Abstract
Background
Pupils’ aggressive behaviour towards teachers is widespread. This study examined the prevalence and types of pupils’ aggressive behaviour towards teachers across different grades and the associations between social capital and the risk of pupils’ aggressive behaviour in a one-year follow-up study design.
Methods
We collected questionnaire data from 2018 to 2019 and 2019–2020. A total of 1198 teachers from 94 schools in Denmark participated in both rounds. The associations were analysed via multilevel logistic regressions.
Results
Teachers are frequently exposed to pupils’ aggressive behaviour, particularly those teaching the youngest pupils. Being called degrading names was the most frequent type of harassment, being yelled in a threatening manner was the most frequent type of threat, and being hit was the most frequent type of physical violence. Social capital (bonding) at baseline was statistically significantly associated with a reduced risk of pupils’ harassment and threats one year later. Furthermore, social capital (linking) at baseline was statistically significantly associated with a reduced risk of threats each year.
Conclusion
Social capital can reduce the risk of pupils’ aggressive behaviour in the form of harassment and threats. Therefore, fostering a supportive climate in which teachers can openly discuss pupils’ aggressive behaviour with their colleagues and supervisors is recommended.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12889-025-23360-5.
Keywords: Teachers, Pupils' aggressive behaviour, Social capital
Introduction
Pupils’ aggressive behaviour towards teachers is a common phenomenon in schools across countries and cultural settings [1]. For instance, a survey of 9,370 pre-kindergarten to 12th-grade teachers in the US revealed that 44% were exposed to obscene remarks or gestures, 25% were verbally threatened and 19% were physically attacked [2]. In a South Korean study among 1,054 teachers, 27% reported being exposed to verbal victimisation, and 19% victimised by students throwing, kicking, or destroying objects [3]. In Europe, teacher victimisation appears to be less frequent than in the US, but the findings remain concerning. For instance, a Swedish study found that nearly every third teacher had experienced at least one incident of student generated teacher victimisation during the last 12 months [4]. In Denmark, a survey has revealed significant levels of teacher-targeted aggression, with 17.4% of primary school teachers reporting verbal threats and 14.3% encountering physical violence [5]. Together, these findings highlight that teacher victimisation is a global issue, though its frequency varies across regions.
Given the varied prevalence rates of teacher victimisation globally, further investigation into this issue is imperative. A meta-analysis including 24 studies from around the world has sought to address this issue [6]. The seven studies that provided information for the meta-analysis revealed a pooled prevalence of 53% for any type of pupils’ aggressive behaviour towards teachers reported over the last 2 years; however, the prevalence ranged widely, from 20 to 75% [6]. This large variance might be due to differences in factors such as measurements, time frames, or countries [7–9]. While these findings highlight the widespread nature of teacher victimisation, understanding its full impact requires a closer examination of the specific types of aggression that teachers face.
However, teacher-directed violence is not defined consistently in the literature [7, 8], which means it is not always clear what is meant by pupils’ aggressive behaviour towards teachers, i.e., whether it encompasses harassment, threats, or physical violence. For instance, one study assessed pupils’ aggressive behaviour using only two items: whether a teacher had been threatened or physically attacked during the last 12 months [10] or at any point in their career [11]. In contrast, another study employed a more comprehensive 24-item measure covering physical victimisation, verbal victimisation, social victimisation, cyber victimisation, sexual harassment and personal property offences [12]. Two recent studies have used 11 items to assess the frequency of verbal and threatening aggression and physical violence during the school year, finding that pupils’ aggressive behaviour is most frequently expressed through harassment, which can include obscene remarks or gestures, intimidation, verbal threats, and physical attacks [2, 9]. Furthermore, teachers have also reported instances of damage to their personal property or school property [13]. A report from the Danish Ministry of Education used 12 items to assess the frequency of violence, 7 items to assess threats, and 8 items to assess harassment [14]. The report found that being called degrading names (e.g., whore, bitch, idiot) was the most frequent type of harassment; being yelled at in a threatening manner was the most common type of threat; and being pushed was the most common type of violence. The studies demonstrated that teachers may be exposed to many different kinds of harassment, threats, and violence, highlighting the need for a nuanced understanding and measurement of pupils’ aggression toward teachers. With overly simplistic questions regarding pupils’ aggressive behaviour, there is a risk of misclassifying and underreporting pupils’ aggressive behaviour, as it is up to the respondent to define what constitutes violence and threats. Based on a review, a model was proposed that classifies violent acts against teachers into three main categories: harassment, verbal threats, and physical violence [8]. A precise understanding of the specific types and frequency of pupils’ aggressive behaviour is essential for informing targeted intervention strategies that can significantly enhance educator safety [8, 15]. On this background, harassment, threats, and physical violence were selected as focal points in this present study because they represent distinct and significant forms of pupils’ aggressive behaviour that teachers regularly encounter.
In addition, there is considerable variation in the findings regarding the grade levels at which pupils’ aggressive behaviour is most prevalent. Some studies suggest that middle school pupils are at greater risk of exhibiting aggressive behaviour than high school or elementary school pupils [16, 17]. Conversely, other research indicates that the highest risk of physical aggressive behaviour occurs during the elementary grades [18–20] and middle and high school teachers were more likely to report verbal harassment compared to elementary school teachers [19]. This inconsistency highlights the need for further research to clarify these patterns and determine which prevention measures should be prioritised and tailored to address aggressive behaviour most effectively across different grade levels.
There are also conflicting results regarding demographic and geographic risk factors. Some studies reported a greater frequency of pupils’ aggressive behaviour towards male teachers [21, 22], whereas another study reported that female teachers experienced a greater frequency of pupils’ aggressive behaviour. another study found that sex was not associated with an increased risk of violence, threats, or harassment [23].
Given these gaps, this study aims to provide a nuanced analysis specifically focusing on municipal primary and lower secondary schools in Denmark to enrich the understanding of this issue. A deeper insight into this phenomenon is particularly important due to its potential consequences for both teachers and pupils.
