This chapter examines how safe, positive school environments and supportive relationships promote the full range of social and emotional skills, albeit with different aspects supporting different skills. The chapter starts with students’ sense of belonging and emotions at school, followed by teacher-student and peer-to-peer relationships, including how teachers’ coping strategies relate to students’ social and emotional learning. The chapter then addresses bullying, revealing unique data on perpetration versus victimisation. It explores how bullying manifests differently across participating countries and subnational entities and key strategies to tackle this.
Nurturing Social and Emotional Learning Across the Globe
Findings from the OECD Survey on Social and Emotional Skills 2023
Abstract
Policy insights
Results from the Survey on Social and Emotional Skills (SSES) 2019 and 2023 suggest the following recommendations for policy makers and practitioners to leverage school climate as a source of healthy social and emotional development and skills:
Build schools into hubs of community: Students aged 15 with a greater sense of belonging at school reported higher social and emotional skills, particularly sociability, emotional regulation skills and trust. Similarly, both 10 and 15-year-old students who experienced more positive and fewer negative emotions at school demonstrated higher skills, especially optimism and emotional control. To improve students’ social and emotional skills, schools should focus on nurturing a sense of community. Evidence-based, contextualised, whole-school approaches to improving school environments can help foster a wider sense of belonging and strengthen students’ skills.
Address site-specific sources of negative emotions: Students report different mixes of emotions in each site, especially variations in confidence, motivation, anxiety and anger. Policymakers and educators must examine the particular experiences that students report in their site, to develop strengths and address sources of excessive negative emotions. Holistic approaches to teaching and school environments alongside explicit social and emotional learning can boost students’ skills, development and perceptions of school.
Improve school experiences for vulnerable groups: Low academic performers in most sites report less sense of belonging at school, fewer positive and more negative emotions, especially anger, compared to their advantaged peers. Fifteen-year-old girls compared to boys show the same pattern, but with especially low confidence. Socio-economically disadvantaged students report less belonging. Yet results vary considerably by site and suggest that these inequalities are not inevitable outcomes. To address such disparities, schools must include targeted efforts to promote positive experiences and belonging for struggling groups, especially low performers and girls.
Promote relationships to promote skills: Students aged 10 and 15 who report better relationships with their teachers demonstrate higher skills, especially higher task performance and open-mindedness skills. In contrast, stronger peer-to-peer relationships correlate in particular with better social skills and better trust across both age groups. However, 15-year-olds perceive less personal concern from their classmates and teachers compared to 10-year-olds. To address this, schools should create opportunities for students to form meaningful relationships with teachers and classmates, fostering complementary skill sets. Special attention should be given to promoting relationships in secondary schools.
Enhance teachers’ coping strategies to support well-being and model healthy emotional regulation: The use of effective strategies for managing work stress varies significantly among teachers across different sites, ranging from 68% in some sites to as low as 15% in others. Proven effective approaches, such as approaching problems optimistically and practising good health habits, can enhance teachers’ own capacities and skills, particularly in emotional regulation. Notably, teachers in Kudus (Indonesia) and Peru reported the highest use of these proven skills and reported better well-being. To improve teachers’ social and emotional skills, schools should focus on reducing teachers’ work-related stress and promoting the use of effective coping strategies. By doing so, teachers can improve their well-being and become better role models for positive emotional regulation in the classroom, creating a more supportive learning environment for students.
Prioritise creating safe and secure school environments for student success: Bullying remains a significant concern across all sites, with some areas facing particularly high levels. Bulgaria, Delhi (India) and Kudus (Indonesia) report the highest incidence of bullying. Notably, in these sites, school leaders report very low levels of bullying, suggesting a concerning normalisation of such behaviours. To address this issue, sites with significant bullying problems should urgently implement comprehensive anti-bullying strategies. These approaches should target all relevant stakeholders, including school staff, students and parents, and draw on evidence-based effective practices. However, creating and maintaining safe school environments is crucial for all education systems, not just those with high bullying rates. Such environments support students’ academic and socio-emotional development, as well as their overall health and well-being.
Tailor anti-bullying interventions to local contexts: The nature and prevalence of bullying vary significantly across different sites, necessitating context-specific approaches. In certain sites, a large proportion of students are involved in bullying, with Delhi (India) having the highest involvement. The roles of perpetrator and victim often overlap: in Delhi (India) and Bulgaria, most bullying perpetrators also report being victims, while in Chile and Peru, only a minority of students report this dual role. The type of bullying also differs across sites: in Gunma (Japan) and Peru, social or verbal bullying is much more common, with physical bullying being rare. Contrastingly, in Bulgaria, physical bullying is almost as common as social or verbal bullying. These variations can significantly impact intervention effectiveness. Therefore, anti-bullying strategies should be carefully tailored to local contexts. For instance, in sites where most students involved in bullying are both victims and perpetrators, approaches targeting specific roles may be less effective. In areas with very high bullying rates, whole-school interventions addressing broader school climate issues are likely to be more beneficial than resource-intensive individual interventions.
Children learn, socialise and navigate their emotions better in safe, supportive environments, and environments become safer and more supportive when children and adults are able to manage their emotions and socialise well (Jones, McGarrah and Kahn, 2019[1]; Cefai et al., 2018[2]; Charlton et al., 2021[3]). As a social species, human brains are wired to prioritise emotions and social dynamics above academic learning (Rogers and Thomas, 2023[4]), which may be particularly true for adolescents (Chatterjee Singh and Duraiappah, 2020[5]). Positive, safe school environments (where students feel physically, socially, emotionally and intellectually safe, accepted and supported) are therefore essential for all types of learning.
The results from the SSES 2019 and SSES 2023 demonstrate how different elements of the school environment from teacher-student relationships to bullying, play different roles in students’ social and emotional development. It illustrates how building schools’ sense of community and confronting the complexity of bullying are essential for fostering students’ socio-emotional skills. However, equity remains a problem. Girls and low academic performers report a lower sense of belonging and fewer positive school experiences than their peers across sites, while socio-economically disadvantaged students report less belonging.
Holistic approaches that address multiple elements necessary for creating safe and supportive school environments will likely work best to support students’ full range of skills. These approaches include:
building schools’ sense of community and students’ positive experiences
promote relationships and teachers’ coping strategies in order to develop skills
creating safe and secure school environments.
The data suggest that sites vary in their ability to realise the nurturing components of school environments. Moreover, tackling equity issues within sites could help level out differences in skill levels observed for different student groups (e.g. advantaged versus disadvantaged students, boys versus girls). School climate and students’ skills can form a virtuous circle: healthier school environments promote social and emotional skills, while stronger skills, in turn, contribute to better environments (Osher and Berg, 2018[6]).
Students who are happy and feel part of a positive, supportive community show greater cooperation and readiness to learn (Cantor et al., 2019[7]). They also dare to express their wishes and opinions and find it easier to cope with failure, anxiety or stress (Chatterjee Singh and Duraiappah, 2020[5]).
These findings adds to research that suggests that a positive, safe school environment is crucial for students’ social and emotional development and skills for all students (Rogers and Thomas, 2023[4]; Center on the Developing Child, 2009[8]; Green and García-Millán, 2021[9]).
This section examines the SSES 2019 and SSES 2023 data on students’ holistic experiences of their school environment. It shows how building schools’ sense of community, enhancing students’ positive experiences and addressing sources of negative experiences can bolster their social and emotional development. For example, students who felt greater belonging at school reported being better able to engage with others, while students with more positive emotions at school reported better emotional regulation. However, students are not benefiting equally and the specific mix of students’ emotions varies by site. Low academic performers and girls report a lower sense of belonging and more negative emotions than their peers. In some sites, students struggle with motivation, in others, anger or anxiety. Both whole-school and targeted support are needed to build an inclusive, positive school community for all.
Students who feel they belong and have more positive emotions at school report higher social skills and better emotional regulation
Feeling included and valued and that school is a place of positive experiences and healthy challenges are key to a nurturing school environment and academic learning (Ibarra, 2022[10]; Rogers and Thomas, 2023[4]; Cantor et al., 2019[7]). When students feel connected and safe, “they can build social capital and more readily use adults as social models, accept feedback, and navigate and persevere through challenges” (Osher and Berg, 2018, p. 5[6]). Thus, students’ sense of belonging and the emotions they have while at school can be considered good indicators of a school’s overall climate.
