‐Cabrera M. M., Koch Serey A., Santander Ramírez N., Cortés Silva V., and Iturra Lara R.. 2025. “Psychometric Properties of the Trait Meta‐Mood Scale (TMMS‐24) in the Chilean Child and Adolescent Population.” Salud, Ciencia y Tecnología 5: 1376. 10.56294/saludcyt20251376. [DOI] [Google Scholar]
- Mayer, J. D. , Salovey P., and Caruso D. R.. 2004. “Emotional Intelligence: Theory, Findings, and Implications.” Psychological Inquiry 15, no. 3: 197–215. [Google Scholar]
- Moore, S. E. , Norman R. E., Suetani S., Thomas H. J., Sly P. D., and Scott J. G.. 2017. “Consequences of Bullying Victimization in Childhood and Adolescence: A Systematic Review and Meta‐Analysis.” World J Psychiatry 7, no. 1: 60–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Olweus, D. 1993. Bullying at School: What we Know and What we Can Do. Blackwell. [Google Scholar]
- Orth, U. , and Robins R. W.. 2014. “The Development of Self‐Esteem.” Current Directions in Psychological Science 23, no. 5: 381–387. [Google Scholar]
- Ramos, R. , Giménez A. I., Muñoz‐Adell M. A., and Lapaz E.. 2006. A‐EP: Cuestionario de Autoestima Para Educación Primaria. TEA Ediciones. [Google Scholar]
- Smith, P. K. , López‐Castro L., Robinson S., and Görzig A.. 2019. “Consistency of Gender Differences in Bullying in Cross‐Cultural Surveys.” Aggression and Violent Behavior 45: 33–40. [Google Scholar]
- Tsaousis, I. 2016. “The Relationship of Self‐Esteem to Bullying Perpetration and Peer Victimization Among Schoolchildren and Adolescents: A Meta‐Analytic Review.” Aggression and Violent Behavior 31: 186–199. [Google Scholar]
- UNESCO . 2019. Behind the Numbers: Ending School Violence and Bullying. UNESCO. [Google Scholar]
- Vargas‐Martínez, A. M. , Guillén‐Martínez D., Domingo Pérez T., et al. 2025. “Analysis of the Management Function of the School Nurse in Spain: Use of Records and Interaction With Primary Care, the Educational Community and Other Institutions.” Enfermería Clínica 35, no. 5: 502200. 10.1016/j.enfcli.2025.502200. [DOI] [PubMed] [Google Scholar]
- Yosep, I. , Hikmat R., and Mardhiyah A.. 2023. “School‐Based Nursing Interventions for Preventing Bullying and Reducing Its Incidence on Students: A Scoping Review.” International Journal of Environmental Research and Public Health 20, no. 2: 1577. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zafra‐Agea, J. A. , Domingo Pérez T., Lima‐Serrano M., et al. 2025. “Availability, Perceived Need and Satisfaction for School Nurses in Spain: A Cross‐Sectional Study Involving the School Community.” BMC Nursing 24: 1364. 10.1186/s12912-025-03970-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zych, I. , Farrington D. P., Llorent V. J., and Ttofi M. M.. 2019. “Protecting Children Against Bullying and Cyberbullying: The Contribution of Emotional Intelligence.” Child Abuse & Neglect 92: 140–149. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Exploratory posttest multilevel analysis of individual heterogeneity and discriminatory accuracy (maihda) across bullying, self‐esteem, and emotional intelligence outcomes.
Table S2: Sensitivity analysis of pretest–posttest changes accounting for classroom clustering.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.