Pupils’ aggressive behaviour not only disrupts the learning environment but can also have significant effects on teachers’ well-being and health. Studies have shown that exposure to pupils’ aggressive behaviour is positively associated with burnout [12, 24, 25], anxiety and stress [2] and victimised teachers have been found to be more likely to change jobs or leave the profession [2]. Not only is serious victimisation, such as physical violence, related to teacher turnover, but relatively minor pupil aggressions towards teachers are also associated with turnover and burnout [24, 26]. Furthermore, it can be assumed that a dose‒response effect exists, whereby the health consequences are more severe, the more often the teacher is exposed to pupils’ aggressive behaviour. Therefore, it is crucial to identify and strengthen protective factors within the school environment to reduce the risk of pupils’ aggressive behaviour.
Social capital as a protective factor against pupils’ aggressive behaviour
Research highlights the importance of social support in helping teachers manage pupils’ aggressive behaviour and that teachers rely on collaboration, open communication, and supportive colleagues to handle challenging behaviours effectively [27, 28]. Teachers who feel supported by school administrators and work closely with principals and colleagues are less likely to experience aggression from pupils [29, 30]. The findings suggest that collaboration, supportive interactions, open communication, and involvement from school administration can reduce the risk of pupils’ aggressive behaviour. This dynamic aligns with the concept of social capital, which, unlike social support that tends to be an reactive individual-level resource accessed during times of challenge, functions at a collective level. In line with this, a qualitative study based on teachers from nine schools revealed that, alongside formal procedures for addressing pupils’ aggressive behaviour, teaching teams created and implemented various informal strategies to prevent pupils’ aggressive behaviour. These teams fostered trustful collaboration by exchanging knowledge, experience, and shared responsibility, which laid the foundation for the development of these informal approaches [31]. This dynamic can be connected to the concept of social capital [32]. The collective resources within teaching teams enable them to effectively develop solutions for managing pupils’ aggressive behaviour. Social capital refers to the resources embedded in social relationships among colleagues, teams, and leadership, facilitating trust, collaboration, and mutual support. These collective resources enhance both individual well-being and organisational effectiveness [33]. The concept of social capital encompasses social networks, norms, and trust developed among colleagues (bonding) and between colleagues and supervisors (linking) [32–34].
In Danish schools, teachers are organised into grade-level teams. Social capital bonding will measure the degree of social capital within teams, while social capital linking will assess the level of social capital between teams and the school leadership. Insights from other occupational fields, such as healthcare and social work, suggest that higher levels of social capital can significantly reduce the risk of violence and aggression towards employees [35, 36]. In healthcare, for example, high levels of social capital within work units, as opposed to low levels, are associated with a lower prevalence of workplace violence [36], whereas social educators (professional who works in educational, social, and community settings to support individuals’ development, well-being, and social integration) who experience high levels of social capital (bonding and linking) report lower levels of threats and violence even when workloads are high [35].
These studies suggest that if a school community is built on mutual support, trust, and open communication, it might similarly help reduce pupils’ aggressive behaviour towards teachers.
The findings of Andersen et al. [31] were based on a qualitative study with a limited sample, leaving uncertainty about whether these patterns apply to a larger population of teachers. To address this limitation, the present study uses a quantitative approach to assess whether the identified protective factor of social capital is generalisable to a broader teacher population. Understanding if social capital influences the likelihood of aggressive behaviour can provide valuable insights for developing prevention strategies and help school principals and educators create safer and more supportive school environments.
Study aims
This study aims to address critical gaps by providing a nuanced analysis of the types and frequency of pupils’ aggressive behaviour toward teachers, examining variations across grade levels, and exploring the potential protective role of social capital within school environments.
More specifically, this study has the following aims.
To examine the type and frequency of pupils’ aggressive behaviour towards teachers in municipal primary and lower secondary schools in Denmark.
To examine whether teachers’ experiences of their aggressive behaviour differ depending on the grade they teach.
To examine whether social capital (bonding) is associated with the risk of teachers being exposed to pupils’ aggressive behaviour.
To examine whether social capital (linking) is associated with the risk of teachers being exposed to pupils’ aggressive behaviour.
Methods
Study population
The study sample consisted of a cohort of schoolteachers from municipal primary and lower secondary schools (preschool to 9th grade) in Denmark. The baseline survey was conducted between September 2018 and January 2019, with a follow-up 1 year later, i.e., from September 2019 to January 2020. Previous research has explored different relationships between the demands, control, and support dimensions at work and indicators of mental health with 1-year, 2-year, and 3-year time lags. The results revealed that the strongest effects were found for a 1-year time lag. Therefore, we chose a 1-year time lag [37].
To base the study on a sample representing Danish primary and lower secondary schools as accurately as possible, we invited schools from all five regions of Denmark, both small and large schools. The selection of schools took into account the geographical differences in inhabitants so that relatively more schools were invited from regions with the most inhabitants. The teachers’ e-mail addresses were obtained from the participating schools. At baseline, 4,935 teachers from 105 schools were invited to participate, and 2,336 completed the electronic questionnaire sent via e-mail (response rate of 47.3%). At follow-up, 94 of the 105 schools from the baseline survey agreed to participate. The follow-up questionnaire was sent to 3,883 teachers, 1,830 of whom responded (response rate of 47.1%). We found no difference in the distribution of gender or age among the teachers who responded to the baseline survey compared with those who responded to both rounds [38].
The final study population for the present study included participants who completed questionnaires at both baseline and follow-up (n = 1,198).
Measures
Social capital
Social capital at work refers to the value and benefits that individuals and organisations derive from their social networks, relationships, and interactions within the workplace. It encompasses trust, helping behaviour, shared norms, and collaboration across different levels of an organisation. Social capital was measured at baseline via two scales, the Danish social capital questionnaire [39]: the scale for measuring bonding (relations between people who belong to the same group or team) and the scale for measuring linking (relations between a work team and management). Both scales consist of four items each and are measured on a 5-point Likert scale ranging from a very high degree (4) to a very low degree (0).
We conducted a confirmatory factor analysis (CFA) to assess the factor structure of the social capital measure. Our results support the expected two-factor structure, with four items loading onto the bonding dimension and four items loading onto the linking dimension, respectively. The model demonstrated a good fit to the data: χ²(19) = 104.65, p <.001, CFI = 0.982, RMSEA = 0.062 (95% CI: 0.051, 0.074), SRMR = 0.028.
Social capital bonding
The items measuring bonding asked about whether the group/team helps colleagues who have too much to do, agree on what is most important in the work tasks, trust each other and if the group has a sense of unity and cohesion. A total sum score with a possible range of 0 to 16 was calculated for each participant.