In SSES 2019 and 2023, students reported on their sense of belonging at school. In 2023, they also reported on the emotions they felt at school. For the former, students were asked whether they felt like they belonged at school, felt liked or lonely, made friends easily or felt left out. For the latter, students were asked how often they had experienced specific positive and negative emotions at school in the past four weeks, such as confidence, happiness, anxiety or anger.
Students who felt greater belonging, more positive emotions and fewer negative emotions at school reported higher levels of all socio-emotional skills, on average across all sites. Figure 3.1 shows the strength of the relationships between students’ skills and their sense of belonging, positive emotions and negative emotions while at school for 2023 sites (see also Tables A3.3, A3.9, A3.15).
Standardised regression coefficients of individual skills on the 15-year-old students’ sense of belonging and emotions at school indices, average across sites
Note: Significant average coefficients at a threshold of p < 0.05 are coloured, non-significant coefficients are outlined Models control for gender, educational achievement, socio-economic and migrant status. See Annex A for information about how the indices were calculated. Social and emotional skills are listed in order of the size of effect of students’ sense of belonging.
Source: OECD, SSES Databases (2023) Tables A3.3, A3.9 and A3.15
SSES findings for 15-year-olds suggest that sense of belonging and positive emotions at school each foster particular skills. Belonging promotes emotional regulation and social skills, such as sociability, optimism, energy and trust. Positive emotions and experiences at school (e.g. enjoyable social interactions, greater achievements) promote optimism and energy, but they also relate strongly to achievement motivation, persistence and curiosity. These are key skills for academic success, even when controlling for academic achievement, gender and socio-economic or migrant status (see Figure 3.1 Note).
Yet SSES data also show the harm of too many negative experiences. 15-year-olds students with more negative emotions at school report especially low levels of emotional control, stress resistance, optimism and energy—some of the same skills needed to build belonging. This finding may be particularly important for adolescence. This is a developmentally malleable period, but positive affect and mental health also typically decline (Goldbeck et al., 2007[11]; Marquez et al., 2024[12]; OECD, 2024[13]). Improving students’ experiences at school (both social and academic) may help mitigate this, while nurturing key skills for their adult lives.
The strength of the relationships between skills, belonging and emotions at school vary by site, although significant relationships emerged for all measures and skills in almost all sites, except for empathy, assertiveness and negative emotions. For example, the relationships between positive emotions at school and skills are at least twice as strong in Jinan (China) as they are in Delhi (India) for all skills except empathy and tolerance (Table A3.9). Additionally, sense of belonging dropped from 2019 and 2023 in Bogotá (Colombia) and Helsinki (Finland), the two sites that participated in both rounds of SSES (see Box 3.3). These variations between sites suggest that, while experiences at school impact students’ social and emotional skills, this interacts with factors outside school—and in some sites more than others.
Students’ emotions at school affects their skills at both ages 10 and 15, for almost all skills in all sites that assessed both cohorts. Additionally, there was no significant shift from ages 10 to 15 in the relationships between skills and negative or positive emotions in most sites. Creativity, curiosity and empathy were more related to positive emotions at age 10 than 15 in most sites. In a few cases, though, the reverse. The relationship between trust and negative emotions was stronger at 15 than at 10, on average and in half of sites (Tables A3.15). Energy and positive emotions also showed a stronger relationship at age 15 (Table A3.9).
By contrast, age differences in the relationship between belonging and skills depended on the site. For example, in 2019 Houston (United States) showed no significant age difference for any skill except curiosity, while Suzhou in 2019 (China) showed a stronger relationship at age 10 compared to 15 for almost all skills (Table A3.3). This indicates that the secondary school environments in Houston (United States) are able to nurture belonging and thereby promote skills just as much as their primary schools, while these effects fade in Suzhou (China).
These results suggest that school environment matters for all ages, but its influence changes from childhood to adolescence and some systems are leveraging its potential more than others. Positive experiences in school may particularly improve adolescents’ energy and capacity for trust. These skills are key skills for well-being and life satisfaction (OECD, 2024[13]). For sense of belonging, sites differ in their ability to maintain its influence at age 15, although it relates to skills at both ages and in all sites. On the one hand, the varying sizes and structures of secondary schools may interact with different expectations of adolescence in each site (e.g. levels of independence, whether extracurriculars occur inside or outside schools). The social role of school may decline in some sites but not others. On the other, the SSES shows that outcomes are not guaranteed. Some systems can do more to strengthen their secondary schools’ capacity to nurture belonging and skills.
SSES data demonstrates that school environment can significantly promote skills necessary for academic, social and personal success. Schools’ capacities to build a sense of community are important levers. Holistic approaches like integrating positive experiences into daily school life and creating opportunities for social connection should supplement structured programmes, like Aulas en Paz (Box 3.1).
Aulas en Paz (“Classrooms in Peace”) is a multi-component, evidence-based programme that seeks to promote peaceful coexistence and reduce aggression in primary school environments, especially in vulnerable communities with histories of violence and substance abuse. Inspired by international programmes and designed by the University of the Andes Colombia, it includes: 1) a universal curriculum to develop citizen competencies; 2) parent workshops and home visits to parents of children with the greatest need; and 3) positive reinforcement in extra-curricular peer groups, each with two children who present aggressive behaviours. It has demonstrated effectiveness in children ages 7-10 in Colombia. Activities seek to promote socio-emotional competencies, such as empathy, anger management, creative generation of alternatives and healthy assertiveness.
Aulas en Paz deliberately leverages the school community by partnering with teachers, families and students. It builds on a tradition of peace education in Colombia that seeks to reconcile after decades of civil war and to reduce community and domestic violence. Exposure to violence is related to higher levels of aggression and emotional dysregulation in children. Hence, Aulas en Paz seeks to break the cycle by supporting children, schools and families in building healthy alternative strategies.
Since 2008, Aulas en Paz has been implemented in 17 departments in Colombia as well as in Chile, Mexico and Peru. It has demonstrated effectiveness even in difficult conditions. A two-year, quasi-experimental evaluation with 1 154 students from 55 classrooms in 7 Colombian public schools located in neighbourhoods with youth gangs and drug cartels found improved prosocial behaviour and assertiveness as well as reduced aggression and verbal victimisation. This was despite partial implementation, as half of the activities were not used. To date, in Colombia, the programme has reached 9 112 teachers, 266 450 students between second and fifth grades, and at least 20% of the families of these students.
Its low cost has helped make Aulas a competitive option in low- and middle-income countries.
Source: Universidad de los Andes Colombia (2023[14]), “Aulas en Paz”, https://imagina.uniandes.edu.co/hub/aulas-en-paz/; Chaux et al. (2017[15]), “Classrooms in peace within violent contexts: Field evaluation of Aulas en Paz in Colombia”, https://doi.org/10.1007/s11121-017-0754-8.
Inequality in students’ sense of belonging and positive emotions at school is a consistent problem in some sites, while it only affects some groups in other sites
A key challenge of education and policy making is to improve equity and ensure all children can learn and thrive. It is concerning, therefore, that the SSES 2023 shows that different student groups do not benefit equally from a nurturing school environment. Figure 3.2 shows how, depending on the site, 15-year-old students’ sense of belonging, positive emotions and negative emotions at school vary widely by gender, socio-economic status and educational achievement.
Some sites show consistent, significant differences the three measures, such as Chile, Helsinki (Finland), Jinan (People’s Republic of China, hereafter “China”) and Spain. In almost all these sites, girls, low academic performers and socio-economically disadvantaged students report significantly less belonging and fewer positive emotions. Girls and low academic performers also report significantly more negative emotions (Tables A3.2, A3.5 and A3.11).
In contrast, some sites show no differences between some groups. For example, in Delhi (India) and Peru, there are no significant differences between disadvantaged and advantaged students, and Delhi (India) shows no gender differences in positive emotions or belonging. In Turin (Italy) and Bogotá (Colombia), there are no disparities between high and low academic performers in belonging and positive emotions.
Importantly, however, no site achieves equity for all groups in all three measures (belonging, positive emotions and negative emotions).
