The findings of this study extend current understanding of child and family predictors of later involvement in bullying. They show that early aggressiveness, good motor functioning and SES had an impact on involvement in bullying during early adolescence. Early anxiety decreased the risk of being a bully/victim at age 10/11. Children from intact families were more likely to be uninvolved at age 13/14.
Early Child and Family Risk Factors for Being a Bully, Victim, or Bully/victim
This study confirms previously described sex differences in bullying, i.e. that boys are more likely than girls to be bullies and bully/victims [
27,
28]. At early adolescence, girls were more likely than boys to be victims.
Early childhood anxiety decreased the likelihood of being a bully/victim at age 10/11, but had no statistically significant effects on any of the other outcomes at ages 10/11 and 13/14. One could hypothesize that the anxiety of these children makes it very unlikely for them to become a bully at that age, even in case of victimization. The effect of this early life factor seems to diminish when the child ages, and enters secondary school.
The findings further suggest that aggressive preschoolers were more likely to be a bully at age 10/11, a bully/victim at ages 10/11 and 13/14, or a victim at age 13/14. This confirms previous cross-sectional studies that consistently showed that proactive aggression is a main characteristic of bullies [
7,
27,
29]. Similarly, the association between early aggression and later victimization has been documented before [
6,
8,
30]. Our findings add to the available evidence that these effects persist during the transition from primary to secondary school. Moreover, low aggressiveness seems to be very predictive for being uninvolved.
Another noteworthy result of our study is that motorically skilful preschoolers were more likely to become a bully at age 10/11 while less motorically able preschoolers were more prone to victimization at age 10/11 as well as at age 13/14. Evidence on the association of preschool motor functioning and involvement in bullying later on was limited to one study showing a positive association between physical condition and bullying [
31]. Our findings indicate that these associations may have their origins in early life, likely due to the important role of motor skills in a child's emotional, behavioral and social functioning [
32,
33]. Motorically able children may receive more positive social feedback and recognition from peers, which is likely to improve their self-image and popularity among peers. These are frequently reported characteristics of bullies. In addition, good motor skills may provide children with physical means to bully [
34].
Poor motor skills have been shown to result in poor performance in both individual and team games and sports, which may reduce children's sense of competence. This in turn reduces success within peer groups and may increase the likelihood of victimization [
35]. Motorically able children may receive more positive social feedback and recognition from peers, which may improve their self-image and popularity among peers which may lead to bullying [
36]. In addition, good motor skills may provide children with physical means to bully. The reverse may increase the likelihood of victimization. Third, positive features, such as motor abilities, may create a context within which negative features, such as bullying, are interpreted in a more positive light [
37], again making bullying more likely. Our results also may be interpreted as that motor ability is more important at primary school (T1) than at secondary school (T2) which fits with the much higher emphasis on cognition at secondary school, compared with the stronger emphasis on play and physical activity at primary school.
Concerning family characteristics, parental SES seems to be associated with involvement in bullying. In line with previous work we found that children of lower SES were more often bullies and victims [
31]. The pathways leading to this require additional study.
At both time points, parental mental health was not associated with involvement in bullying. This finding contradicts previous research into the association between certain parental mental health characteristics such as depression and the quality of peer relations of their offspring [
38]. An explanation might be that the measures of bullying that we used are not affected by the mental state of the adolescent itself, which may partially reflect the parental mental state. This would imply that previous findings are due to information bias.
Strengths and limitations of the study
Notable strengths of our study are its large population-based sample of preadolescent boys and girls, and its focus on both bullying and victimization. Next to this, its longitudinal nature makes that our findings about early risk factors are more robust.
The main limitation of the present data is that we used retrospective reports of preschool behavior. As a result, parental report may have been affected by inaccuracies in their memory. This may have introduced additional random error or recall bias, if parental memory artefacts were affected by bully/victim status. Our use of peers as informants regarding bullying makes such a bias less likely, though.
Second, parents were asked to rate their child's early behavior relative to its peers, which may be subject to recall bias. However, previous studies have shown a good parental recall of early life factors such as maternal smoking during pregnancy, maternal smoking during pregnancy, for gestational age, and for birth weight [
39‐
41]. Despite this, our results thus need confirmation by use of prospectively collected data.
Implications
The results of the present study imply that preschool behavioral, emotional and motor problems, and family characteristics are related to involvement in bullying at a later age. Prevention of bullying and its consequences can be enhanced by focusing on risk groups in early life.
One of the main findings of this study is the predictive value of motor performance on involvement in bullying at an advanced age. Because it is inadequate and nearly impossible to intervene on the often superior physical status of bullies, the focus has to be on children with a poor motor performance which often persist throughout adolescence and into adulthood [
42]. Next to this, additional longitudinal research incorporating more detailed measures on motor performance is needed to assess the way in which motor skills affect involvement in bullying and have the potential to prevent victimization.