Cronbach’s alpha = 0.89.
Social capital linking
The items measuring linking are assessed if the supervisor takes the employees’ needs and views into account when making decisions, helps to solve concrete problems in everyday life, has a great understanding of the work they do and whether the relationship with the supervisor is characterised by mutual respect and recognition. A total sum score with a possible range of 0 to 16 was calculated for each participant.
Cronbach’s alpha = 0.85.
Outcome: pupils’ aggressive behaviour towards teachers
On the basis of the definition of work-related violence by Wynne et al. 1997 [40], we applied a broad approach and measured three types of aggressive behaviour: (1) harassment, (2) threat of violence and (3) physical violence towards teachers. All the items were measured on a 5-point Likert scale ranging from never (0) to daily (4). The threats and violence items were inspired by a Swedish study [41] and utilised in several studies [42–44]. The harassment scale was specifically revised for this study. Because the measures of harassment, threats, and violence are based on a formative rather than a reflective model, traditional factor analysis techniques such as EFA and CFA—which assume underlying latent constructs—are not appropriate for this study.
Pupils’ aggressive behaviour towards teachers: harassment
Harassment was measured by 8 items asking about experiencing harassment, such as being called degrading things or subjected to unpleasant practical jokes during class or breaking or persistent harassment or disturbances in teaching directed towards you as a teacher in the last 12 months. A sum score, with a possible range of 0 to 20, was calculated for each participant. The scale was then dichotomised into ‘not exposed’ (0) and ‘exposed at least once during the last year’ (> 0).
Pupils’ aggressive behaviour towards teachers: threats
Threats were measured by 6 items asking about having experienced threats such as being threatened with objects, being threatened with beatings, being scolded, or shouting during the previous year. A possible range of 0 to 12 was calculated for each participant. The variable was dichotomised into ‘not (0) and ‘exposed at least once during the last year (> 0).
Pupils’ aggressive behaviour towards teachers: violence
Violence was measured by 11 items asking about having experienced, e.g., physically violent behaviour, including being hit, pushed, or kicked by pupils during the last 12 months. A sum score ranging from 0 to 44 was calculated for each participant. The scale was then dichotomised into ‘not exposed’ (0) and ‘exposed at least once during the last year’ (> 0)s.
Potential confounders
We considered age and sex as potential confounders. Information was obtained from the baseline questionnaire.
Covariables
The information about pupils’ baseline aggressive behaviour (harassment, threats and violence) used in the adjusted models was obtained from the baseline questionnaire. Furthermore, the question regarding the grade level at which teachers primarily teach is taken from the baseline questionnaire.
Statistical analyses
We used a chi-squared test to conduct a pairwise comparative analysis to determine whether there were statistically significant differences in the frequency of harassment, threats, and violence across grade levels.
We created a plot of standardised residuals against standardised predicted values and applied a logarithmic transformation, finding that the data did not meet the statistical assumptions of normality and homoscedasticity required for linear regression [45]. Therefore, the associations were examined using multilevel logistic regression to account for the possibility that responses from teachers within the same school were not independent.
Crude and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated for the associations between social capital (bonding and linking) and the three types of aggressive behaviour (harassment, threats, and violence) one year later, considering potential confounders and pupils’ aggressive behaviour at baseline. Missing data were handled using listwise deletion. All the statistical analyses were performed via the statistical software package SPSS (Version 20) (IBM Corp, Armonk, NY, USA).
Results
We found no difference in the distribution of gender or age among the teachers who responded to the baseline survey compared with those who responded to both rounds. The population characteristics and social capital work environment characteristics at baseline for those participants who answered the questionnaires at both baseline and follow-up are presented in Table 1 which shows that almost three-quarters of the participants were women (74.1%). Most participants were in the 41–50 years (37.1%) and 51–60 years (26.3%) age groups, whereas fewer participants were between 21 and 30 years of age (6.6%).
Table 1.
Population characteristics and exposure to harassment, threats, and violence (n = 1,198)
| n (%) | |
|---|---|
| Gender (at baseline) | |
| Women | 888 (74.1) |
| Men | 296 (24.7) |
| Missing | 14 (1.2) |
| Age (at baseline) | |
| ≤ 21–30 | 79 (6.6) |
| ≤ 31–40 | 241 (20.1) |
| ≤ 41–50 | 445 (37.1) |
| ≤ 51–60 | 315 (26.3) |
| > 60 | 98 (8.2) |
| Missing | 20 (1.7) |
| Haressment (at baseline) | |
| Exposed | 764 (63.8) |
| Not exposed | 399 (33.3) |
| Missing | 35 (2.9) |
| Threats (at baseline) | |
| Exposed | 513 (42.8) |
| Not exposed | 658 (54.9) |
| Missing | 27 (2.3) |
| Violence (at baseline) | |
| Exposed | 518 (43.2) |
| Not exposed | 630 (52.6) |
| Missing | 47 (3.9) |
| Haressment (at follow-up) | |
| Exposed | 747 (62.4) |
| Not exposed | 426 (35.6) |
| Missing | 35 (2.9) |
| Threats (at follow-up) | |
| Exposed | 487 (40.7) |
| Not exposed | 682 (56.9) |
| Missing | 29 (2.4) |
| Violence (at follow-up) | |
| Exposed | 463 (38.6) |
| Not exposed | 688 (57.4) |
| Missing | 47 (3.9) |
| Social capital (at baseline) | Mean/SD |
| Social capital (bonding) | 11.3/2.7 |
| Social capital (linking) | 10.7/2.9 |
At baseline, harassment was reported by 63.8% of participants, 42.8% reported being exposed to threats, and 43.2% reported being exposed to violence. At follow-up, the frequencies were nearly the same, except for violence, which decreased from 43.2 to 38.6%. More data are needed to determine whether this decrease is merely a fluctuation or represents a trend.
Appendices a–c show the frequencies at baseline and follow-up for the specific types of harassment, threats of violence or physical violence. Teachers can be exposed to many kinds of pupils’ aggressive behaviour during a 12-month period; some behaviours occur very rarely, whereas others occur more frequently. Being called degrading things (e.g., whore, bitch, idiot) was the most frequent type of harassment, with 53.1% being exposed to this harassment at least once during the past 12 months at baseline. Being yelled at in a threatening manner was the most frequent type of threat, with 40.9% being exposed to it at least once during the past 12 months at baseline. Finally, being hit was the most frequent type of physical violence, with 29.5% being exposed to it at least once during the past 12 months at baseline. The frequencies of harassment, threats, and violence remained largely unchanged from baseline to follow-up.