Differences in 15-year-olds’ sense of belonging, positive emotions and negative emotions at school indices, by gender, socio-economic status and educational achievement
Note: Significant average differences at a threshold p < 0.05 are coloured, while non-significant average differences are outlined. High and low academic performers are students respectively in the top or bottom quarter of educational achievement in their site. High/low performers in Gunma (Japan) and Kudus (Indonesia) cannot be defined because of the grades distributions. Advantaged and disadvantaged students are students respectively in the top or bottom quarter of the index of economic, social and cultural status (ESCS) in their site. Sites are listed in alphabetical order.
Source: OECD, SSES 2023 Database Tables A3.2, A3.5 and A3.11.
Students in Spain and Ukraine feel more confident but less motivated, while students in Italian sites feel more anxious
Students’ perceptions and experiences at school affect their development, including skills. As Figure 3.1 shows, higher levels of negative emotions are linked to lower levels of all social and emotional skills for both age groups, particularly emotional regulation. SSES data also shows that students’ emotions vary by site. It also shows wide variation, up to 30 percentage points (Table A3.4). Policy makers and education leaders should examine the composition of students’ emotions in their systems to identify specific needs and skills to target.
In SSES 2023, students were asked how often they had felt various positive or negative emotions at school in the last four weeks, such as confident, motivated, happy, anxious or angry. Students rated frequency with statements from “never or almost never” to “about half the time” to “all or almost all of the time” (see Tables A3.4, A3.10). Figure 3.3 shows results for four emotions, with the percentage of 15-year-olds in each site that reported feeling that emotion over half the time.
This varying mix of students’ feelings suggest different underlying causes, as well as different needs. Other SSES data may help illuminate. For example, high levels of anger in Bulgaria could partly be explained by high levels of bullying, especially physical bullying, compared to other sites (see below). Students in Gunma (Japan) have the highest test and class anxiety of all sites (OECD, 2024[13]), which may help explain their low confidence and high anxiety (24% report feeling anxious most of the time, see Table A3.10).
Cultural differences likely also affect how students report on their own emotions. For example, students in Helsinki (Finland) and Jinan (China) report low levels of positive emotions, but they also report low levels of negative ones. In comparison, students in Bogotá (Colombia) and Kudus (Indonesia) report high levels of confidence, motivation and happiness, but also relatively high levels of anxiety and anger (Tables A3.4, A3.10). These results indicate different attitudes towards reporting intense emotions.
Policy makers and education leaders need to consider emotions prevalent in their schools to better identify and address relevant issues in the school environment, such as bullying or stressful exams. For example, students in Gunma (Japan) might benefit from approaches that address sources of anxiety, while Spain and Ukraine might focus on boosting student engagement through new teaching approaches. This can also help improve social and emotional skills. Approaches should be tailored to site needs, combining holistic efforts to address students’ experiences—such as anxiety or motivation—with explicit social and emotional learning. Social and emotional learning is more effective when it is both explicit and properly contextualised (Yaeger, 2017[16]; Jones et al., 2021[17]; Durlak et al., 2011[18]).
Percentage of 15-year-old students who reported they have felt the following ways more than half the time while at school
Note: Sites are listed in descending order of the percentage of students who reported they have felt motivated more than half the time.
Source: OECD, SSES 2023 Database Tables A3.4 and A3.10
Low performers struggle more with anger at school, girls with confidence, but no pattern emerges for disadvantaged students
SSES data reveal that the emotions of particular groups also need attention, because they do not always match expectations or population patterns. This is notably true for anger and confidence. Figure 3.6 shows the percentage differences between two groups for two emotions: on the left, differences in anger between low and high performers and, on the right, differences in confidence between boys and girls.
Percentage of students who reported they felt either angry or confident more than half the time, by academic achievement (left) and gender (right), by site
Note: Only differences that are statistically significant at a threshold of p<0.05 are noted next to site names. High/low performers refer to students scoring in top/bottom quarter in math and reading distribution in their own sites. High/low performers in Gunma (Japan) and Kudus (Indonesia) cannot be defined because of the grades distributions. In Ukraine and Spain, the percentages for high performers is missing due to the small sample size (fewer than 30 students and 5 schools). Sites are listed in alphabetical order.
Source: OECD, SSES 2023 Database Tables A3.4, A3.6, A3.10 and A3.14
In most sites, low performers report feeling significantly angrier compared to high performers, but not more anxious, less happy or less motivated (Tables A3.8, A3.14). On average, 19% of low performers report feeling angry most of the time on average, compared to 12% of high performers. However, this varies by site. In Delhi (India) and Sobral (Brazil), 29%—almost one-third—of low performers feel angry most of the time. Jinan (China), Peru, Spain and Ukraine report the lowest levels of anger among low performers, at 11-13%.
In contrast, girls struggle with lack of confidence in all sites except one, but not with interest in school (Table A3.6). For example, only 32% of girls report feeling confident more than half the time on average, compared to 47% of boys. These differences were also present at age 10 in most sites that surveyed that age, but the gap widens at age 15 in all sites except Ukraine. Again, however, some sites struggle more than others. Dubai (United Arab Emirates), Helsinki (Finland), both Italian sites, and Spain present the largest gender gaps at age 15—over 22 percentage points. Delhi (India) is the only site with no significant gender difference in confidence at age 15.
Yet disparities in emotions disappear when comparing socio-economically advantaged and disadvantaged students. Most sites show no significant differences in either positive or negative emotions between these groups. Only feeling interested showed a slight significant difference in half of sites, favouring advantaged students. Yet some sites show the opposite pattern. In Bogotá (Colombia), Chile and Emilia-Romagna (Italy), for example, advantaged students reported higher levels of anxiety than disadvantaged ones (Table A3.13).
These findings show that policymakers and educators must not impose homogenous solutions on students. Equity requires adapting approaches not just to context, but groups within that context. This matters for social and emotional learning. Universal programmes that have positive effects for some groups can have null or adverse effects for others (Daley and McCarthy, 2021[19]; Rowe and Trickett, 2018[20]). Effective social and emotional learning must balance whole-school expectations with the needs of different students. Designing initiatives explicitly for equity can help (Jones et al., 2021[17]). This means incorporating student voice in programme design (Cefai et al., 2018[2]) and using contextualised, inclusive approaches that affirm student identities and address their lived experiences (Yaeger, 2017[16]; Cantor et al., 2019[7]; Wigelsworth, 2016[21]). This can boost students’ engagement and intervention effectiveness, especially for adolescents (Yaeger, 2017[16]).
Strengthening schools’ sense of community and increasing positive experiences can help foster students’ social and emotional skills – as well as their academic development. Whole-school approaches to social and emotional learning can enhance both school environments and students’ skills even better than isolated lessons, when they are effectively implemented (Cefai et al., 2018[2]; Jones et al., 2021[17]). However, any intervention must be contextualised and respond to students’ lived experiences, including those of particular groups (e.g. low performers, girls) (Denham, 2018[22]; Yaeger, 2017[16]; Jones et al., 2021[17]).
Addressing sources of negative emotions in each site, such as low engagement or a culture of stressful exams, can improve school communities and students’ skills. Integrating pedagogy for engagement, appropriate student agency and culturally relevant material can enhance motivation, belonging and positive emotions at school (Gutman and Schoon, 2013[23]; Rogers and Thomas, 2023[4]). These should be paired with explicit social and emotional learning that targets skills relevant to students’ experiences, such as emotional regulation or assertiveness (Cefai et al., 2018[2]; Durlak et al., 2011[18]). Delhi’s (India) Happiness Curriculum (OECD, 2024[13]) or the United Kingdom’s .b Mindfulness programme are two examples (Clarke et al., 2015[24]).
Finally, low-stakes, convenient monitoring and assessment platforms, like Chile’s Diagnóstico Integral de Aprendizajes (DIA) (see Chapter 2), can encourage schools to engage with this topic, track their progress and tailor their efforts.
Relationships are perhaps the most critical component of a nurturing school environment. As Chatterjee Singh and Duraiappah put it, “When children feel comfortable with their teachers and peers, they are more willing to grapple with challenging material and persist at difficult learning tasks” (2020, p. 75[5]), including social and emotional learning. According to the SSES data, schools can promote skills through promoting quality relationships at school. Although all good relationships can promote skills, improving students’ relationships with teachers seems more important for skills related to learning, innovation and achievement. Students’ relationships with their peers, on the other hand, relate more to sociability and emotional regulation skills. However, sites vary in their ability to promote relationships, especially in secondary schools, where students perceive less care and concern from teachers and classmates.
