School Bullying and Dating Violence in Adolescents: A Systematic Review and Meta-Analysis

Abstract

Involvement in bullying perpetration or victimization could be risk factors for perpetration or victimization in early romantic relationships that emerge within an evolving peer group. Nevertheless, research on this topic is still in its early stages. This study was conducted to fill these gaps in knowledge through a comprehensive research synthesis. After systematic searches and application of inclusion and exclusion criteria, 23 projects that reported relations between bullying (perpetration and victimization) and dating violence (perpetration and victimization) were included in this meta-analysis. Bullying perpetration was related to dating violence perpetration (odds ratio [OR] = 1.98), and this relation held after adjusting for covariates (OR = 1.29). Moderator analyses showed that this effect was significant for males and females and stronger in cross-sectional studies. There was also a significant relation between bullying perpetration and dating violence victimization (OR = 2.59), but this was much weaker after controlling for covariates (OR = 1.09) and stronger for males and in longitudinal studies. Bullying victimization was related to dating violence victimization (OR = 2.51), also after adjusting for covariates (OR = 1.96), stronger for females and longitudinal projects. The relation between bullying victimization and dating violence perpetration was not statistically significant (unadjusted OR = 1.43, adjusted OR = 1.01). More research is needed to confirm these results, but this meta-analysis suggests that bullying and dating violence could be different behavioral manifestations, in different evolutionary moments and in different contexts, of the same underlying antisocial or violent dispositions, although longitudinal studies are needed to confirm this. Tailored and comprehensive interventions could be useful to tackle both problem behaviors.

Keywords

bullying dating violence systematic review meta-analysis adolescence

Research suggests that school bullying could be related to teen dating violence, but studies focused on this relation yield inconclusive results. Many antisocial behaviors share similar risk and protective factors (Catalano et al., 2012), and both bullying and dating violence are prevalent in adolescence and emerge in the evolving peer group. New research focused on the relation between bullying and dating violence and possible moderators, such as gender or the type of bullying or dating violence (e.g., physical, verbal, sexual), is urgently needed to confirm this possible relation. A research synthesis that summarizes the existing evidence, identifies what is known, and suggests what should be studied next can be especially useful to understand whether bullying and dating violence are related. The current study includes a systematic review and a meta-analysis on the topic, including an analysis of the overall effect sizes of the relations between bullying and dating violence and moderator analyses.

Bullying is defined as a direct or indirect intentional aggressive behavior that is repeated over time on a long-term basis, with power imbalance between perpetrators and victims (Olweus, 1999; Smith, Cowie, Olafsson, & Liefooghe, 2002). These characteristics make it possible to distinguish bullying from other types of aggressive behaviors such as sporadic fights among young people who have similar physical, psychological, or social strength. Research in this field has identified specific bullying roles (Salmivalli, Lagerspetz, Björkqvist, Österman, & Kaukiainen, 1996) that are usually divided into perpetrators, victims, bully/victims, and bystanders.

The number of studies about bullying has increased greatly in the past decades (Zych, Ortega-Ruiz, & Del Rey, 2015a), and research on school bullying is now considered to be an important subfield in studies focused on violence. It was found that bullying is prevalent across the world (Craig et al., 2009). A meta-analysis conducted by Modecki, Minchin, Harbaugh, Guerra, and Runions (2014), that included more than 80 projects from different countries, showed that around 35% of students report bullying victimization and around 35% report perpetration. Rates of bullying perpetration and victimization are usually higher for boys when compared to girls (Smith, López-Castro, Robinson, & Görzig, 2018) and higher in primary school when compared to adolescence (Zych, Ortega-Ruiz, & Del Rey, 2015b).

Several studies have reported detrimental short- and long-term consequences of bullying such as internalizing problems, social maladjustment, and poor health (Analitis et al., 2009; Wolke, Copeland, Angold, & Costello, 2013). A meta-analysis of longitudinal studies conducted by Ttofi, Farrington, and Lösel (2012) found that both bullying perpetration and victimization predicted violence later in life. Specifically, bullying perpetration predicted increased violence about 6 years later by about two thirds and bullying victimization predicted increased violence by about one third. Thus, bullying perpetration and victimization are related to other antisocial behaviors, but research about the role of bullying within a pattern of antisocial behaviors is still in its early stages. Given that both bullying and dating violence are present and prevalent in adolescence, the current study aims to discover whether school bullying is related to another antisocial behavior—dating violence.

