2024
272
14102022
22022024
Agnès Ros Morente
Universidad de Lleida, España
Raquel Gomis
Universidad de Lleida, España
María Priego
Universidad de Lleida, España
Gemma Filella
Universidad de Lleida, España
Abstract: School violence has received increased interest from researchers during the last decades. According to the literature, violence in schools is related to various negative outcomes, such as a negative climate, decreased well-being of the students, or bad academic performance. Numerous studies have tried to detangle the underlying variables that can improve coexistence and prevent peer aggression and violence. Emotional competencies seem to constitute an interesting factor in this deed, increasing well-being and positive coexistence in class. The present study aims at analyzing the role of emotional competencies when preventing violence, as well as the role that anxiety plays in this relationship. The sample was constituted of 767 Spanish primary school students of 4th, 5th, and 6th grade from various public and private centers of a few northen Spain cities. Variables of emotional competencies, anxiety, and violence were gathered with self-informed questionnaires (CDE 9-13, STAI and CUVE-R). A pre-post design with experimental and control group was carried out. Results show that, as it was expected, emotional competencies are inversely correlated to violence. Also, anxiety plays an important role in the relationship between both factors, although no full mediation was found. Several interaction effects between pre and post evaluation were found in different variables, such as emotional awareness, social competence, life and well-being competencies and the global score of emotional competence. According to our results, students decreased their anxiety levels after intervention, in a significant way. Also, it was observed that higher levels of emotional competencies are directly related to lower levels of violence in post evaluation. All in all, interventions, like the one described in the present article, may be beneficial and show that the improvement of the management of emotions may be a good solution to prevent violent behaviors in school.
Keywords: school coexistence, violence, emotional competences, anxiety, school climate.
Resumen: Durante las últimas décadas, la comunidad científica ha mostrado interés en la investigación de la violencia escolar. De acuerdo con la literatura, la violencia en los centros escolares está relacionada con varios aspectos negativos, como son un clima de aula negativo, un menor bienestar de los estudiantes o unos malos resultados académicos. Varios estudios han intentado averiguar las variables que subyacen a esta relación, previniendo así, las conductas violentas entre los estudiantes. El presente estudio tiene como objetivo analizar el rol de las competencias emocionales para prevenir la violencia, así como el papel que juega la ansiedad. la muestra fue constituída por 767 alumnos de 4º, 5º y 6º de diferentes centros privados y públicos de ciudades del norte de España. Se recogieron datos de las variables de competencias emocionales, ansiedad y violencia con cuestionarios (CDE 9-13, STAI and CUVE-R). Se realizó un diseño pre-post con grupo control. Tal y como se esperaba, las competencias emocionales estan inversamente relacionadas con la violencia. Además, el grado de ansiedad juega un papel muy relevante en la relación. No se observó un efecto mediador de la ansiedad completo. Diferentes efectos de interacción fueron detectados entre los momentos pre y post en diferentes variables como fueron, consciencia emocional, competencia social o la competencia para la vida y el bienestar. De acuerdo con los resultados, los estudiantes reducieron los niveles de ansiedad después de la intervención de manera significativa. También, se observaron mayores niveles de violencia en aquellos individuos con menos manejo de sus competencias emocionales. Así pues, a modo de conclusion, los resultados parecen indicar que las intervenciones como la realizada en el presente estudio, podrían ser de gran beneficio y mostrar mejoras en el manejo de las emociones, previniendo, de este modo, las conductas violentas en los centros escolares.
Palabras clave: convivencia escolar, violencia, competencias emocionales, ansiedad, clima.
Estudios
An examination of the relationship between and intervention with emotional competences and violence. The role of anxiety
Análisis de la relación e intervención entre las competencias emocionales y la violencia. El papel de la ansiedad
Agnès Ros Morente
Raquel Gomis
María Priego
Gemma Filella
Recepción: 14 Octubre 2022
Aprobación: 22 Febrero 2024
Publicación: 28 Junio 2024
School violence and its prevention have become subjects of great international scientific interest, given the latest numbers referring to aggression. It is important to note that reports, like the one of UNESCO, in 2019, indicate that almost one in three students (32%) has suffered an episode of bullying.