Students who have better relationships with their teachers are more open-minded, motivated and persistent
Students’ relationships with their teachers and adults at school affect numerous student outcomes (Poulou, 2017[25]; Cantor et al., 2019[7]), including social and emotional outcomes (Cefai et al., 2018[2]; Wigelsworth et al., 2022[26]). This is also true for adolescents (Yaeger, 2017[16]).
In SSES 2023, students were asked whether their teachers were respectful, friendly, or mean and if they showed concern for students. Figure 3.5 shows the strength of the relationship between 15-year-old students’ relationships with their teachers and peers, and their social and emotional skills.
Students with better relationships with their teachers reported higher levels of all social and emotional skills, with particularly strong associations in achievement motivation, persistence, curiosity and optimism. Notably, this correlation holds true even when controlling for academic achievement, and similar patterns are observed among 10-year-old students (see Table A3.17). The relationships with achievement motivation, persistence and curiosity are also significantly stronger for student-teacher relationships than peer-to-peer relationships (see Tables A3.19, A3.17).
Standardised regression coefficients of 15-year-old students’ individual skills on teacher-student and peer-to-peer relationships, average across sites
Note: All coefficients are statistically significant with a threshold of p < 0.05. Models control for gender, educational achievement, socio-economic and migrant status. See Annex A for information about how teacher-student and peer-to-peer relationship indices were calculated. Social and emotional skills are listed by the size of the effect on teacher-student relationships.
Source: OECD, SSES Database 2023 Tables A3.17 and A3.19.
These findings suggest that high-quality teacher-student relationships play a crucial role in fostering students’ ability to strive for excellence, engage with learning, persevere through challenges and maintain optimism. They also imply that teacher-student relationships should go beyond promoting achievement—partly because broadening such relationships will, in fact, enhance learning and achievement. Other research finds that positive, safe relationships with teachers enable curiosity, motivation and academic learning (Rogers and Thomas, 2023[4]; Cantor et al., 2019[7]).
Teachers’ own social and emotional capacities and skills are a key factor in good teacher-student relationships and effective social and emotional learning (Green and García-Millán, 2021[9]; Cefai et al., 2018[2]; Jones, McGarrah and Kahn, 2019[1]). Further results from the SSES 2023 discussed in Box 3.2 show that teachers’ capacity to cope with stress – one aspect of emotional regulation – differs across sites.
Teachers play a pivotal role in creating an environment that nurtures social and emotional learning as well as academic learning in schools (Wigelsworth et al., 2022[26]; Cefai et al., 2018[2]; OECD, 2021[27]). To do this, they must have adequate emotional capacity and robust social and emotional skills themselves (CASEL, 2024[28]; Jones et al., 2021[17]; Green and García-Millán, 2021[9]). Coping strategies for managing stress can help teachers maintain their own well-being and teach social and emotional skills (Maricuțoiu et al., 2023[29]) (see also Chapter 2). Teachers who cope well and regulate their emotions better can establish better relationships with their students and to be better able to model how to manage one’s own emotions (Aldrup, Carstensen and Klusmann, 2024[30]).
Only 30% of teachers use effective coping skills in some sites, while in others, almost 70% do so
In SSES 2023, teachers reported on their work-related stress and, separately, on the extent (from “not at all” to “to a large extent”) to which they use different coping strategies for such stress, such as maintaining good sleep habits or approaching problems optimistically (see Table A3.20).
Results show that teachers’ use of coping strategies was significantly related to their well-being at work in most sites and on average (see Table A3.21). Notably, across most sites the relationships were significant and strongest for strategies related to managing emotions and physical health: approaching problems optimistically, setting realistic goals and maintaining good health and sleep habits.
Figure 3.6 shows how the use of these strategies varies by site. It illustrates teachers’ average use of coping strategies across sites and the range between highest and lowest scoring sites. In Peru, Delhi (India) and Kudus (Indonesia), over 50% of teachers of 15-year-olds reported that they approach problems optimistically to a large extent, and close to 50% or more teachers maintain good health habits. By contrast, this is 35% or less in other sites (Table A3.20).
Teachers in Kudus (Indonesia) and Peru reported high use of proven effective strategies. For example, 59% or more reported maintaining good health habits and approaching problems optimistically. They also set realistic goals more often. These approaches can be classed respectively as lifestyle, emotion-focused and action-focused strategies, which all have proven benefits (Buettner et al., 2016[31]; Madigan and Kim, 2021[32]; Lee et al., 2023[33]; Corbett et al., 2022[34]). Conversely, fewer teachers from Kudus (Indonesia) and Peru reported using less effective strategies, such as working more hours, than in other sites.
Percentage of teachers of 15-year-olds who used the following coping strategies to a large extent, average across sites
Note: Top and bottom sites are those sites with respectively the highest and lowest percentages of teachers who reported using a given strategy “to a large extent.”
Source: OECD, SSES 2023 Database Table A3.20.
Peer-to-peer relationships are particularly important for developing trust, engaging with others and regulating emotions
Youth naturally orient more towards their peers as they age (Denham, 2018[22]; Immordino-Yang, Darling-Hammond and Krone, 2019[35]). During late childhood and adolescence, in particular, peer relationships become crucial for young people’s development – and for effective social and emotional learning (Yaeger, 2017[16]; Cefai et al., 2018[2]).
Students in SSES 2023 were asked about the quality of their relationships with their classmates, e.g. if classmates were friendly, respectful, mean, or showed concern for other students. Similar to teacher-student relationships, 15-year-old students with better relationships with their peers reported higher levels of all social and emotional skills across all sites. This was also true at age 10 (see Table A3.19).
Yet, Figure 3.5 shows that some skills showed significantly stronger relationships with peer-to-peer relationships than with teacher-student relationships for 15-year-olds: trust, engaging with others (sociability, energy, assertiveness) and emotional regulation skills (optimism, emotional control, stress resistance). This pattern also appears in most sites (see Tables A3.19, A3.17).
SSES findings suggest that relationships with peers and teachers serve different purposes. Good relationships with teachers are particularly relevant for persistence, motivation and curiosity, while positive relationships with peers seem to help refine social skills and the management of emotions. Positive relationships with peers also seem important for developing trust, which facilitates well-being and learning (OECD, 2024[13]; Center on the Developing Child, 2009[8]). Such positive relationships are also necessary for civic engagement, encouraging individuals to consider the collective good rather than solely focusing on their own interests (Balliet and Van Lange, 2013[36]; Talò, 2018[37]).
However, the strength of the relationship between skills and peer-to-peer relationships varies between sites. In Dubai (United Arab Emirates), Gunma (Japan) and Jinan (China), relationships between skills and peer-to-peer relationships are strongest. The relationships for teacher-student relationships and skills were strongest in Delhi (India), Dubai (United Arab Emirates) and Jinan (China) (Tables A3.19 and A3.17). These variations may reflect how relationships in school vary in nature and intensity between contexts, due to factors like class sizes, school structures and expected roles of teachers or students.
Perception of care and concern in school relationships drops as students age
Positive relationships strengthen development more when they are infused with safety, reliability and care for the individual. These features enable children and youth to explore the world, withstand adversity and develop healthy self-concepts (Center on the Developing Child, 2009[8]; Cantor et al., 2019[7]). Consequently, it is concerning that 15-year-olds in SSES 2023 perceived significantly less care and personal concern from both teachers and peers in their relationships with them, compared to 10-year-olds.
Figure 3.7 shows the percentage of 15- and 10-year-olds that either agreed or strongly agreed with each of the listed statements. SSES 2023 asked students, on the one hand, if teachers and peers were generally respectful (e.g. teachers and classmates are “respectful” and “friendly”) and on the other, if they showed care and concern, such as “the students at my school are interested in other students’ well-being”.
Percentage of 10-year-olds and 15-year-olds who agreed or strongly agreed with the following statements regarding perceived relationship with teachers
Note: Only age differences that are statistically significant at a threshold of p <0.05 are noted by site names. Sites are listed in alphabetical order.
Source: OECD, SSES 2023 Database Tables A3.16 and A3.18
Two patterns emerge. Figure 3.7 illustrates that 15-year-olds perceive similar levels of respect from teachers and classmates compared to 10-year-olds. However, they perceive less care and concern. Whereas 81% of 10-year-olds agree or strongly agree that their teachers would be concerned if the student were upset, only 65% of 15-year-olds agree on average – a statistically significant drop of 16 percentage points (Table A3.16).