It is reasonable to suggest that bullying and dating violence are related, taking into account the life-course development of aggressive behavior. From a developmental perspective, Dunphy (1963) suggested that the early romantic relationships generally emerge in the broader context of adolescent peer relationships. Thus, behaviors acquired and established within the peer group can be easily transferred to the new context of intimate relationships. Peers are crucial agents in the socialization process of children and adolescents. Thus, the psychosocial dynamics of school bullying, in which the peer group can reinforce aggressive behaviors of the perpetrators (Salmivalli, 2010), might make violence appear socially desirable.

Using a social learning framework, Malik, Sorenson, and Aneshensel (1997) found that exposure to violence in different contexts is related to dating violence perpetration and victimization. They concluded that this might be caused by normative beliefs about aggression being perceived as justified. After a literature review on the topic, Leen et al. (2013) concluded that peer attitudes toward violence and peer influence are the most important risk factors for adolescent dating violence. Nevertheless, the cross-sectional nature of most of the studies only makes it possible to approach prediction on a theoretical basis. It is also possible that bullying and dating violence share early risk factors, and if these factors are accounted for, the contribution of bullying to later teen dating violence could be small.

Many antisocial behaviors share risk and protective factors (Catalano et al., 2012). Although both stability and change over time in aggressive behavior were described in different research studies, interpersonal differences in aggression are relatively stable (Eisner & Malti, 2015; Zych et al., 2018). Thus, it is possible that the perpetration of bullying and dating violence are two expressions of one underlying construct in different contexts and stages of human development.

In the context of peer relationships and friendship, the first intimate partner relationships appear in adolescence. Many of these relationships are prosocial and desirable (Viejo, Ortega-Ruiz, & Sanchez, 2015), but some adolescent couples display aggressive behaviors toward each other. Dating violence is defined as sexual, physical, or emotional aggressive behavior perpetrated in an intimate (stable or unstable) relationship (Sánchez, Ortega-Rivera, Ortega-Ruiz, & Viejo, 2008).

The prevalence rates of dating violence vary greatly among the studies, depending on the way in which it is measured and defined (Jennings et al., 2017). A meta-analysis focused on teen dating violence found that around 20% of adolescents are involved in physical dating violence and around 9% are involved in sexual dating violence (Wincentak, Connolly, & Card, 2016). Sex differences in intimate partner violence are usually found, with slightly more women using physical aggression and more men causing injury to their female partners according to meta-analytic findings (Archer, Ireland, Monks, & Gilling, 2000). A meta-analysis of research on teen dating violence found that girls perpetrated more physical dating violence than boys, boys perpetrated more sexual dating violence than girls, there were no gender differences regarding physical victimization, and girls were more sexually victimized than boys (Wincentak et al., 2016). A study with a national sample of adolescents in the United States showed that serious dating violence victimization, such as being a victim of physical and sexual assault, was much more common in girls than boys (Wolitzky-Taylor, et al., 2008). Both boys and girls can be perpetrators and victims, and research suggests that simultaneous involvement in perpetration and victimization is common (Menesini, Nocentini, Ortega-Rivera, Sanchez, & Ortega, 2011).

A recent meta-analysis of 100 primary studies showed that, overall, anti-bullying interventions are effective in decreasing bullying (Gaffney, Ttofi, & Farrington, 2019). A systematic review of studies about dating violence in young people showed that most of the interventions yielded mixed results with significant desirable short-term but often not long-term results (Jennings et al., 2017). A recent meta-analysis showed that school programs have desirable effects on dating violence attitudes and knowledge but not on the behaviors (De La Rue, Polanin, Espelage, & Pigott, 2017), although some programs decreased self-reported dating violence behaviors (e.g., Wolfe et al., 2009).