As shown in Table 2, many teachers in lower and middle grades are exposed to harassment. Furthermore, many teachers in the lower grades are exposed to threats and especially violence. There appears to be a clear trend indicating that the younger the pupils are, the more harassment, threats, and violence they direct towards teachers. However, teachers who teach mixed-level groups are also frequently subjected to harassment, threats, and violence. Nevertheless, there are significantly fewer teachers in these groups, and the results should therefore be interpreted with caution.
Table 2.
Prevalence of teachers’ reports of pupils’ aggressive behaviour at baseline, by grade level taught
| Baseline harassment | Baseline threats | Baseline violence | |||||||
|---|---|---|---|---|---|---|---|---|---|
| No | Yes | Total | No | Yes | Total | No | Yes | Total | |
| n (%) | n (%) | n (%) | n(%) | n (%) | n (%) | n (%) | n (%) | n (%) | |
| Teaching lower grades (1–3 grades) | 113 (28.0) | 291 (72.0) | 404 (100) | 193 (47.0) | 218 (53.0) | 411 (100) | 135 (33.7) | 266 (66.3) | 401 (100) |
| Teaching middle grades (4–6 grades) | 105 (31.3) | 230 (68.7) | 335 (100) | 184 (54.6) | 153 (45.5) | 337 (100) | 173 (53.1) | 153 (46.9) | 326 (100) |
| Teaching higher grades (7–9 grades) | 141 (46.5) | 162 (53.3) | 303 (100) | 213 (70.8) | 88 (29.2) | 301 (100) | 249 (82.5) | 53 (17.5) | 302 (100) |
| Teaching both lower grades and middle grades | 11 (36.7) | 19 (63.3) | 30 (100) | 19 (63.3) | 11 (36.7) | 30 (100) | 19 (65.5) | 10 (34.5) | 29 (100) |
| Teaching both middle grades and higher grades | 12 (27.9) | 31 (72.1) | 43 (100) | 23 (53.5) | 20 (46.5) | 43 (100) | 32 (74.4) | 11 (26.6) | 43 (100) |
| Teaching all grades | 5 (31.2) | 11 (68.8) | 16 (100) | 7 (43.8) | 9 (56.2) | 16 (100) | 6 (37.5) | 10 (62.5) | 16 (100) |
Table 3 presents a comparison of reported harassment, threats, and violence across grade levels. Due to an insufficient number of participants in the mixed-grade group, these cases were excluded from the analysis. The comparison between lower and middle grades reveals that teachers in the lower grades report significantly higher levels of threats and violence. Similarly, when comparing lower and upper grades, teachers in the lower grades report significantly more harassment, threats, and violence. A comparison between middle and upper grades also indicates that teachers in the middle grades report significantly higher levels of harassment, threats and violence compared to higher grades.
Table 3.
Pairwise comparative analysis of harassment, threats, and violence involving grades 1–3
| Comparison | Pupils’ aggressive behaviour | Value | df | P value |
|---|---|---|---|---|
| Lower grades vs. middle grades | Harassment | 1.00 | 1 | 0.32 |
| Threats | 43.24 | 1 | 0.04* | |
| Violence | 27.72 | 1 | 0.00* | |
| Lower grades vs. higher grades | Harassment | 25.29 | 1 | 0.00* |
| Threats | 40.18 | 1 | 0.00* | |
| Violence | 165.41 | 1 | 0.00* | |
| Middle grades vs. higher grades | Harassment | 15.50 | 1 | 0.00* |
| Threats | 17.67 | 1 | 0.00* | |
| Violence | 27.72 | 1 | 0.00* |
The analyses are based on chi-square tests
Note: ‘Value’ denotes the chi-square test statistic, indicating the extent to which observed frequencies differ from expected frequencies under the null hypothesis
Outcomes for harassment, threats, and violence were treated as binary variables (i.e., all nonzero counts were coded as 1). *P <.05
Table 4 shows that high social capital (bonding, i.e., social relations between teachers) was statistically significantly associated with a reduced risk of pupils’ aggressive behaviour in the form of harassment and threats. High social capital linking (i.e., social relations between teachers and their supervisors) were statistically significantly associated with a reduced risk of threats. The remaining associations between social capital bonding or social capital linking and pupils’ aggressive behaviour were statistically nonsignificant but the estimates pointed in the same direction.
Table 4.
Associations between social capital and risk of pupils’ aggressive behaviour
| Harassment | Threats | Physical violence | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Unadjusted | Model 1 | Model 2 | Unadjusted | Model 1 | Model 2 | Unadjusted | Model 1 | Model 2 | |
| OR (95% CI) | OR (95% CI) | OR /95% C)] | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Social capital bonding | |||||||||
| Low | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
| High |
0.71 (0.52–0.96) |
0.69 (0.51–0.94) |
0.63 (0.45–0.90) |
0.65 (0.49–0.87) |
0.65 (0.49–0.87) |
0.61 (0.4–0.84) |
0.92 (0.68–1.23) |
0.88 (0.65–1.19) |
0.74 (0.53–1.05) |
| Social capital linking | |||||||||
| Low | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| High |
1.12 (0.83–1.53) |
1.12 (0.82–1.51) |
1.21 (0.85–1.71) |
0.65 (0.49–0.87) |
0.65 (0.48–0.87) |
0.60 (0.44–0.84) |
0.92 (0.68–1.23) |
0.88 (0.65–1.19) |
0.74 (0.52–1.05) |
Model 1: Adjusted for gender and age
Model 2: Adjusted for gender and age and baseline harassment, threats and violence
Bold type: Statistical significant associations (p <.05)
Discussion
Our study shows that teachers in municipal primary and lower secondary schools in Denmark are often exposed to pupils’ aggressive behaviour harassment, threats of violence and physical violence during their work and are most frequently exposed to pupils’ aggressive behaviour in classes with the youngest pupils.