School violence is related to many unhealthy outcomes, such as mental disorders symptoms (Fabbri et al., 2022; Gong et al., 2022; Kim et al., 2020) less academic achievement (Kim et al., 2020), or worsened psychologyical well-being (Fabbri et al., 2022). At a pedagogical level, there are also many negative consequences, such as, a bad school climate (Benbenishty et al., 2016). In this sense, studies show that the more school violence, the worse climate is observed in the school. This fact has several negative implications. For example, different studies have shown that when the school environment is negative, the centre becomes more vulnerable to situations of bullying (Wang et al., 2014) as well as to the social and emotional development of students (Lester et al., 2017), and academic performance is affected (Benbenishty et al., 2016).
Differently, a positive school climate, with good coexistence and a good classroom climate, promotes students’ motivation to learn, at the same time that it improves social skills (Curby et al., 2009; Wang et al., 2014). In addition to that, healthy friendship relationships and emotional support seem to promote a better learning process and higher academic achievement (Kim et al., 2020; Reyes et al., 2012).
When studying school climate, results at all ages point at the importance of variables such as emotional competencies or anxiety (Coyle et al., 2021; Gong et al., 2022). Studies show that these variables have a great impact in school violence, students’ well-being, and anxiety, among others (Ros-Morente et al., 2017). For this reason, they are of great interest in educational research at an international level, since an improvement in emotional skills could mean an improvement in other variables and in the school climate itself (Martínez-Sánchez, 2019). In addition, it is a fact that education of these skills helps the emotional development of students and good school performance (Nasir & Masrur, 2010).
Also, some studies have shown that emotional competencies could predict anxiety levels in students, as a predictor of the level of psychological well-being (Ros-Morente et al., 2017). Regarding the relationship of anxiety, school climate and school violence, there are studies that find significant relationships between students with high levels of anxiety and higher levels of aggressive behavior, pointing to emotional regulation as an explanation (Fernández-Sogorb et al. al., 2020). That is, those students that have more problems regulating their emotions will tend to use aggression to manifest their emotional state.
Thus, if emotional competencies are linked to anxiety levels (Ros-Morente et al., 2017) and anxiety levels are related to psychological well-being and more aggressive behaviors at school, and, as a consequence, to greater school violence (Fernández-Sogorb et al., 2020), one can assume that training the management of emotions will have a great benefit when promoting favorable school climate and reducing the rates of violence in the centers.
With this in mind, authors have proposed different training programs, which have been developed based on these emotional competencies. Its results, so far, are promising (Corcoran et al., 2018; Durlak et al., 2011; Taylor et al., 2017), showing improvements in emotional skills, prosocial behavior and behavioral problems of students, as well as in academic performance (Weissberg et al., 2015) and reducing the anxiety of the students (Keefer et al., 2018; Sánchez-Gómez et al., 2020).
When talking about emotional competencies, however, it is important to clarify and identify the theory which frames the different studies, since there is no general consensus on the conceptualizations, models and tools used to measure intellectual, educational and emotional skills (Pérez-González et al., 2020; Nelis et al., 2009).
This study is based on the emotional competencies model of the Psychopedagogical Orientation Research Group (GROP), which suggests five emotional competencies which can be trained to improve emotional skills and psychological well-being. These group of cometences are: emotional awareness, emotional regulation, social competence, emotional autonomy, and skills for life and well-being (Bisquerra and Pérez-Escoda, 2007). Emotional awareness refers to the skill to identify one’s emotions and the emotions of others. Emotion regulation is when a person can manage the emotions that is feeling at a certain moment. Emotional autonomy describes aspects related with self-esteem. Social competence concern prosocial behaviour and skills for the life and well-being describes the ability to function effectively in life (Bisquerra & Pérez-Escoda, 2007).
The objective of this study is to evaluate an emotional skills program developed for students from 4th to 6th grade) aiming to improve emotional skills, as well as reducing anxiety. The aim is to observe the relationship between school violence, emotional skills, and anxiety with a sample of school children in the fourth, fifth and sixth grades of primary education (8 to 12 years old) before and after an intervention to work on emotional competencies.