Both anti-bullying and anti-dating violence interventions are crucial for the desirable social and emotional development of young people, but research on these two types of aggressive behavior is rather compartmentalized, making it difficult to see a bigger picture regarding the prevention of patterns of antisocial behaviors. Understanding the relation between bullying and dating violence, including perpetration and victimization, could be a landmark in tailored and comprehensive developmental prevention programs for violence that could be implemented in schools. Thus, the current study was conducted to synthesize knowledge about the relation between bullying and dating violence through a systematic review and a meta-analysis of cross-sectional and longitudinal studies, including direct and unique relations and adjustments for confounders, together with moderators.

A meta-analysis focused on these two types of antisocial behaviors is especially important because projects on school bullying and dating violence are currently two important subfields of research about violence and they are objects of developmental interventions. Bullying and dating violence are both relational antisocial behaviors and their association might be especially important when analyzing desirable and undesirable interpersonal interactions. To our knowledge, this is the first systematic review and meta-analysis focused on the relation between school bullying and dating violence.

Method

Systematic Searches

Systematic searches were conducted in databases such as Web of Science (searches in “topic” including MEDLINE and SciELO), Google Scholar (title), and Scopus (titles, abstracts, and key words). Articles were searched with no time restriction up to October 2016. A combination of key words was used, including “bullying,” “bully,” “peer violence,” and “peer victimization” combined with “dating,” “romantic,” “couple,” and “partner.” The total numbers of located studies, studies excluded and retained for full-text screening after reviewing abstracts, and titles and studies included and excluded in the final stage are shown in Figure 1.

Figure 1.

A flowchart showing different stages of literature searches.

Inclusion and Exclusion Criteria

Studies were included if they measured the relation between bullying and teen dating violence. Studies were excluded if they did not present data on this relation.

Research with school students was included, and studies conducted in other contexts such as street violence were excluded (e.g., Antônio, Koller, & Hokoda, 2012).

Only studies that explicitly focused on bullying and not on other types of school violence were included. Other types of school violence were excluded (e.g., Cava, Buelga, & Carrascosa, 2015).

Only studies conducted with children and adolescents up to age 21 were included. Research with adult participants was excluded (e.g., Rodríguez Franco, Antuna Bellerin, Lopez-Cepero Borrego, Rodríguez Díaz, & Bringas Molleda, 2012).

Regarding dating violence, included studies focused on dating, excluding research with married couples or couples who lived together.

All types of bullying and dating violence were included (e.g., physical, psychological, and sexual).

Only empirical studies with primary data were included, and theoretical research was excluded.

Studies were included if their publication language was English or Spanish.

All the searches, screening, and coding were done twice by two independent researchers (I.Z. and E.V.). Information such as study year, location, participants, and roles in bullying and dating violence were coded. Also, all the coefficients used for the analyses of unadjusted and adjusted effect sizes were coded. All the discrepancies were discussed and resolved.

Data Analyses

Effect sizes were described using the odds ratios (OR) with 95% confidence intervals (CI). An OR above 1 indicated that participants who were involved in a bullying role had higher odds of involvement in a dating violence role. An OR below 1 indicated that participants involved in a bullying role had lower odds of involvement in a dating violence role. An OR of 1 indicated no relation between these two aggressive behaviors and a CI that included 1 indicated a nonsignificant relation.

ORs were calculated for each of the included studies and overall effect sizes were calculated for relations of bullying perpetration and victimization with dating violence perpetration and victimization. Unadjusted effect sizes were calculated from univariate statistics from the primary studies (e.g., correlations, comparing means) and adjusted effect sizes were calculated from multivariate statistics (regression coefficients) to discover unique relations among variables.

Comprehensive meta-analysis software was used to create forest plots and to calculate overall effect sizes under fixed and random effects methods. In fixed effects, studies are weighted according to their sample sizes (larger studies have larger weights), but this method has important limitations because it assumes a normal and homogeneous distribution of the effect sizes. The random effects method adjusts for heterogeneity, but all studies, large and small, can have similar weights. Therefore, multiplicative variance adjustment (MVA) was used to overcome these difficulties (Farrington & Welsh, 2013). Unlike the random effects method, the MVA method exactly fits the data and exactly adjusts for the heterogeneity of the effect sizes. The MVA method yields the same weighted mean effect size as the fixed effects method but has a wider confidence interval because the fixed effects variance is multiplied by (Q/df). This method was first used in the medical field by Shore, Gardner, and Pannett (1993). Thus, MVA adjusts exactly for heterogeneity and appropriately gives more weight to larger studies. This correction was made using formulas (see Farrington & Welsh, 2013; Gaffney, Ttofi, & Farrington, 2018; Zych, Baldry, Farrington, & Llorent, 2018, for details). MVA ORs are used for the interpretation of the results, but fixed and random methods ORs are also presented for interested readers. Moderators such as gender, design (longitudinal vs. cross sectional), type of dating violence, and location were tested. When available, adjusted coefficients were used for moderator analyses.