High levels of social capital among colleagues (bonding) significantly reduce the risk of harassment and threats. Additionally, social capital between teachers and their supervisors (linking) significantly reduces the risk of threats. Both social capital linking and bonding trend toward reducing the risk of physical violence; however, the associations were not statistically significant.
In the following, we discuss our results in relation to the four aims of the study.
Our first aim was to examine the frequency and type of pupils’ aggressive behaviour towards teachers in Danish schools and investigate whether teachers’ experiences of pupils’ aggressive behaviour differ depending on the grades they teach. As documented in other studies, we found that many teachers are exposed to pupils’ aggressive behaviour. The results from the appendices demonstrated that teachers may be subjected to a wide range of harassment, threats, and acts of violence, highlighting the complex and multifaceted nature of pupils’ aggressive behaviour towards teachers. This emphasises the need for a nuanced understanding of the different types of pupils’ aggressive behaviour as well as the importance of developing precise measurement tools to assess their frequency and impact. A deeper insight into these dynamics can help inform targeted prevention strategies and intervention programs aimed at fostering a safer and more supportive working environment for teachers.
Harassment was the most frequent type of aggressive behaviour, as found in other studies [21, 28], whereas exposure to physical violence occurred less often. Our results correspond with those of a recent meta-analysis [6], which reported low prevalence rates for the more intrusive types of aggressive behaviour. These findings are important for the development of tailored preventive strategies, as these strategies must specifically target the most prevalent types of students’ aggressive behaviour, including harassment. This focus is particularly crucial because even less overt forms of aggressive behaviour can negatively impact teachers’ well-being, especially given their greater frequency [24]. Tailoring preventive measures that address both frequent, lower-level aggression and infrequent but severe incidents, it is possible to foster a safer school environment for teachers.
One important factor to consider when examining teachers’ exposure to pupil aggression is the broader educational context. As mentioned, many teachers experience aggressive behaviour from pupils, and there may be multiple explanations for this trend. One significant factor is the implementation of inclusive education, which aims to integrate all pupils, regardless of their abilities or needs, into the general classroom setting. In Denmark, this philosophy has been widely adopted within the public school system to ensure that pupils with special needs or disabilities have access to the same educational opportunities as their peers [46]. Nevertheless, despite its progressive principles, the implementation of inclusive education in Danish public schools faces considerable challenges. While these reforms promote educational equity, they have also introduced difficulties. Teachers often report feeling insufficiently trained to manage the complexities of inclusive classrooms, leading to increased stress, potential conflicts, and aggressive behaviour [14, 47] However, this is not the only explanation. A report from the Danish Ministry of Education indicates that younger pupils without special needs struggle to handle demands from teachers and may respond with kicking or hitting [14].
Our second aim was to examine whether teachers’ experiences of their aggressive behaviour differed depending on the grades the teachers taught. In line with previous studies [18, 20], we found that episodes of aggressive behaviour were most often carried out by the youngest pupils (preschool to 3rd grade). Our findings indicate that pupils in grades 1–3 may encounter distinct challenges related to aggressive behaviours, which differ from those experienced by students in middle and higher grades. This finding is comparable to findings from a representative study among 2,000 German teachers from a variety of schools, which revealed that elementary school teachers reported having been exposed to physical violence from pupils to a greater degree than teachers who taught in higher grades [48]. A similar result was reported in a smaller Danish study among 19 schools [14]. One explanation for this finding could be that, especially in the lower grades, it is important for teachers to be able to draw students’ attention and motivate pupils. Furthermore, the teacher must support the pupils by being attuned to the pupils’ needs and functioning as caregivers in the absence of their parents [49, 50]. However, the support necessary for managing a class with more than 20 pupils, some of whom might need special attention, is not always provided, making it difficult for teachers to react in a timely manner to all situations that can lead to more aggressive behaviour. Our results may have implications for school policies and prevention, as they suggest that preventive measures and support systems should be targeted early in a child’s education to address aggression before it escalates. By understanding that younger pupils are more likely to engage in such behaviours, schools can develop age-appropriate strategies, including teacher training in de-escalation and classroom management techniques [51], to mitigate these issues and protect teachers from early exposure to pupils’ aggressive behaviour. Compared to older pupils, early-grade pupils may display aggressive behaviour in ways that require tailored interventions and support strategies.
Our third and fourth aims were to investigate whether social capital was associated with reporting less aggressive behaviour from pupils. We found that social capital bonding (social capital among colleagues) was statistically significantly associated with a decreased risk of both harassment and threats. Furthermore, both dimensions of social capital (bonding and linking) were associated with a reduced risk of pupils’ aggressive behaviour in the face of threats. The mitigating effect of social capital is in line with previous findings from other occupations, showing that social capital reduces the risk of work-related violence in hospitals and among special educators [35, 36]. One explanation may be that a high level of workplace social capital, characterised by strong, well-functioning relationships among coworkers and between employees and their supervisor, enhances task coordination and work planning [52] and improves the identification of relational issues between pupils and teachers. High bonding social capital means that teachers and the school principal have supportive, trusting relationships with one another [33]. This might enable more trustful and open communication, which makes it easier to share strategies for managing pupils’ aggressive behaviour. When social capital is high, the degree of shared understanding is high, which may also include a more consistent approach to managing pupil behaviour [31]. Consistency in rules, expectations, and consequences across teachers reduces ambiguity for pupils, decreasing the likelihood of frustration, boundary testing, and aggression, which may be especially beneficial for pupils with special needs [53]. Additionally, linking social capital may ensure that teachers have access to support from school principals, including guidance and resources to address aggressive behaviour. This is likely because supervisors can provide teachers with instrumental, task-related support, including direct assistance, resource planning, feedback, and guidance [54].
The development of social capital at both the bonding (among teachers) and linking (between teachers and supervisors) levels may have practical implications for fostering a positive and resilient school environment. By reinforcing trust and collaboration across these levels, social capital contributes to a resilient school climate, enhancing teachers’ capacity to manage pupils’ aggressive behaviour proactively and effectively. Through team meetings, structured peer collaboration strengthens bonding social capital, enabling teachers to share strategies and create a unified approach to behaviour management [31]. Investing in social capital in these ways may contribute to a safer, structured environment, reducing incidents of harassment and threats and enhancing teacher resilience, well-being, and retention.
