A Look into the New TIMSS Data on Bullying and Learning

Physical, emotional and sexual violence in and around schools is inherently harmful, with far-reaching consequences for children’s well-being and development. Among the many forms of violence that children may experience, peer-to-peer bullying is the most widely measured, offering valuable insights into its potential effects on children’s learning, physical and mental health.

In this blog, we draw on the 2023 Trends in International Mathematics and Science Study (TIMSS)—released in February this year—to take a closer look at the links between bullying and learning. Using data from 21 middle-income countries, we highlight four key findings. First, lots of children experience bullying. Second, bullying is on the rise almost everywhere. Third, bullying is bad for learning (as other studies have shown). Fourth, most head teachers don’t take bullying seriously. This last finding underscores the need for interventions that engage the school staff to raise awareness about school-based violence and its consequences (as implemented in countries such as Peru, Mozambique, and Uganda).

1. Both boys and girls face frequent bullying, but boys are bullied more often than girls

School violence affects girls and boys in different ways, influenced by a mix of individual, community and institutional factors that influence why, where, and how children experience violence. Bullying is one common form of violence, and TIMSS data shed light on its prevalence. Across 21 middle-income countries, boys, on average, experience more bullying than girls. Among 4th graders (aged 9 to 10 years), 63 percent of boys reported experiencing bullying, while 55 percent of girls reported the same. We observe a similar result among 8th graders (aged 13 to 14 years). However, bullying tends to increase among older girls (64 percent), with a smaller gap compared to older boys (68 percent). Figure 1 shows that these patterns apply across all the middle-income countries included in the analysis, except for older students in Côte d’Ivoire and Brazil.

Our findings align with descriptive statistics from existing studies, indicating that bullying is high for both girls and boys but more prevalent among boys, particularly at younger ages. For girls, bullying—especially some forms of emotional bullying—increases during adolescence. Unfortunately, the TIMSS data does not provide information on the perpetrators (e.g., their gender or age) or where bullying happens (e.g., classroom, playground, or on the way to school), limiting our ability to further explore the dynamics of violence. Yet, a few studies show that gender stereotypes and socially expected behaviours influence bullying perpetration among boys and girls. These findings reinforce the importance of considering social norms and beliefs in the design of school-violence prevention interventions.

Figure 1. Bullying by sex and grade, 2023

Note: TIMSS data asks about eight forms of emotional and three forms of physical bullying. Using this information, the indicator of emotional or physical bullying is a binary variable that takes the value of one if the student reported experiencing at least one form of emotional or physical bullying over the last year. Data on 4th-grade and 8th-grade students are available only for 18 and 13 middle-income countries, respectively.

2. Bullying is on the rise in almost all countries

When comparing bullying data with TIMSS’s 2019 cycle, we note that in 2023, more students reported experiencing physical and emotional bullying from peers. On average, the percentage of 4th graders reporting bullying increased from 45 percent to 56 percent, and among 8th graders, from 60 percent to 64 percent. Across countries, the increase in bullying ranged from 4 to 23 percentage points for 4th graders, and from 4 to 18 percentage points for 8th graders. Noticeably, South Africa and Malaysia are among the only countries with a reduction in prevalence, yet the share of students experiencing bullying in these countries remains the highest. The persistently high rates—which may also reflect a better understanding of what constitutes bullying and/or a greater willingness to report such experiences over time—underscore the importance of strengthening education sector efforts to improve school environments. At the First Global Ministerial Conference to End Violence against Children, 45 countries committed to implementing actions to foster safe school environments. Of the 21 countries included in this analysis, only two committed to this, showing the need to strengthen efforts to tackle violence at school.

Figure 2. Bullying by grade, 2019 and 2023

3. Bullying is associated with worse learning outcomes

We know that violence is bad for learning, and while we still need to understand all the mechanisms linking bullying to educational outcomes, existing evidence offers important insights. For example, experiencing bullying may lead to poorer mental health, increased anxiety and sleep disruptions, and it is associated with lower academic engagement, self-esteem, and self-efficacy—factors that, in turn, are linked to poorer academic achievement. Additionally, some evidence shows that bullying is correlated with higher absenteeism, which can further hinder children's ability to learn.