All studies provided data for the calculation of the effect sizes. Nevertheless, Espelage, Low, Anderson, and De La Rue (2014) provided data to calculate effect sizes where results were significant but not where results were nonsignificant (they were reported as “nonsignificant” without details). Following Wilson, Gottfredson, and Najaka (2001), given that direction and sample size were known, it was assumed that the effect was between zero and the largest nonsignificant effect size. Thus, the midpoint value was used and sensitivity analyses were run to confirm that this did not substantially affect the weighted mean effect size.

The fail-safe N test was used to search for possible publication bias (Lipsey & Wilson, 2000). This test makes it possible to calculate the number of nonsignificant results that would need to be included in a meta-analysis to make the p value nonsignificant.

Results

The systematic review included 23 studies. Most of them were conducted in the United States (n = 15) and Canada (n = 6), with only one study published in Norway and one in South Africa. Seven studies focused on bullying perpetration only, nine focused on victimization only, and seven focused on both perpetration and victimization. Ten studies focused on dating violence victimization only, six on perpetration only, and seven studies reported results on both dating violence victimization and perpetration (see details in Table 1).

Table 1.

Primary Studies Reporting the Relation Between Bullying and Dating Violence With Adjusted and Unadjusted Coefficients.

Study and Location	Participants	Roles in Bullying	Roles in Dating Violence	Main Findings
Coker et al. (2014), USA	14,190 secondary education students, Grades 9–12, enrolled in 26 schools (44.4% boys)	Perpetration and victimization	Perpetration and victimization (physical and psychological)	Among 1,620 children who were victims at schools (more than twice), 906 were dating violence victims and 554 were dating violence perpetrators. Among 12,549 children who were not victims of school bullying, 3,821 were dating violence victims and 2,306 were perpetrators of dating violence. Among 1,241 students who bullied others in school (more than twice), 716 were dating violence victims and 547 were dating violence perpetrators. Among 12,933 students who did not perpetrate school bullying, 4,015 were dating violence victims and 2,313 were dating violence perpetrators. No adjusted coefficients or gender differences were available.
Connolly, Pepler, Craig, and Taradash (2000), Canada	Among 1,758 adolescent participants of the study (M _age = 12.6 years, SD = 1.2, 49.7% boys), 110 bullies and 79 comparison adolescents were selected in this analysis	Perpetration	Perpetration and victimization (physical and social dating violence)	Among school bullies (n = 110), 40 adolescents were perpetrators of dating violence and 44 were victims of dating violence. Among comparison adolescents (n = 79), 7 were perpetrators of dating violence, and 9 were victims of dating violence. Adjusted coefficients and gender comparison were not available.
Connolly et al. (2015), Canada	509 students in Grades 7–8 (M _age = 12.37, SD = 0.63) enrolled in four schools (49% boys)	Victimization	Victimization (physical, emotional, and verbal)	There was no significant relationship between bullying victimization and dating violence victimization (r = .01, ns). Adjusted coefficients and gender comparison were not available.
Cutbush, Williams, and Miller (2016); Cutbush and Williams (2016); and Miller et al. (2013), USA	612 middle school students, Grade 7 (49.6% boys) followed up 6 months and 1 year later	Perpetration and victimization	Perpetration and victimization (physical, psychological, electronic, and sexual)	In boys, bullying perpetration was related to electronic dating violence perpetration (r = .20, p < .05) and victimization (r = .25, p < .05), emotional dating violence perpetration (r = .21, p < .05) and victimization (r = .26, p < .05), and physical dating violence perpetration (r = .15, p < .05) and victimization (r = .08, ns) measured 1 year later. In boys, bullying victimization was related to electronic dating violence perpetration (r = .16, p < .05) and victimization (r = .21, p < .05), emotional dating violence