For all those reasons, it is hypothesized that after the intervention with the emotional education program and in the post-test evaluation, emotional competencies of students would improve. Also, it is expected a reduction in anxiety. Likewise, it is expected to observe a negative relationship between emotional competencies and the presence of school violence, also presenting an indirect relationship between them through anxiety.
The sample was comprised by 767 Spanish primary school students. Between the participants, 311, 192 and 265 were coursing 4th, 5th, and 6th grade, respectively, from various public and private centers of a few northern Spain cities (Lleida, Barcelona, Donosti, and A Coruña). The experimental group consisted in 616 participants and the control group was composed by 151 participants. The mean age was 9.92 (SD= .93) for the experimental group and 9.87 (SD = .87) for the control group. Related to sex, there were 292 girls and 324 boys in the experimental group and 85 girls and 66 boys in the control group. The sample was recruited using a convenience sampling technique, due to the prior agreement established with the schools that showed voluntary interest in participating in the project. Students answered a paper-and-pencil survey two times, before and after the intervention (pretest and post-test), with parents’ and teachers’ consent. The majority of schools expressed their willingness to be involved in the program, and a few of them agreed to participate in the control condition with the understanding that they could access the program for their own use after the study. Since we wanted to prioritize the experimental group, these conditions resulted in a difference in sample size between the two groups.
Emotional competencies. The 41-item Emotional Development Questionnaire (CDE 9-13, Pérez-Escoda et al., 2021) was used to measure the five emotional competencies developed by the GROP (Bisquerra & Pérez-Escoda, 2007): emotional awareness, emotion regulation, emotional autonomy, social competence, and life and well-being competences, as well as the global emotional competence. Items were measured on a 11-point Likert scale with 0 indicating never and 10 suggesting most of the time. Example items are: “I can easily describe my feelings” and “I get angry very often”. Related to reliability, for the present sample, each competence reported the following Cronbach’s α values: emotional awareness (Cronbach’s αpre= .80; αpost= .81;), emotional regulation (αpre= .74; αpost= .75), emotional autonomy (αpre= .65; αpost= .64), social competence (αpre= .73; αpost= .72) and life, well-being competences (αpre= .73; αpost= .70) and global emotional competence (αpre= .93; αpost= .92).
State anxiety. The 20-item subscale from the Spanish adaptation of the State-Trait Anxiety Inventory (STAI, Spielberger et al., 1970; Buela-Casal et al., 2016) was used to measure state anxiety. In the Spanish version, items were measure using a 4-point scale, that ranges from 0 (indicating not at all) to 3 points (indicating almost always). Example items are: “I find myself restless” and “I am worried”. Reliability and fit measures for the present sample were the following: αpre= .84; αpost= .85.
Presence of violence in school. The 31-item School Violence Questionnaire-Revised (CUVE-R, Álvarez-García et al., 2011) was used to measure the presence of different types of violence in the school environment. It comprises six factors: verbal violence from students towards students, verbal violence from students towards teachers, direct physical violence and threats between students, indirect physical violence by students, social exclusion and disruption in the classroom; as well as the total level of school violence. The questionnaire includes items measured on a 5-point Likert scale with 1 indicating never and 5 indicating always. Sample items include: “Some students hit their classmates on the school grounds” and “Some students neither work nor allow others to work”. The evaluation of this measure was performed only after the intervention, using the scale to explore the total level of school violence as a latent variable. This variable showed an adequate reliability, with a Cronbach’s α value of.89.
First, each member of the research group contacted schools that showed interest in participating Emotion Education Program. An educational agreement was formed between the university and the schools, having the consent from both teachers and parents. After securing their unanimous approval, both emotional competences questionnaire and state anxiety questionnaire were administered to children at two separated times (before and after the intervention). The questionnaire to evaluate school violence was administered only one time (after the intervention). International Helsinki recommendations were followed and institutional approval was also obtained from the relevant ethics committees to safeguard the identity and well-being of participants both during the data collection and the intervention. Before the intervention, teachers from both the experimental and control group schools did a training session to be able to distribute the questionnaires. Only the teachers of the experimental group underwent a 30-hour training for the intervention, explaining them the program in detail.