A similar pattern emerges for peer-to-peer relationships across all sites, although with smaller significant decreases of 8-12% (Table A3.18). This shift appears for all six sites that surveyed both age cohorts, although the size of the change varies. In Sobral (Brazil), there is a 22 percentage-point drop in students who agree that their teachers would be concerned if they arrived upset to class, compared to 12 percentage-point drops in Jinan (China) and Helsinki (Finland) (Table A3.16).
This finding has implications for policy and practice, but it should also be placed in its developmental context. Adolescents naturally seek independence from adults and broaden their social circles (Yaeger, 2017[16]; Denham, 2018[22]), which can create emotional distance between adolescents and their school relations, especially with teachers. Lower life satisfaction has also been shown to mediate adolescents’ perception of relationships in school (Boruchovitch et al., 2021[38]). Finally, brain development at this age creates heightened social sensitivity, particularly to social rejection (Chatterjee Singh and Duraiappah, 2020[5]). Thus, this perceived lack of concern may partially reflect such normal changes.
Yet Figure 3.7 suggests that some sites can and should invest more in promoting relationships in schools, especially in secondary education. Primary schools often emphasise relationships and social and emotional learning more. Schools and classes are often smaller than in secondary schools; students spend more time with the same teacher(s) and classmates and they change class groupings less often. Teachers in primary schools also report feeling more prepared to teach social and emotional skills than teachers in secondary schools (see Chapter 2) (OECD, 2023[39]). Although adolescent needs differ, they still require the care and stability of strong relationships, which can benefit mental health and mitigate the effects of toxic stress and adversity (Yaeger, 2017[16]; Center on the Developing Child, 2015[40]; Steponavičius, Gress-Wright and Linzarini, 2023[41]). Secondary education, in particular, should create more opportunities for students to form meaningful, sustained relationships with staff and peers, such as through relationship mapping, school- or year-wide projects, regular advisory, “tutor group” or “homeroom” periods, or extra-curricular activities.
Small interactions, daily culture and formal structures are all important for enhancing relationships in schools (Osher and Berg, 2018[6]). School structures must support teachers in their relationship work and allow for sufficient time. Training teachers on how their interactions, language and coping strategies can foster care, respect and emotional regulation in their classrooms will benefit both staff and students (Green and García-Millán, 2021[9]; Jennings et al., 2019[42]; Ritchart, 2015[43]).
Reducing teachers’ non-teaching workload can create more time and capacity for cultivating relationships with students (OECD, 2021[44]). Integrating regular relationship-building opportunities into school schedules, such as “homeroom” periods or extra-curricular activities, can provide all students with chances to form connections outside of classes while allowing staff to monitor vulnerable students. Strategies like “relationship mapping” can leverage existing relationships without requiring costly new interventions (Harvard Graduate School of Education, 2024[45]).
The significant negative impacts of bullying1 and other forms of violence in schools on students’ health, well-being and education outcomes are well established (Moore et al., 2017[46]). Making schools safe learning environments is, therefore, a global priority in education. A significant minority of students are impacted by bullying; however, the prevalence and nature of bullying behaviours vary between schools and education systems. Examining the nature of the problem is an essential step in designing, implementing and monitoring effective approaches to reduce violence and its consequences.
On average across sites, 30% of 15-year-olds are involved in bullying, either as a victim, a perpetrator, or both
Bullying directly impacts a large minority of students. On average across sites, 30% of 15-year-old students report being bullying victims or perpetrators, or both. Students are considered victims of bullying if they report experiencing unkind or aggressive behaviours, such as being hit, pushed or made fun of a few times a month or more in the previous year. Similarly, bullying perpetrators are students who said they enacted these behaviours a few times a month or more in the same timeframe. Bullying behaviours include verbal or social bullying, such as being made fun of or left out of things, and physical bullying, such as being hit, pushed or threatened.
The extent of bullying varies considerably across sites (see Figure 3.8 and Table A3.34). Students in Delhi (India) report the highest levels of both bullying victimisation and perpetration, where 64% of students are involved in bullying. Levels of bullying are also higher than average in Bulgaria and Kudus (Indonesia), where approximately 40% of students report being involved. Students in Jinan (China) and Gunma (Japan) report the lowest levels of bullying among SSES sites; however, a significant minority of students, between 16% and 19%, remain directly involved. While addressing bullying is, therefore, a particular priority in those sites with very high levels, it remains an issue that has long-term and wide-ranging impacts on many students in all systems. Box 3.3 examines how levels of bullying have changed in Bogotá (Colombia) and Helsinki (Finland) since 2019.
Percentage of 15-year-olds who reported being a victim of bullying, a perpetrator or both
Note: The percentages of involvement in bullying refer to students who reported being involved in bullying - victim and/or perpetrator - in at least one of the items several times a month or more often. Sites are listed in descending order of the total involvement in bullying.
Source: OECD, SSES 2023 Database Table A3.34
Many students are both bullying victims and perpetrators, particularly in Delhi (India) and Bulgaria
Many students who are victims of bullying also report bullying other students and vice versa. On average, across sites, students involved in bullying can be considered as victims only (14% of all students), both victims and perpetrators (11%) and perpetrators only (5%) (see Figure 3.8 and Table A3.34). Most perpetrators also report being victims in all sites, meaning that students who are perpetrators only are the smallest group involved in bullying (fewer than 5% of all students in most sites and fewer than 10% in Bulgaria, Italian sites [Emilia-Romagna and Turin], Kudus (Indonesia) and Ukraine). On the other hand, just under half of all victims of bullying also report bullying other students on average across sites.
Of these three experiences (victims, perpetrators, or both), there is the greatest variation across sites in the proportion of students who report being both victims and perpetrators. This group ranges from 5% of all students in Peru and Jinan (China) to 39% in Delhi (India). In fact, all the elevated levels of bullying in Bulgaria and almost all in Delhi (India), compared to the average across sites, can be accounted for by this group of students (see Figure 3.8 and Table A3.34). On the other hand, there is less overlap between these experiences in other sites. In both Chile and Peru, only 26% of bullying victims reported also being perpetrators, for example (see Table A3.34).
Social and verbal bullying is most common, and students who are bullied physically are also usually bullied verbally or socially
Verbal and social bullying are more common than physical bullying across all sites; 28% of all students are involved in some form of verbal or relational bullying, while 16% are involved in physical bullying on average across sites (see Table A3.38). However, the extent of this difference and the overlap between these experiences among students vary. Levels of verbal or relational bullying were only slightly higher than physical bullying in Bulgaria (27% and 24% of all students reported being victims of these forms of bullying, respectively, see Figure 3.9 and Table A3.38). In addition, while sites with high levels of verbal or social bullying tend to also have high levels of physical bullying, there are exceptions. For example, in Bogotá (Colombia), Chile and Dubai (United Arab Emirates), physical bullying victimisation was below the average across sites, but verbal or social bullying were similar to or above average (see Figure 3.9 and Table A3.38).
Most students involved in physical bullying are also involved in verbal or social bullying. On average, across sites, 12% of students are victims of verbal or social bullying only, 10% are victims of both verbal or social and physical bullying, and just 3% are victims of physical bullying only (see Figure 3.9 and Table A3.39). This means that instances of students only being involved in physical bullying are rare, while it is common for students to only experience or perpetuate verbal or social bullying.
Understanding the nature of bullying is key to designing effective anti-bullying approaches
A range of anti-bullying interventions are effective in reducing bullying perpetration and victimisation in schools worldwide (Gaffney, Ttofi and Farrington, 2021[47]; Gaffney, Farrington and Ttofi, 2019[48]). Differences in the nature of bullying between education systems or individual schools, such as those discussed in this chapter, may impact the effectiveness of interventions. For example, in schools where most students are involved in bullying, efforts to involve peer bystanders (students who are witnesses but not involved in bullying) may be less effective, as these students are in the minority. Understanding the problem’s scale and nature is an essential step in choosing or designing an anti-bullying approach.
Percentage of 15-year-old students who reported being victims of verbal or social and/or physical bullying
Note: The percentages of victim of bullying refer to students who reported being victim in at least one of the items several times a month or more often. Sites are listed in descending order of the total percentage of victims of bullying.