perpetration (r = .18, p < .05) and victimization (r = .21, p < .05), and physical dating violence perpetration (r = .12, ns) and victimization (r = .12, ns) measured 1 year later. In girls, no significant relationships were found between bullying perpetration and electronic dating violence perpetration (r = .05) and victimization (r = .00), emotional dating violence perpetration (r = .07) and victimization (r = .07), and physical dating violence perpetration (r = .08) and victimization (r = .00) measured 1 year later. In girls, bullying victimization was not significantly related to electronic dating violence perpetration (r = .05) and victimization (r = .08), emotional dating violence perpetration (r = .02) and victimization (r = .05), and physical dating violence perpetration (r = .06) and victimization (r = .04) measured 1 year later. Bullying perpetration predicted dating violence perpetration 1 year later (β = .39, SE = .10 in girls and β = .12, SE = .13 in boys) after controlling for dating violence 6 months earlier, gender, alcohol use, race, and sexual harassment.
Debnam, Waasdorp, and Bradshaw (2016), USA	17,780 adolescents, Grades 9–12, enrolled in 58 schools (50.1% boys)	Victimization (relational, verbal, and physical)	Victimization (physical and emotional)	Relational bullying victimization was related to higher levels of physical dating violence (OR = 1.36, 95% CI [1.13, 1.62]) and emotional dating violence (OR = 1.94, 95% CI [1.66. 2.26]). Verbal bullying victimization was related to higher levels of physical dating violence (OR = 1.34, 95% CI [1.14, 1.58]) and emotional dating violence (OR = 1.51, 95% CI [1.28, 1.78]). Physical bullying victimization was related to higher levels of physical dating violence (OR = 2.40, 95% CI [2.03, 2.84]) and emotional dating violence (OR = 1.79, 95% CI [1.52, 2.10]). These coefficients were adjusted for variables such as age, sex, race, student’s and teacher’s intervention in bullying; perception of school safety, free, and reduced priced meals; percentage of minority; school size; and the percentage of suspensions in the school. Unadjusted coefficients or coefficients for boys and girls were not available.
Ellis and Wolfe (2015), Canada	A subsample of 88 (of the 585) students (50% boys) who reported being involved in a stable dating relationship and agreed be observed while performing a task (M _age = 15.19 years, SD = 0.82)	Perpetration	Perpetration and victimization	Bullying perpetration was related to more dating violence perpetration (r = .15, p < .001) and to more dating violence victimization (r = .11, p < .001). After controlling for age and sex, bullying was significantly related to higher levels of dating violence perpetration in boys (β = .09, p < .001), but not in girls (β = .02, p > .05). After controlling for age and sex, differences between girls and boys were not significant and therefore were not reported. For the whole sample, bullying was not significantly related to dating violence victimization (β = .11, p > .05).
Espelage and Holt (2007), USA	684 adolescents in Grades 7–12 (47% boys) enrolled in two schools	Perpetration and victimization	Victimization (physical and emotional)	Regarding physical dating victimization, children not involved in bullying scored lower (M = 1.30, SD = 3.66) than bullies (M = 3.50, SD = 5.59) and bully/victims (M = 7.93, SD = 12.18) but not significantly lower than victims (M = 2.42, SD = 4.66). Regarding emotional abuse, children not involved in bullying scored lower (M = 2.97, SD = 4.61) than bullies (M = 4.84, SD = 4.85), bully/victims (M = 7.13, SD = 6.85) but not significantly lower than victims (M = 4.73, SD = 6.17). Adjusted coefficients or gender differences were not reported.
Espelage, Low, Anderson, and De La Rue (2014), USA	1,162 students enrolled in four schools, Grades 9–11 and 1 year later (51% boys)	Perpetration	Perpetration (physical, verbal/emotional relational, and sexual)	After controlling for family conflicts, sibling aggression and delinquent friends, in girls, bullying perpetration was related to more verbal dating violence (β = .20, p < .01 in Grades 9–11, nonsignificant 1 year later), physical dating violence was related to more bullying perpetration (nonsignificant in Grades 9–11, β = .19, p < .01 1 year later), and sexual dating violence was found to be related to more bullying (β = .27, p < .01 in Grades 9–11; β = .22, p < .01, 1 year later). In boys, verbal dating violence was related to more bullying (β = .26, p < .01 in Grades 9–11, nonsignificant 1 year later), physical dating violence was related to more bullying (β = .22, p < .01 in Grades 9–11, nonsignificant 1 year later), and sexual dating violence was not significantly related to bullying. Unadjusted coefficients and data for nonsignificant relationships were not available.