What do we learn from TIMSS data? TIMSS defines student competencies in science and math based on four achievement score thresholds (from low to advanced), with the lowest indicating students who fail to reach a basic level of understanding in math or science (low international benchmark) at their grade level. Using these scores, we explore links between learning outcomes and bullying. Among 4th graders who experienced bullying, 35 percent scored below minimum proficiency, compared to 25 percent among those who didn’t. We see a similar pattern for 8th graders (differences are statistically significant for all countries in the analysis). These numbers reflect the negative relationship between learning and bullying, which is also documented in studies using similar assessment data.

In many countries, the achievement gap between children who were bullied and those who were not is slightly wider in science than in math. Moreover, students who experienced physical bullying were more likely to score below basic competency compared to those who experienced emotional bullying. This highlights that different forms of violence may have varying impacts and underscores the importance of deepening our understanding of the relationship between bullying and other forms of violence—including corporal punishment by teachers, sexual violence, and gang violence—and their impacts.

Figure 3. Share of students below minimum proficiency, broken down by whether the student experienced bullying

4. Not all schools view bullying as a severe problem

Despite high levels of bullying across countries, not all head teachers identified it as a problem. TIMSS data asks head teachers whether intimidation and injuries from fights between students are a problem in their schools. Using this information as a proxy for bullying, we observe that only 25 percent and 34 percent of head teachers consider this to be a moderate or severe problem among 4th- and 8th-grade students, respectively. Head teacher views varied across countries with similar bullying prevalence (Figure 5). For example, in Morocco, where 73 percent of 8th graders reported experiencing bullying, 61 percent of headteachers said bullying was a moderate to severe problem, while in Türkiye, which had similar levels of bullying relative to Morocco (see Figure 2), only 33 percent of head teachers said bullying was a problem in their schools.

We explore this further using a school-level regression with country fixed effects and find that, in both 4th and 8th grade, high bullying levels are not correlated with the likelihood of head teachers reporting bullying as a severe problem. Multiple factors might influence this, including other competing challenges at the school level, as well as the lack of awareness of the magnitude and consequences of bullying. For example, teachers are more likely to identify and intervene in overt forms of bullying, such as hitting or pushing, but fail to recognize emotional bullying. They can also miss symptoms associated with bullying, such as suicidal ideation. In existing school violence interventions, a whole school approach, involving school staff and teachers, is often key to reducing overall prevalence.

Figure 4. Principals who responded to student bullying as not a problem, a minor, moderate, or serious problem

Note: Head teachers of 4th graders are asked about the degree of which intimidation or verbal abuse (including texting, emailing, etc.) and physical fights among students is a problem. Head teachers of 8th graders are asked about the degree of which intimidation or verbal abuse (including texting, emailing, etc.) and physical injury among students is a problem. We use these options to proxy their perception of bullying.

Safer schools for better learning

Insights from TIMSS 2023 emphasise what we already know: many children experience bullying. The data also show us that over the last few years, bullying has been on the rise. Students who experience bullying are falling behind in developing basic subject competencies compared to their peers. Despite this, not all head teachers consider bullying a problem in their schools. Bullying’s pervasiveness makes it easy to overlook, but its consequences can last a lifetime.

The latest TIMSS results reiterate a dimension of violence that needs to be further unravelled to tackle the roots of why it happens and how to prevent it. While this blog focuses on insights related to bullying and learning, it is important to recognise that other forms of violence—such as sexual abuse— have severe and lasting effects. Much of the existing evidence centres on bullying, in part because it is more frequently and systematically measured. In contrast, the sensitive nature of asking about sexual violence means that less data is available. However, the limited evidence we do have highlights its serious impact. A lack of data should not lead to inaction—addressing all forms of violence must remain a priority.

Building on what we know works to prevent violence, working with civil society organisations and strengthening cross-sectoral collaboration are key to preventing all forms of violence in and around schools. The education sector must play a central role in ensuring child safety, and efforts such as the new Ministerial Taskforce to End Violence against Children can be instrumental in making the elimination of violence a funded priority in the national and global education agendas.

Many thanks to Dave Evans and Radhika Nagesh for their feedback on this post.