Source: OECD, SSES 2023 Database Table A3.39
In sites with high levels of bullying and where most students involved are both victims and perpetrators, a comprehensive whole-school approach that involves all students as well as parents, teachers, and other school staff, as well as the wider community, may be particularly relevant. For example, informal peer involvement (such as discussions in class and role-playing activities) and providing information to parents are two anti-bullying intervention components associated with greater reductions in both perpetration and victimisation outcomes compared to others (Gaffney, Ttofi and Farrington, 2021[49]). One aspect these features have in common is that they do not target individual bullies or victims, which may be challenging or even counter-productive when bullying is widespread and systemic.
School leaders often do not recognise problems with bullying in their schools, particularly in sites where students report high levels of bullying
A whole-school approach to reducing bullying and improving school climate can be effective. However, such transformative approaches require committed leadership to involve, co-ordinate and motivate staff, students, parents and the wider community (Pearce et al., 2022[50]). This can be a challenge where school leaders lack the necessary information, training or resources to identify and tackle bullying.
Results from SSES 2023 suggest that students’ experiences and school leaders’ assessments of bullying levels in their schools often do not align. Such discrepancies are particularly evident in sites where students report the highest levels of bullying. Fewer than 5% of school leaders in Delhi (India) and Kudus (Indonesia) said intimidation or bullying occurs monthly or more among students in their school, while fewer than 10% reported this in Bulgaria. These assessments stand in stark contrast to reports from students in these sites (see Figure 3.10 and Table A3.22).
Percentages of 15-year-old students who reported being victim of physical bullying and of students in school where principals reported bullying several times a month or more often, across sites
Source: OECD, SSES 2023 Database Tables A3.22 and A3.40.
There are several potential explanations for discrepancies between student and school leader responses. While students were asked if they had experienced different forms of aggressive behaviours, these experiences may not meet the threshold for bullying according to principals or even by students themselves. For example, although behaviours such as making fun of others and hitting or pushing are problematic, if students participated willingly or were of equal strength or status, such behaviour may not be considered bullying. In addition, many incidences of bullying may be managed between students themselves or by classroom teachers or other staff, meaning principals only hear of a minority of cases. However, even when these differences are considered, such discrepancies in those sites with the highest levels of bullying remain stark. This misalignment between students’ experiences and school leaders’ assessments may also contribute to the bullying problem. In sites where very few principals recognise there are issues with bullying, this suggests that acts of violence experienced by students, such as hitting and pushing or threatening others, may have become normalised in the school. Leadership is key to improving school climate and driving an effective whole-school approach, a strategy that engages all stakeholders (Gaffney, Ttofi and Farrington, 2021[49]). For such an approach to work, acknowledgement of the problem by school leadership is an important first step in addressing it.
While levels of bullying in schools were relatively low in most sites, according to school leaders, Helsinki (Finland), Spain and Chile stand out as particular exceptions (see Figure 3.10). In these sites, over 40% of students are in schools where school leaders said bullying occurs monthly or more often in their school. While levels of bullying are, therefore, a cause for concern for many school leaders in these sites, this also suggests they tend to be better informed of bullying incidents. In fact, Helsinki (Finland) is the only site where there was a clear relationship between principals’ assessments of bullying levels and the experiences reported by students in these schools (see Table A3.42). For example, 5% of students report being physically bullied in schools where their principals say it happens less than monthly, 11% in schools where it happens monthly, and 15% where it happens weekly or more often in Helsinki. This suggests that school leaders in Helsinki are better informed about the scale of bullying in their schools than in other sites.
Students involved in bullying tend to have lower levels of social and emotional skills, with variance in skill levels between victims and perpetrators
Students who are bullied, bully others, or take on both roles often share common background characteristics and risk factors. One such risk factor is low levels of social and emotional skills: students who are bullying victims, perpetrators, or both tend to have lower levels of almost all social and emotional skills compared to students not involved in bullying (see Figure 3.11 and Tables A3.35, A3.36 and A3.37). Both victims and perpetrators, including students who are both, tend to have lower responsibility, emotional control and trust. However, differences emerge depending on how students are involved in bullying for other skills. Firstly, victims of bullying tend to have poorer levels of all emotional regulation skills (emotional control, optimism and stress resistance) and energy compared to students not involved in bullying. However, students who are only perpetrators have similar levels of optimism, stress resistance and energy to those not involved in bullying in most sites. For sociability, perpetrators (including those who are also victims) have similar levels to those not involved in bullying in most sites, whereas in around half of sites, students who are only victims have poorer levels. Differences depending on students’ role in bullying are also found for empathy. Victims of bullying typically have higher levels of empathy in around half of sites (Bogotá [Colombia], Bulgaria, Chile, Italian sites [Emilia-Romagna and Turin], Sobral [Brazil], Spain and Ukraine), while perpetrators (including those who are also victims) have lower levels of this skill in almost all sites. For tolerance, bullying perpetrators tend to have lower levels in most sites, while students who are only victims have broadly similar levels to students not involved in bullying.
These results establish a clear link between students’ social and emotional skills and their involvement in bullying, suggesting fostering these skills can help create safer schools. At the same time, environments where students feel safe are key to supporting students’ social and emotional development. Wider research suggests that low empathy is a particularly important risk factor for involvement in bullying (Zych, Ttofi and Farrington, 2016[51]). Higher levels of empathy among victims who are not also perpetrators suggest this skill might help students avoid retaliating or becoming bullies themselves. Social and emotional learning approaches can therefore complement other bullying prevention strategies, although they should not be a stand-alone approach. Wider research finds that other well-established interventions can be more effective than social and emotional learning at reducing bullying (Gaffney, Ttofi and Farrington, 2021[49]; Vreeman and Carroll, 2007[52]). This may reflect that, while low social and emotional skills are a risk factor for bullying involvement, the precise mechanisms of these relationships are complex. For example, students with low levels of optimism and energy may be a target for bullies, but exposure to aggression can itself also undermine student well-being, leading to lower levels of these skills. It is also unclear from wider longitudinal research if high empathy protects against bullying or whether bullying causes low empathy (Zych, Farrington and Ttofi, 2019[53]). If low levels of skills are largely a consequence of bullying or other adverse experiences (rather than a cause), building students’ social and emotional skills alone may have less impact on bullying outcomes.
Standardised regression coefficients of 15-year-old students' individual skills on the index of bullying, average across sites
Note: Significant coefficients at a threshold of p < 0.05 are coloured, non-significant coefficients are outlined. Models control for gender, educational achievement, socio-economic and migrant status. Reference group is the students who reported not being victim or perpetrator of bullying. See Annex A for information about how the bullying indices were calculated. Social and emotional skills are listed by the size of the effect on being both victim and perpetrator of bullying.
Source: OECD, SSES 2023 Database Tables A3.35, A3.36 and A3.37.
Social and emotional learning programmes may also be more successful at reducing bullying in some sites or schools than others. In sites with the highest levels of bullying (Bulgaria and Delhi [India]), no differences in levels of empathy, engaging with others skills (sociability, energy, assertiveness), optimism or self-control are found between students who are bullying perpetrators only and those not involved in bullying. This suggests different levels of these skills among students are not a significant contributor to their involvement in bullying. In these sites, the normalisation of bullying behaviours in the school may play a more important role. Therefore, education systems’ strategies for bullying prevention should integrate social and emotional learning into comprehensive anti-bullying strategies. Implementing practices with a strong evidence base, such as having clear school rules and working with parents as part of a whole-school approach (Gaffney, Ttofi and Farrington, 2021[47]), can help reduce bullying and create a school environment where students feel safe and secure. Such an environment creates the ideal hub for students to develop both their academic and social and emotional skills.
Boys are more involved in bullying than girls, particularly in physical bullying and as perpetrators
Overall, boys are more likely to be involved in bullying than girls. Around one-third of boys (34%) were involved in bullying, meaning they reported being a victim, perpetrator, or both, compared to approximately one-quarter of girls (26%), on average across sites (see Table A3.34). However, boys are not over-represented in all bullying roles. A similar proportion of girls and boys report being bullying victims only. On average, across sites, 13% of girls and 14% of boys reported being victims only of social or verbal bullying, while 6% of girls and 8% of boys reported being victims only of physical bullying (see Figure 3.12). The heightened involvement of boys is, therefore, mostly due to more boys being perpetrators, either in isolation or in addition to being victims. One explanation for this finding is that boys tend to only be victims of bullying from other boys, while girls can often be victims of both boys and girls. This would be consistent with findings from the Programme of International Assessment (PISA) that show single-sex girls’ schools tend to have lower levels of bullying than gender-balanced schools, while single-sex boys’ schools have higher levels (OECD, 2019[54]). Alternatively, perhaps girls are less likely to report taking part in bullying than boys.