Strengths and limitations
The present study has several important strengths. The study is based on a longitudinal design with a large number of municipal primary and lower secondary schools, including small and large schools from all parts of the country. Furthermore, the items measuring pupils’ aggressive behaviour are rather detailed, decreasing the risk of misclassification. We used logistic regressions as most respondents report either never experiencing the aggressive behaviour or experiencing it rarely. Therefore, we transformed the data into a binary format. This approach helps to more accurately capture meaningful patterns in the data by categorising respondents into two groups: those who experienced the pupils’ aggressive behaviour and those who did not. This method minimises the influence of the overrepresentation of zero responses and enhances the robustness of our analysis.
However, the results of the present study should be considered in light of several limitations. First, the participants may not be representative of the respective population of teachers, since several schools refused to participate; thus, we cannot rule out some potential selection bias, which may reduce the external validity of the findings. Furthermore, if only employees with an interest in the topic responded (i.e., those who were most exposed to pupils’ aggressive behaviour), this could also have induced potential selection problems. However, the distribution of gender and the age of our population of teachers corresponds to that of the members of the Danish Teacher Association (representing 96% of all teachers in Denmark), indicating that the selection bias was only minor. Furthermore, we found no differences in the distributions of gender, age, or job title among the teachers who responded to the baseline survey and those who responded to both rounds.
Additionally, only healthy employees participated in the current study. Those on sick leave, including sick leave due to pupils’ aggressive behaviour or time pressure, did not participate. This may have led to an underestimation of the associations between time pressure at work and pupils’ aggressive behaviour.
Second, the data came entirely from self-reports, which may introduce mono-method bias due to unmeasured third variables [55]. A further limitation was the use of dichotomised measures, i.e., threats and violence, which implies a loss of information and perhaps a simplification of the relationships in question. However, dichotomising also has several advantages because measures such as odds ratios often provide more realistic and meaningful assessments of the strength of relationships [56]. By converting continuous variables into categorical variables, dichotomising can simplify complex data, making it easier to interpret and communicate results to nonspecialists.
A final limitation concerns recall bias; that is, the answers concerning the frequency of exposure in a questionnaire may be influenced by the fact that participants have difficulties remembering which and how many episodes of aggressive behaviour they have experienced during the past 12 months [57]. However, by asking about what type of harassment, threats and violence teachers experienced during the past 12 months, we asked them to think about very specific situations and not aggressive behaviour in general. The similarities between our results and those of a smaller study in a different sample of schools led to similar results [14]; therefore, we suggest that the effect of recall bias was small.
Conclusion
Teachers in Danish municipal primary and lower secondary schools are frequently exposed to pupils’ aggressive behaviour, particularly in the lower grades. The findings from this study indicate that teachers may be subjected to various types of harassment, threats, and violence, underscoring the complex and multifaceted nature of pupils’ aggressive behaviour toward teachers.
The results further demonstrate that higher levels of bonding social capital at baseline were significantly associated with a reduced risk of harassment and threats one year later, while linking social capital was linked to a reduced risk of threats. Additionally, both bonding and linking social capital showed a trend toward reducing the risk of violence over the same period.
To effectively address pupil aggression in Danish primary and lower secondary schools, fostering strong social capital among school staff is recommended. Supportive and collaborative relationships among colleagues have the potential to reduce the frequencies of harassment and threats, thereby contributing to a safer school environment. Ultimately, enhancing social capital within schools not only benefits teachers but may also contribute to a more positive and supportive learning environment for pupils.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Abbreviations
- OR
Odds ratio
Author contributions
L.P.A. and T.N.W. made substantial contributions to the application, conception, design and acquisition of the data. L.P.A. and T.N.W. analysed and interpreted the data. L.P.A. drafted the article. T.N.W. critically revised the article for important intellectual content. L.P.A. and T.N.W. read the final manuscript and approved the version to be published.
Funding
Study supported by: The Danish Working Environment Research Fund. Grant number 14-2014-03/2014416500.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
All methods were carried out in accordance with relevant guidelines and regulations (declaration of Helsinki). Use of the data was carried out under the same conditions and with the same purpose as collected. Participation in the study was voluntary and collected data were treated confidentially. The study was approved by the Danish Data Protection Agency prior to collecting the data (1-16-02-287-18). According to Danish Law -Act on Research Ethics Review of Health Research Projects, (The Act on Processing of Personal Data - Act No. 429 of 31 May 2000) questionnaire and register-based studies require neither approval by ethical or scientific committees.
Consent for publication
Not applicable.
Informed consent
Informed consent was obtained from all the subjects involved in the study. The collected information was only used if the participant had given informed consent by ticking off a box with the following text at the beginning of the questionnaire: “I have been informed that I can receive a copy of my own answers and can have them deleted if I wish, and hereby agree to participate in the survey.” I chose to press “submit” after finishing the questionnaire.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Moon B, McCluskey J. School-based victimization of teachers in korea: focusing on individual and school characteristics. J Interpers Violence. 2016;31(7):1340–61. [DOI] [PubMed] [Google Scholar]
- 2.McMahon SD, Swenski T, Bare K, Valido A, Asad S, Reddy LA, et al. Teacher-directed violence and anxiety and stress: predicting intentions to transfer and quit. School Psychol. 2023;39(5):530–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Moon B, Morash M, Jang JO, Jeong S. Violence against teachers in South korea: negative consequences and factors leading to emotional distress. Violence Vict. 2015;30(2):279–92. [DOI] [PubMed] [Google Scholar]
- 4.Hellfeldt K, Andershed H, Göransson S, Meehan A, Sverke M. Teacher-victimization in Swedish schools: Identification of risk factor. In: 6th EFCAP Congress, Young victims and young offenders: Prevention and intervention within families and institutions; 2018.