Foshee, Benefield, et al. (2016), USA	2,581 dating adolescents in Grades 8, 9, and 10 (13–16 years old) enrolled in three rural schools, followed up in the next semester. Percentages of girls and boys not reported	Perpetration and victimization	Perpetration (physical)	Bullying perpetration was related to higher levels of dating violence perpetration in the next semester (r = .04, p < .01). Bullying victimization was not related to higher levels of dating violence perpetration (r = .01, p > .05). After controlling for anger, depression, anxiety, lifetime dating violence, sex, race, parent education, grade, family aggression, and family structure, bullying perpetration (β = .01, p > .05) or victimization (β = .00, p > .05) did not predict dating violence perpetration.
Foshee, McNaughton, et al. (2016), USA	399 adolescents (M _age = 13.6 years, 36% boys)	Perpetration	Perpetration (physical)	Bullying was related to higher levels of physical dating violence (r = .16, p = .001). No adjusted coefficients or gender differences were reported.
Foshee et al. (2014), USA	1,154 adolescents (47% boys) who did not report dating violence perpetration in Grade 6, followed up 2 years later	Direct and indirect perpetration	Perpetration (physical)	Direct (r = .16, p < .001) and indirect (r = .08, p = .01) bullying perpetration was related to higher levels of dating violence perpetration reported 2 years later. After controlling for sex, race, parent education, living in a one parent family, and family conflict, indirect bullying did not significantly predict dating violence perpetration (OR = 1.26, 95% CI [0.93, 1.73]) and direct bullying predicted higher levels of dating violence perpetration (OR = 1.36, 95% CI [1.11, 1.68].
Hellevik and Øverlien (2016), Norway	549 dating adolescents, 50.2% girls, M _age = 15.2 years (SD = 0.90)	Victimization	Victimization (psychological, physical, digital, and sexual)	Bullying victimization was related to psychological (OR = 2.97, 95% CI [1.94, 4.54]), physical (OR = 2.97, 95% CI [1.69, 5.22]), digital (OR = 2.37, 95% CI [1.58, 3.55), and sexual (OR = 2.98, 95% CI [1.85, 4.82]) dating victimization. After controlling for gender, violence at home, sending sexual message, having older parents, aggressive friends, and academic achievement, bullying victimization was related to psychological (OR = 1.75, 95% CI [1.08, 2.85]) and sexual (OR = 1.86, 95% CI [1.06, 3.25]) dating victimization. Relationship to physical (OR = 1.56, 95% CI [0.81, 3.01]) and digital (OR = 1.33, 95% CI [0.82, 2.16]) dating violence was no longer significant.
Holt and Espelage (2003), USA	503 students in Grades 9–12 (45% boys) enrolled in one school	Victimization	Victimization (physical and emotional)	In boys, bullying victimization was related to more physical dating victimization (r = .14, p < .05) and not significantly related to emotional dating victimization (r = .07, ns). In girls, bullying victimization was significantly related to physical (r = .12, p < .05) and emotional (r = .27, p < .01) dating victimization. No adjusted coefficients were provided.
Leadbeater, Banister, Ellis, and Yeung (2008), Canada	664 adolescents (M _age = 15.5, SD = 1.93, 49% boys)	Perpetration and victimization (physical and relational)	Perpetration (emotional) and victimization (emotional and physical)	Correlations were reported between physical dating victimization and physical peer victimization (r = .16, ns), relational peer victimization (r = .19, p < .05), physical bullying perpetration (r = .19, p < .05), and relational bullying perpetration (r = .21, p < .05). Correlations were also described between emotional dating victimization and physical peer victimization (r = .15, ns), relational peer victimization (r = .07, ns), physical