Percentage of 15-year-old students who reported being victim of bullying, perpetrator or both, by form of bullying and gender, average across sites
Note: The percentages of involvement in bullying refer to students who reported being involved in bullying - victim and/or perpetrator - in at least one of the items several times a month or more often.
Source: OECD, SSES 2023 Database Table A3.38.
Low academic attainment is also associated with greater involvement in bullying, both as a victim and perpetrator (see Tables A3.25 and A3.31). This is particularly the case in Bulgaria and Delhi (India), sites with high levels of bullying. Furthermore, disadvantaged students and students from migrant backgrounds tend to be more involved in bullying than other students in some sites. In Bogotá (Colombia), Delhi (India), Dubai (United Arab Emirates), Helsinki (Finland) and Sobral (Brazil), disadvantaged students had lower levels of bullying perpetration than advantaged students, while advantaged students had lower levels of both perpetration and victimisation than disadvantaged students in Bulgaria. However, in all these sites except Bogotá (Colombia), the link between socio-economic background and bullying involvement was weaker than that with academic achievement.
Bogotá (Colombia) and Helsinki (Finland) participated in SSES 2019 and 2023, enabling comparisons of students’ sense of belonging at school and levels of bullying between these years.
In both cities, 15-year-old students’ overall sense of belonging at school was lower in 2023 than in 2019 (see Table A3.1). In both Bogotá and Helsinki, the proportion of students who feel like an outsider, awkward and out of place, and lonely at school increased and fewer students said they make friends easily. In Bogotá, the largest shift was in the proportion of students who agreed they made friends easily: 66% of students agreed with this in 2023, down from 75% in 2019. While in Helsinki, 19% of students agreed they felt lonely in school in 2023, up from 14% in 2019.
Students in both rounds of SSES were asked how often they were made fun of, threatened, hit or pushed, or had their belongings taken away. In both cities, a similar proportion of 15-year-old students said they were made fun of in 2023 compared to 2019 (between 12% and 13% of students at least a few times a month in both sites across these years) (see Table A3.23). While this experience remained stable, levels of all types of physical bullying were greater in Helsinki in 2023 compared to 2019. In 2023, 7% of students in Helsinki said they were threatened at least a few times a month in 2023, up from 3% in 2019. Students in Helsinki were also more likely to say that other students had hit or pushed them and had taken away or destroyed their belongings in 2023. In Bogotá, a similar proportion of students said they were hit or pushed or had their belongings taken or destroyed in both 2019 and 2023, while they were slightly more likely to report being threatened in 2023 (5% compared to 3%).
Source: OECD, SSES 2019 and 2023 Databases Tables A3.1 and A3.23.
Following Chapter 2’s discussion on enhancing social and emotional education in schools, Chapter 3 expanded the scope to the holistic school environment. It discussed how improving students’ sense of belonging, positive experiences at school and relationships can strengthen their social and emotional skills and overall development. It also provided new insights about bullying, such as the frequent overlap between perpetrators and victims or misalignment between students’ and principals’ perceptions of bullying in their schools. The SSES results discussed in this chapter show that many systems can do more to build strong, healthy school communities, particularly in secondary schools. Solutions like creating more opportunities for staff and students to develop relationships, addressing site-specific sources of students’ negative emotions (anxiety, anger), and tackling bullying directly, can improve not only school climate and safety, but students’ and staff’s socio-emotional development.
The next chapter will shift attention to students’ home environments and their impact on students’ social and emotional skills, especially the effects of gender stereotypes, beliefs and home gender roles.
Online tables for each chapter can be accessed via the StatLink.
References
[30] Aldrup, K., B. Carstensen and U. Klusmann (2024), “The role of teachers’ emotion regulation in teaching effectiveness: A systematic review integrating four lines of research”, Educational Psychologist, Vol. 59/2, pp. 89-110, https://doi.org/10.1080/00461520.2023.2282446.
[36] Balliet, D. and P. Van Lange (2013), “Trust, conflict, and cooperation: A meta-analysis”, Psychological Bulletin, Vol. 139/5, pp. 1090-1112, https://doi.org/10.1037/A0030939.
[38] Boruchovitch, E. et al. (2021), “Adolescents’ mental health at school: The mediating role of life satisfaction”, Frontiers in Psychology, Vol. 1, http://www.frontiersin.org.
[31] Buettner, C. et al. (2016), “Teachers’ social–emotional capacity: Factors associated with teachers’ responsiveness and professional commitment”, Early Education and Development, Vol. 27/7, pp. 1018-1039, https://www.tandfonline.com/doi/abs/10.1080/10409289.2016.1168227.
[7] Cantor, P. et al. (2019), “Malleability, plasticity, and individuality: How children learn and develop in context”, Applied Developmental Science, Vol. 23/4, pp. 307-337.
[28] CASEL (2024), Strengthen Adult SEL, https://schoolguide.casel.org/focus-area-2/overview/.
[2] Cefai, C. et al. (2018), Strengthening Social and Emotional Education as a Core Curricular Area Across the EU: A Review of the International Evidence, http://www.ppmi.lt.
[40] Center on the Developing Child (2015), InBrief: Resilience Series, Harvard University, https://developingchild.harvard.edu/resources/inbrief-resilience-series/.
[8] Center on the Developing Child (2009), Young Children Develop in an Environment of Relationships, Harvard University, https://developingchild.harvard.edu/resources/wp1/.
[3] Charlton, C. et al. (2021), A Systematic Review of the Effects of Schoolwide Intervention Programs on Student and Teacher Perceptions of School Climate, SAGE Publications Inc.
[5] Chatterjee Singh, N. and A. Duraiappah (2020), Rethinking Learning: A Review of Social and Emotional Learning for Education Systems, UNESCO, https://mgiep.unesco.org/rethinking-learning.
[15] Chaux, E. et al. (2017), “Classrooms in peace within violent contexts: Field evaluation of Aulas en Paz in Colombia”, Prev Sci 18, pp. 828–838, https://doi.org/10.1007/s11121-017-0754-8.
[24] Clarke, A. et al. (2015), What works in enhancing social and emotional skills development during childhood and adolescence? A review of the evidence on the effectiveness of school-based and out-of-school programmes in the UK, World Health Organisation Collaborating Centre for Health Promotion Research, National University of Ireland Galway, http://www.nuigalway.ie/hprc/.
[34] Corbett, L. et al. (2022), “Characteristics and effectiveness of physical activity, nutrition and/or sleep interventions to improve the mental well-being of teachers: A scoping review”, Health Education Journal, Vol. 81/2, pp. 196-210, https://doi.org/10.1177/00178969211062701.
[19] Daley, S. and M. McCarthy (2021), “Students With Disabilities in Social and Emotional Learning Interventions: A Systematic Review”, Remedial and Special Education, Vol. 42/6, pp. 384-397, https://doi.org/10.1177/0741932520964917.
[22] Denham, S. (2018), “Keeping SEL developmental: The Importance of a developmental lens for fostering and assessing SEL competencies”, Establishing Practical Social-Emotional Competence Assessments Work Group.
[18] Durlak, J. et al. (2011), “The Impact of Enhancing Students’ Social and Emotional Learning: A Meta-Analysis of School-Based Universal Interventions”, Child Development, Vol. 82/1, pp. 405-432, https://srcd.onlinelibrary.wiley.com/doi/10.1111/j.1467-8624.2010.01564.x.
[48] Gaffney, H., D. Farrington and M. Ttofi (2019), “Examining the effectiveness of school-bullying intervention programs globally: A meta-analysis”, International Journal of Bullying Prevention, Vol. 1/1, pp. 14-31, https://doi.org/10.1007/s42380-019-0007-4.
[47] Gaffney, H., M. Ttofi and D. Farrington (2021), “Effectiveness of school‐based programs to reduce bullying perpetration and victimization: An updated systematic review and meta‐analysis”, Campbell Systematic Reviews, Vol. 17/2, https://doi.org/10.1002/cl2.1143.