The intervention was conformed by two programs, one that involved 25 classroom activities integrated in the Emotional Education program developed by the GROP (López-Cassà, 2023), and other that implied the interaction with the video game Happy 8-12 (Filella et al., 2018). Both the Emotional Education program and the video game are focused on improving children’s emotional competences and are based in the emotional competences model of Bisquerra and Pérez-Escoda (2007), developed by the GROP. Throughout the entire course, the students, together with their teachers, engaged in various activities for approximately one hour per week. These activities were tailored to each course’s age group and involved an active and practical approach, including group dynamics, role-playing, discussions, relaxation exercises, and interactive games. The primary objective was to train and strengthen emotional competences, encouraging the active participation of all students.
To further reinforce emotional competences and promote assertive conflict resolution, the students used the video game “Happy 8-12,” specifically designed for this purpose. The game featured 25 conflicts set in familiar contexts such as school, home, or the playground. The conflict resolution process involved several steps: first, identifying the conflict; second, recognizing the emotions inspired by the conflict; third, selecting strategies to resolve the conflict; and fourth, identifying the emotions experienced after providing an answer. In the video game, students had the option to choose passive, aggressive, or assertive responses to conflicts. However, the game actively encouraged the selection of assertive responses by awarding higher points and offering additional mini-games.
Descriptive statistics were reported for each emotional competence (emotional awareness, emotional regulation, emotional autonomy, social competence, and life and well-being competences), the global emotional competence, state anxiety, and the presence of violence in school. Normality and homogeneity of variance were checked. Variables with a non-normal distribution were transformed using a two-stage approach (Templeton, 2011). To do so, total scores for each measure were used. To explore reliability, Cronbach’s α was reported, with values higher than .60 considered adequate (Taber, 2018).
A repeated measures mixed ANOVA was performed to explore changes related to the intervention, considering time (pretest and post-test) and group (experimental and control), using course and sex as covariates. Transformed values for each variable were introduced in the mixed model. Missing cases were previously eliminated from the analysis. Since the sample size for each group was unbalanced, and this condition could be a potential source of bias, between-subject differences for the group variable were examined using a t-test, to assess possible pretest disparities.
Afterward, a multigroup mediation analysis was carried out to explore the relationship between global emotional competence at Time 1 (pretest) and the presence of violence at Time 2 (post-test), mediated by anxiety state at Time 2. The analysis evaluated both groups to assess possible differences between the variables’ links. As both the exogenous and endogenous variables were composed of several subscales, the total scores for each subscale were used to create our latent variables. Anxiety was included in the model as an observed variable, using its total score as well.
Model fit was examined using the Chi-square test (χ2), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standard Root Mean Square Residuals (SRMR). CFI and TLI values above .90 and RMSEA and SRMR below .08 were considered adequate (Hu and Bentler, 1999). The data analysis was implemented using IBM SPSS Statistics (Version 26) and R (R Core Team, 2022).
First, in Table 1, descriptive statistics for each emotional competences, the global emotional competence and state anxiety were reported for time (pretest and post-test) and group (experimental and control). The measure related to the total presence of violence in school as well as their subscales were reported during post-test.
Next, to explore the effect of the program, examining differences in group, time, and the interaction between group and time, a repeated measure mixed ANOVA was performed using previously normalized variables. Most of the variables reported significant Kolmogorov-Smirnov and Shapiro-Wilk values, except for emotional regulation, with negative skewed distributions. As the sample size was unbalanced between the experimental and the control group, differences related to group were explored by performing a t-test. There were no group differences in most of the variables at the pretest, except for life and well-being competences (t (765) = -2.29, p = .023), with the experimental group reporting lower values.
Subsequently, considering these previous differences, results for the repeated measures ANOVA were explored. Related to the main effect for the variable time, results showed a significant main effect of time in state anxiety (F₍₁,₇₆₀₎ = 4.02, p = .045, ηp² = .005), with a decrease in the mean from pretest to post-test. Nonetheless, no main effect on time was detected in none of the emotional competences: emotional awareness (F₍₁,₇₆₀₎ = .88, p = .348, ηp² < .001), emotional regulation (F₍₁,₇₆₀₎ = .51, p = .475, ηp² < .001), emotional autonomy (F₍₁,₇₆₀₎ = .01, p = .915, ηp² < .001), social competence (F₍₁,₇₆₀₎ = .03, p = .860, ηp² < .001), life and well-being competences (F₍₁,₇₆₀₎ = .01, p = .906, ηp² < .001) and the global emotional competence (F₍₁,₇₆₀₎ = .07, p = .789, ηp² < .001).