- 5.Arbejdstilsynet. National overvågning af arbejdsmiljøet blandt lønmodtagere 2023 [Internet]. Copenhagen: Arbejdstilsynet; 2023 [cited 2025 Apr 29]. Avaiable from: https://at.dk/arbejdsmiljoe-i-tal/national-overvaagning-af-arbejdsmiljoeet-blandt-loenmodtagere/
- 6.Longobardi C, Badenes-Ribera L, Fabris MA, Martinez A, McMahon SD. Prevalence of student violence against teachers: A meta-analysis. Psychol Violence. 2018;9(6):596–610. [Google Scholar]
- 7.McMahon SD, Martinez A, Reddy LA, Espelage DL, Anderman EM. Predicting and reducing aggression and violence toward teachers: Extent of the problem and why it matters. 2017. [Internet]. 2017 [cited 2025 Apr 30]. Available from: https://digitalcommons.sacredheart.edu/cgi/viewcontent.cgi?article=1026&context=socwk_fac
- 8.Reddy LA, Espelage DL, Anderman EM, Kanrich JB, McMahon SD. Addressing violence against educators through measurement and research. Aggress Violent Behav. 2018;(42):9–28.
- 9.McMahon SD, Worrell FC, Reddy LA, Martinez A, Espelage DL, Astor RA, et al. Violence and aggression against educators and school personnel, retention, stress, and training needs: National survey results. Am Psychol. 2024;79(7):903–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Huang FL, Eddy CL, Camp E. The role of the perceptions of school climate and teacher victimization by students. J Interpers Violence. 2017;5(23–24):5526–51. [DOI] [PubMed] [Google Scholar]
- 11.Kapa RR, Luke J, Moulthrop D, Gimbert B. Teacher victimization in authoritative school environments. J Sch Health. 2018;88(4):272–80. [DOI] [PubMed] [Google Scholar]
- 12.Yang C, Jenkins L, Fredrick SS, Chen C, Xie JS, Nickerson AB. Teacher victimization by students in china: A multilevel analysis. Aggress Behav. 2019;45(2):169–80. [DOI] [PubMed] [Google Scholar]
- 13.Astor RA, Benbenishty R, Capp GP, Watson KR, Wu C, McMahon SD, et al. How school policies, strategies, and relational factors contribute to teacher victimization and school safety. J Community Psychol. 2024;52(1):39–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Aust B, Andersen LPS, Laursen LL, Erdogan M, Kristiansen J, Nielsen HB. Undersøgelse af chikane, trusler om vold og fysisk vold rettet mod undervisningspersonale i udvalgte folkeskoler. Det Nationale Forskningscenter for Arbejdsmiljø (NFA) og Styrelsen for Undervisning og Kvalitet, Undervisningsministeriet; 2018.
- 15.Espelage D, Anderman EM, Brown VE, Jones A, Lane KL, McMahon SD, et al. Understanding and preventing violence directed against teachers: recommendations for a National research, practice, and policy agenda. Am Psychol. 2013;68(2):75–87. [DOI] [PubMed] [Google Scholar]
- 16.Casteel C, Peek-Asa C, Limbos MA. Predictors of nonfatal assault injury to public school teachers in Los Angeles City. Am J Ind Med. 2007;50(12):932–9. [DOI] [PubMed] [Google Scholar]
- 17.Dinkes R, Kemp J, Baum K. Indicators of school crime and safety: 2009. NCES 2010-012/NCJ 228478. National Center for Education Statistics; 2009.
- 18.Gerberich SG, Nachreiner NM, Ryan AD, Church TR, McGovern PM, Geisser MS, et al. Case-control study of student-perpetrated physical violence against educators. Ann Epidemiol. 2014;24(5):325–32. [DOI] [PubMed] [Google Scholar]
- 19.McMahon SD, Cafaro CL, Bare K, Zinter KE, Murillo YG, Lynch G, et al. Rates and types of student aggression against teachers: A comparative analysis of US elementary, middle, and high schools. Soc Psychol Educ. 2022;25(4):767–92. [Google Scholar]
- 20.Robers S. Indicators of school crime and safety [Internet]. DIANE Publishing; 2017 [cited 2025 Apr 30]. Available from: https://nces.ed.gov/pubs2016/2016079.pdf
- 21.McMahon SD, Martinez A, Espelage D, Rose C, Reddy LA, Lane K, et al. Violence directed against teachers: results from a National survey. Psychol Sch. 2014;51(7):753–66. [Google Scholar]
- 22.Martinez A, McMahon SD, Espelage D, Anderman EM, Reddy LA, Sanchez B. Teachers’ experiences with multiple victimization: identifying demographic, cognitive, and contextual correlates. J Sch Violence. 2016;15(4):387–405. [Google Scholar]
- 23.Reddy LA, Perry AH, Martinez A, McMahon SD, Bare K, Swenski T, et al. Student violence against paraprofessionals in schools: A Social-Ecological analysis of safety and Well-Being. Behav Sci. 2024;14(12):1181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Winding TN, Aust B, Andersen LPS. The association between pupils´ aggressive behaviour and burnout among Danish school teachers-the role of stress and social support at work. BMC Public Health. 2022;22(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Olivier E, Janosz M, Morin AJ, Archambault I, Geoffrion S, Pascal S, et al. Chronic and temporary exposure to student violence predicts emotional exhaustion in high school teachers. J Sch Violence. 2021;20(2):195–211. [Google Scholar]
- 26.Moon B, Saw G, McCluskey J. Teacher victimization and turnover: focusing on different types and multiple victimization. J Sch Violence. 2020;19(3):406–20. [Google Scholar]
- 27.Stevenson DJ, Neill JT, Ball K, Smith R, Shores MC. How do preschool to year 6 educators prevent and Cope with occupational violence from students? Aust J Educ. 2022;66(2):154–70. [Google Scholar]
- 28.Berlanda S, Fraizzoli M, de Cordova F, Pedrazza M. Psychosocial risks and violence against teachers. Is it possible to promote well-being at work? Int J Environ Res Public Health. 2019;16(22):4439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Huang FL, Eddy CL, Camp E. The role of the perceptions of school climate and teacher victimization by students. J Interpers Violence. 2020;35(23–24):5526–51. [DOI] [PubMed] [Google Scholar]
- 30.McMahon SD, Peist E, Davis JO, McConnell E, Reaves S, Reddy LA, et al. Addressing violence against teachers: A social-ecological analysis of teachers’ perspectives. Psychol Sch. 2020;57(7):1040–56. [Google Scholar]
- 31.Andersen LPS, Winding TN, Grytnes R. Pupils’ challenging behaviour towards their teachers: the role of teacher team collaboration in prevention. A qualitative study. Teach Teach Educ. 2025;155:104883. [Google Scholar]
- 32.Kouvonen A, Oksanen T, Vahtera J, Stafford M, Wilkinson R, Schneider J, Am, et al. J Epidemiol. 2008;167(10):1143–51. [DOI] [PubMed]
- 33.Oksanen T, Kouvonen A, Vahtera J, Virtanen M, Kivimäki M. Prospective study of workplace social capital and depression: are vertical and horizontal components equally important? J Epidemiol Community Health. 2010;64(8):684–89. [DOI] [PubMed] [Google Scholar]
- 34.Meng A, Clausen T, Borg V. The association between team-level social capital and individual‐level work engagement: differences between subtypes of social capital and the impact of intra‐team agreement. Scand J Psychol. 2018;59(2):198–205. [DOI] [PubMed] [Google Scholar]
- 35.Pihl-Thingvad J, Andersen LPS, Pihl-Thingvad S, Elklit A, Brandt LPA, Andersen LL. Can high workplace social capital buffer the negative effect of high workload on patient-initiated violence? Prospective cohort study. Int J Nurs Stud. 2021;120:103971. [DOI] [PubMed] [Google Scholar]
- 36.Török E, Rod NH, Ersbøll AK, Jensen JH, Rugulies R, Clark AJ. Can work-unit social capital buffer the association between workplace violence and long-term sickness absence? A prospective cohort study of healthcare employees. Int Arch Occup Environ Health. 2020;93(3):355–64. [DOI] [PubMed] [Google Scholar]
- 37.de Lange AH, Taris TW, Kompier MA, Houtman IL, Bongers PM. Different mechanisms to explain the reversed effects of mental health on work characteristics. Scand J Work Environ Health 2005;31(1):3–14. [DOI] [PubMed]
- 38.Andersen LP, Nøhr T, Aust B. Chikane, trusler og vold mod undervisere i Den Danske Folkeskole. Herning: Arbejdsmedicin, Regionshospitalet; 2019.