[49] Gaffney, H., M. Ttofi and D. Farrington (2021), “What works in anti-bullying programs? Analysis of effective intervention components”, Journal of School Psychology, Vol. 85, pp. 37-56, https://doi.org/10.1016/j.jsp.2020.12.002.
[11] Goldbeck, L. et al. (2007), “Life satisfaction decreases during adolescence”, Qual Life Res, Vol. 16, pp. 669-979.
[23] Gutman, L. and I. Schoon (2013), The impact of non-cognitive skills on outcomes for young people: Literature review, Institute of Education, University College London, http://www.ioe.ac.uk.
[45] Harvard Graduate School of Education (2024), Relationship Mapping Strategy, https://mcc.gse.harvard.edu/resources-for-educators/relationship-mapping-strategy.
[10] Ibarra, B. (2022), “Understanding SEL to create a sense of belonging: The role teachers play in addressing students’ social and emotional well-being”, Current Issues in Education, Vol. 23/2, https://cie.asu.edu/ojs/index.php/cieatasu/article/view/2049.
[35] Immordino-Yang, M., L. Darling-Hammond and C. Krone (2019), “Nurturing nature: How brain development Is inherently social and emotional, and what this means for education”, Educational Psychologist, Vol. 54/3, pp. 185-204.
[42] Jennings, P. et al. (2019), “Long-term impacts of the CARE program on teachers’ self-reported social and emotional competence and well-being”, Journal of School Psychology, Vol. 76, pp. 186-202.
[17] Jones, S. et al. (2021), Navigating SEL from the Inside Out: Looking Inside and Across 33 Leading SEL Programs: A Practical Resource for Schools and OST Providers Preschool and Elementary Focus, EASEL Lab, Harvard Graduate School of Education.
[1] Jones, S., M. McGarrah and J. Kahn (2019), “Social and emotional learning: A principled science of human development in context”, Educational Psychologist, Vol. 54/3, pp. 129-143.
[33] Lee, E. et al. (2023), “What about Mme? The Importance of teacher social and emotional learning and well-being in the classroom”, Beyond Behavior, Vol. 32/1, pp. 53-62, https://doi.org/10.1177/10742956221145942.
[32] Madigan, D. and L. Kim (2021), “Does teacher burnout affect students? A systematic review of its association with academic achievement and student-reported outcomes”, International Journal of Educational Research, Vol. 105, p. 101714, https://doi.org/10.1016/j.ijer.2020.101714.
[29] Maricuțoiu, L. et al. (2023), “Is teachers’ well-being associated with students’ school experience? A meta-analysis of cross-sectional evidence”, Educational Psychology Review, Vol. 35/1, pp. 1-36, https://doi.org/10.1007/S10648-023-09721-9/TABLES/4.
[12] Marquez, J. et al. (2024), Child and Adolescent Well-Being: Global Trends, Challenges and Opportunities.
[46] Moore, S. et al. (2017), “Consequences of bullying victimization in childhood and adolescence: A systematic review and meta-analysis”, World Journal of Psychiatry, Vol. 7/1, p. 60, https://doi.org/10.5498/wjp.v7.i1.60.
[13] OECD (2024), Social and Emotional Skills for Better Lives: Findings from the OECD Survey on Social and Emotional Skills 2023, OECD Publishing, https://doi.org/10.1787/35ca7b7c-en.
[39] OECD (2023), “Schools as hubs for social and emotional learning: Are schools and teachers ready?”, OECD Education Spotlights, No. 4, OECD Publishing, Paris, https://doi.org/10.1787/f6d12db7-en.
[27] OECD (2021), Embedding Values and Attitudes in Curriculum: Shaping a Better Future, OECD Publishing, Paris, https://doi.org/10.1787/aee2adcd-en.
[44] OECD (2021), Positive, High-achieving Students?: What Schools and Teachers Can Do, TALIS, OECD Publishing, Paris, https://doi.org/10.1787/3b9551db-en.
[54] OECD (2019), PISA 2018 Results (Volume III): What School Life Means for Students’ Lives, PISA, OECD Publishing, Paris, https://doi.org/10.1787/acd78851-en.
[6] Osher, D. and J. Berg (2018), School Climate and Social and Emotional Learning, CASEL, http://www.rwjf.org/socialemotionallearning.
[50] Pearce, N. et al. (2022), “‘It’s all about context’: Building school capacity to implement a whole-school approach to bullying”, International Journal of Bullying Prevention, Vol. 6/1, pp. 53-68, https://doi.org/10.1007/s42380-022-00138-6.
[25] Poulou, M. (2017), “An examination of the relationship among teachers’ perceptions of social-emotional learning, teaching efficacy, teacher-student interactions, and students’ behavioral difficulties”, International Journal of School and Educational Psychology, Vol. 5/2, pp. 126-136, https://www.tandfonline.com/doi/abs/10.1080/21683603.2016.1203851.
[43] Ritchart, R. (2015), Creating Cultures of Thinking: The 8 Forces We Must Master to Truly Transform Our Schools, Jossey-Bass, https://www.wiley.com/en-us/Creating+Cultures+of+Thinking%3A+The+8+Forces+We+Must+Master+to+Truly+Transform+Our+Schools-p-9781118974605.
[4] Rogers, C. and M. Thomas (2023), Educational Neuroscience: The Basics, Routledge.
[20] Rowe, H. and E. Trickett (2018), “Student Diversity Representation and Reporting in Universal School-Based Social and Emotional Learning Programs: Implications for Generalizability”, Educ Psychol Rev, Vol. 30, pp. 559-583, https://doi.org/10.1007/s10648-017-9425-3.
[41] Steponavičius, M., C. Gress-Wright and A. Linzarini (2023), “Social and emotional skills: Latest evidence on teachability and impact on life outcomes”, OECD Education Working Papers, No. 304, OECD Publishing, Paris, https://doi.org/10.1787/ba34f086-en.
[37] Talò, C. (2018), “Community-based determinants of community engagement: A meta-analysis research”, Social Indicators Research, Vol. 140/2, pp. 571-596, https://doi.org/10.1007/S11205-017-1778-Y/TABLES/4.
[14] Universidad de los Andes Colombia (2023), Aulas en Paz, https://imagina.uniandes.edu.co/hub/aulas-en-paz/.
[52] Vreeman, R. and A. Carroll (2007), “A systematic review of school-based Interventions to prevent bullying”, Archives of Pediatrics and Adolescent Medicine, Vol. 161/1, p. 78, https://doi.org/10.1001/archpedi.161.1.78.
[21] Wigelsworth, M. (2016), “The impact of trial stage, developer involvement and international transferability on universal social and emotional learning programme outcomes: A meta-analaysis”, Cambridge Journal of Education, Vol. 46/3, pp. 347-376, https://doi.org/10.1080/0305764X.2016.1195791.
[26] Wigelsworth, M. et al. (2022), “Social and emotional learning in primary schools: A review of the current state of evidence”, British Journal of Educational Psychology, Vol. 92/3, pp. 898-924.
[16] Yaeger, D. (2017), “Social and emotional learning programs for adolescents”, Future of Children, Vol. 27/1, pp. 73-94, https://eric.ed.gov/?id=EJ1145078.
[53] Zych, I., D. Farrington and M. Ttofi (2019), “Protective factors against bullying and cyberbullying: A systematic review of meta-analyses”, Aggression and Violent Behavior, Vol. 45, pp. 4-19, https://doi.org/10.1016/j.avb.2018.06.008.
[51] Zych, I., M. Ttofi and D. Farrington (2016), “Empathy and callous–unemotional traits in different bullying roles: A systematic review and meta-analysis”, Trauma, Violence, and Abuse, Vol. 20/1, pp. 3-21, https://doi.org/10.1177/1524838016683456.
← 1. Bullying is typically defined as unwanted aggressive behaviour that is repeated over time and involves an imbalance of power or strength. However, definitions can vary across cultures and contexts. In the SSES, students are considered a victim or perpetrator of bullying if they report experiencing or carrying out any of the following behaviours at least a few times a month in the past year: being left out of things on purpose; being made fun of; being threatened; taking away or destroying belongings; being hit or pushed around; and having nasty rumours spread about them.