Related to the variable group, no significant effect was found for all the considered variables: emotional awareness (F₍₁,₇₆₀₎ = .14, p = .705, ηp² < .001), emotional regulation (F₍₁,₇₆₀₎ = .82, p = .365, ηp² < .001), emotional autonomy (F₍₁,₇₆₀₎ = .65, p = .418, ηp² < .001), social competence (F₍₁,₇₆₀₎ = .00, p = .928, ηp² < .001), life and well-being competences (F₍₁,₇₆₀₎ = .16, p = .689, ηp² < .001), the global emotional competence (F₍₁,₇₆₀₎ = .04, p = .837, ηp² < .001) and state anxiety (F₍₁,₇₆₀₎ = 3.7, p = .053, ηp² = .005).
In contrast, significant interaction effects between group and time were found for emotional awareness (F₍₁,₇₆₀₎ = 18.86, p < .001, ηp² = .024), social competence (F₍₁,₇₆₀₎ = 12.93, p < .001, ηp² = .017), life and well-being competence (F₍₁,₇₆₀₎ = 23.46, p < .001, ηp² = .030) and the global emotional competence (F₍₁,₇₆₀₎ = 21.69, p < .001, ηp² = .028). This suggested that a crossover effect may have occurred. Taking a look at Figure 1 and the descriptive statistics in Table 1, it can be observed that the experimental group had an increment in emotional awareness, social competence, life and well-being competences and global emotional competence, while the control group decreases or maintains their levels.
Considering our covariables, sex between-subjects differences were found for emotional awareness (F₍₁,₇₆₀₎ = 5.56, p = .019, ηp² = .007) and social competences (F₍₁,₇₆₀₎ = 4.33, p = .038, ηp² = .006) with girls showing higher values. Course differences were also found for emotion regulation (F₍₁,₇₆₀₎ = 20.06, p < .001, ηp² = .026), emotional autonomy (F₍₁,₇₆₀₎ = 11.29, p < .001, ηp² = .015), life and well-being competences (F₍₁,₇₅₉₎ = 25.01, p < .001, ηp² = .032) and the global emotional competence (F₍₁,₇₆₀₎ = 12.75, p < .001, ηp² = .017), with higher values for children that were coursing 4th grade, with the 6th grade reporting the lowest values. Although there are differences related to these covariables, neither course (F₍₁,₇₅₄₎ = 1.31, p = .242, ηp² = .012) nor sex (F₍₁,₇₅₄₎ = .84, p = .553, ηp² = .008) interacted with time showing no within-subjects effects.
The next step was to evaluate if there is any relationship between the previous level of global emotional competence and the presence of violence at the post-test, evaluating a possible mediation through pretest and post-test state anxiety. First, bivariate correlations were checked.
Table 2 shows negative correlations between global emotional competence at pretest, and state anxiety and presence of violence at post-test. Positive correlations were found between post-test state anxiety and presence of violence. Pretest state anxiety didn’t report significant relationships except for post-test state anxiety, so it wasn’t included in the multigroup mediation model (Figure 2).
The model reported good fit measures: χ² = 274.65, df = 98, p < .001, CFI = .96, TLI = .95, SRMR = .03, and RMSEA = .07(.06, .08). As it can be seen in Figure 2, the level of global emotional competence at pretest was negatively related with the presence of violence at posttest in both control (β = –.33, p = .001) and experimental group (β = –.12, p = .004). The global emotional competence was also negatively related with post-test state anxiety for both control (β = –.41, p < .001) and experimental group (β = –.39, p < .001). The model reported small and significant R² for state anxiety (R²ₑₓₚ = .16, p < .001; R²꜀ₒₙₜᵣₒₗ = .16, p < .001) and for presence of violence (R²ₑₓₚ = .02, p < .001; R²꜀ₒₙₜᵣₒₗ = .12, p < .001). No significant relationship between state anxiety and presence of violence at post-test was found. Also, the global emotional competence didn’t show indirect effects on the presence of school violence through state anxiety neither for the experimental (β = –.01, p = .476) or the control group (β = –.02, p = .557). In conclusion, the global emotional competence is negatively associated with the presence of future school violence in both groups as well as with anxiety.