- 39.Borg V, Mateu N, Clausen T. Udvikling af en ny metode til undersøgelse af social kapital på arbejdspladsen [Internet]. Copenhagen: The National Research Centre for the Working Environment; 2014. [cited 2025 Apr 30]. Available from: https://nfa.elsevierpure.com/ws/portalfiles/portal/3649679/Borg_V_Udvikling_af_ny_metode_til_unders_gelse_af_social_kapital_2014.pdf
- 40.Wynne R, Clarkin N, Cox T, Griffeths A. Guidance on the prevention of violence at work [Internet]. Luxembourg: European Commision, DG-V; 1997 [cited 2025 Apr 30]. Available from: https://osha.europa.eu/sites/default/files/2022-03/violence-harassment-report.pdf
- 41.Menckel E, Viitasara E. Threats and violence in Swedish care and welfare. Magnitude of the problem and impact on municipal personnel. Scand J Caring Sci. 2002;16(4):376–85. [DOI] [PubMed] [Google Scholar]
- 42.Rasmussen CA, Hogh A, Andersen LP. Threats and physical violence in the workplace: A comparative study of four areas of human service work. J Interpers Violence. 2013;28(13):2749–69. [DOI] [PubMed] [Google Scholar]
- 43.Gadegaard CA, Andersen LP, Hogh A. Effects of violence prevention behavior on exposure to workplace violence and threats: a follow-up study. J Interpers Violence. 2018;33(7):1096–117. [DOI] [PubMed] [Google Scholar]
- 44.Andersen LP, Hogh A, Biering K, Gadegaard CA. Work-related threats and violence in human service sectors: the importance of the psycho-social work environment examined in a multilevel prospective study. Work. 2018;59(1):141–54. [DOI] [PubMed] [Google Scholar]
- 45.Field A. Discovering statistics using IBM SPSS statistics. 5th ed. London: Sage; 2018. [Google Scholar]
- 46.Østergaard JS. Challenges of inclusive education in Danish public schools: How teachers’ self-efficacy affects their attitudes and concerns about inclusive education. In: 4th Meeting on Quantitative Educational Research in Denmark (QERD): 2024.
- 47.Engsig TT. The evolution of inclusive education in Denmark: A history of inclusions—from rights-based to accountability and competition. In: 2016 International Summit on RTI and Inclusive Education: 2016.
- 48.Roewer A. Das Tabu brechen: Gewalt gegen Lehrkräfte. In: Beckmann U, Bentrup N, Gostovic-Storz M, Schneider W, editors. Das Tabu brechen. Berlin: Verband Bildung und Erziehung e. V.; 2017. [Google Scholar]
- 49.Mantzicopoulos P. Conflictual relationships between kindergarten children and their teachers: associations with child and classroom context variables. J Sch Psychol. 2005;43(5):425–42. [Google Scholar]
- 50.Tucker CM, Zayco RA, Herman KC, Reinke WM, Trujillo M, Carraway K, et al. Teacher and child variables as predictors of academic engagement among low-income African American children. Psychol Sch. 2002;39(4):477–88. [Google Scholar]
- 51.Scrivener J. Classroom management techniques. Cambridge: Cambridge University Press; 2012. [Google Scholar]
- 52.Gloede TD, Hammer A, Ommen O, Ernstmann N, Pfaff H. Is social capital as perceived by the medical director associated with coordination among hospital staff? A nationwide survey in German hospitals. J Interprof Care. 2013;27(2):171–76. [DOI] [PubMed] [Google Scholar]
- 53.Pijl SJ, Frostad P, Flem A. The social position of pupils with special needs in regular schools. Scand J Educ Res. 2008;52(4):387–405. [Google Scholar]
- 54.Kraut AI, Pedigo PR, McKenna DD, Dunnette MD. The role of the manager: what’s really important in different management jobs. Acad Manag Perspect. 2005;19(4):122–29. [Google Scholar]
- 55.Spector PE, Brannick MT. Common method variance or measurement bias? The problem and possible solutions. In: Buchanan DA, Bryman A, editors. The SAGE handbook of organizational research methods. London: Sage Publications; 2009. [Google Scholar]
- 56.Farrington DP, Loeber R. Some benefits of dichotomization in psychiatric and criminological research. Crim Behav Ment Health. 2000;10(2):100–22. [Google Scholar]
- 57.Andersen LP, Mikkelsen KL. Recall of occupational injuries: A comparison of questionnaire and diary data. Saf Sci. 2008;46(2):255–60. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.