The increasing rates of violence in schools have become seriously worrying due to the negative effects they have in the mental health, well-being and academic achievement of students (for example, Coyle et al., 2021; Gong et al., 2022; Kim et al., 2020). For this reason, the subject of violence and aggression in schools has been addressed by scientists during the latest decades. Evidence has repeatedly proved that, contrary to what was initially assumed, scholar violence is related to emotional variables and its management and not with hostility of the participants in conflicts or situations (Pons et al., 2005; Ros-Morente, 2017).
Prior studies have identified variables that constitute risk factors for scholar violence, such as, school climate (Wang et al., 2014). At the same time that, certain variables can be identified as prevention variables, especially when trained in early stages of life. Take as an example emotional competences or emotion regulation (Corcoran et al., 2018; Durlak et al., 2011; Taylor et al., 2017; Ros-Morente et al., 2017), which trained properly and with educational programs, can prevent conflict and violent situations to a great extent (Ros-Morente et al., 2017). This is also the case, for instance, of anxiety (Martínez-Monteagudo et al., 2011) or school climate (Benbenishty et al., 2016).
Given the aforementioned previous findings, the aim of this work was to evaluate a program of emotional competences specifically developed for primary school students and its effects in anxiety (as an indicator of psychological well-being) and scholar violence.
According to our results, students decreased their anxiety levels after intervention, in a significant way. However, no differences were observed for emotional competences at first sight. When taking time into account (pre and post evaluations), emotional competences seemed to importantly increase, such as competences of well-being, social competence and global emotional competence. Although this may seem counterintuitive, previous studies have shown that anxiety state is the variable that is faster to show changes and the most responsive one to interventions and that some of the competences are more resistant to change (e.g., emotion regulation) (Ros-Morente, 2017). For this reason, results were not surprising, and even, expected.
However, it is important to note that, interestingly and as it was mentioned above, there were several interaction effects between group (experimental vs. control) and time (pre vs. post) for different competences, such as, emotional awareness, social competences, life and well-being competences and the global score of emotional competences. In all of these cases, when studying the behavior of competences improvement along time, different changes for the students that received the training and for those who did not show it were observed. In this sense, the level of competence for those students that received the training exponentially increased, while those students that underwent no intervention, showed a decrease in the aforementioned competences. Although there is not a lot of literature regarding the matter of stability of competences among time, it could be hypothesized that those competences that are not trained, not only do not improve, but they can also become less adaptative with time and that interventions would benefit from having more time at their disposal (Collie, 2020; Pekrun, 2009).
Regarding the association of emotional competences and violence, according to our results, higher levels of emotional competences, that is, being more competent emotionally, is directly related to lower levels of violence in the future This is coincident with existing literature, which proves that those students that can manage their emotions better and that show higher levels of assertiveness also exhibit less violence with their peers and teachers in school (Röll, Koglin, & Petermann, 2012).
All in all, this article shows the importance that emotional competences have when it comes to wellbeing (anxiety) and violence. Interventions that improve the management of emotions and emotional competences seem to offer a good solution to prevent violent behaviors in the scholar context. However, more studies are needed in order to find out the behaviour that emotional competences have along time, in order to tackle the interactions, mediations and other pathways towards a better climate and wellbeing of students, with an environment free of violence.
How to reference this article: Ros Morente, A., Gomis, R., Priego, M., & Filella, G. (2024). An examination of the relationship between an intervention with emotional competences and violence. The role of anxiety. Educación XX1, 27(2), 283-299. https://doi.org/10.5944/educxx1.35828
redalyc-journal-id: 706
Esta publicación es parte del proyecto de I+D+i / ayuda RTI2018-098294-B-I00/ PRE2019-091604, financiado/a por MCIN/ AEI/10.13039/501100011033/ FEDER “Una manera de hacer Europa”.