bullying perpetration (r = .17, p < .05), and relational bullying perpetration (r = .24, p < .05). Correlations were reported between emotional dating aggression and physical peer victimization (r = .09, ns), relational peer victimization (r = .10, ns), physical bullying perpetration (r = .17, p < .05), and relational bullying perpetration (r = .26, p < .05). After controlling for age, gender and direct dating aggression, direct peer victimization (β = .18, ns), relational peer victimization (β = −.13, ns), and direct bullying perpetration (β = −.00, ns) were not significantly related to relational dating violence, while relational bullying perpetration was related to higher levels of relational dating violence (β = .21, p < .05). After controlling for age, gender, and relational dating victimization, direct peer victimization (β = −.09, ns), relational peer victimization (β = 19, ns), direct bullying perpetration (β = .06, ns), and relational bullying perpetration (β = .04, ns) were not significantly related to physical dating victimization. After controlling for age and gender, direct peer victimization (β = .02, ns), relational peer victimization (β = −.00), and direct bullying perpetration (β = .14, ns) were not related to relational dating violence perpetration, while relational bullying perpetration (β = .27, p < .01) was significantly related to higher levels of relational dating violence perpetration. Gender differences were not available.
Lozano (2012), USA	1,542 Latin American girls between 14 and 18 years	Victimization	Victimization (physical)	Bullying victimization was related to higher rates of dating victimization, χ²(1) = 19.18, p < .01; yes dating yes bullying = 40, no dating, yes bullying = 192; yes dating no bullying = 97, no dating no bullying = 1,111.
Martin-Storey (2015), USA	10,493 adolescents, Grades 9–12 (48% boys)	Victimization	Victimization (physical and sexual)	After controlling for sexual identity, age, race, the number of fights, the number of sexual partners, and binge drinking, bullying was related to higher rates of dating violence in girls (n = 5,368; β = .22, p < .01) and in boys (n = 5,078; β = .15, p < .01). Unadjusted coefficients were not provided.
Melançon and Gagné (2011), Canada	278 enrolled in three schools (M _age = 14.2, SD = 1.1, 53% boys)	Victimization	Victimization (psychological, physical, and sexual)	Bullying victimization was related to more dating victimization (r = .34, p < .01). No adjusted coefficients or gender differences were available.
Niolon et al. (2015), USA	1,653 dating adolescents in Grades 6, 7, and 8 (47% boys)	Perpetration and victimization	Perpetration (verbal/emotional, physical, threatening, relational, sexual, and stalking)	Correlations were reported between bullying perpetration and threatening (r = .08, ns in girls and r = .04, ns in boys), verbal/emotional dating violence (r = .16, p < .05 in girls and r = .03, ns in boys), relational (r = .00, ns in both genders), physical (r = .09, ns in girls and r = −.03 in boys), sexual (r = .18, p < .05 in girls and r = .05, ns in boys), and stalking (r = .00, ns in girls and r = −.05, ns in boys). Correlations were reported between bullying victimization and threatening (r = −.08, ns in girls and r = −.08, ns in boys), verbal/emotional (r = −.09, p < .05 in girls and r = −.07, ns in boys), relational (r = −.09, p < .05 in girls and r = −.04, ns in boys), physical (r = −.11, p < .05 in girls and r = −.09, ns in boys), sexual (r = −.15, p < .05 in girls and r = −.07, ns in boys), and stalking (r = −.08, ns in girls and r = −.07, ns in boys). Bully/victims scored high on threatening (r = .15, p < .05 in both girls and boys), verbal/emotional (r = .16, p < .05 in girls and r = .27, p < .05 in boys), relational (r = .18, p < .05 in girls and r = .15, p < .05 in boys), physical (r = .18, p < .05 in girls and r = .30, p < .05 in boys), sexual (r = .08, ns in girls and r = .21, p < .05 in boys), and stalking (r = .07, ns in girls and r = .08, ns in boys). After controlling for variables such as geographic area, grade, race, exposure to family and community violence, delinquency, weapon carrying, alcohol and drug use, emotional symptoms, sex initiation, and attitudes toward female and male violence, and the relationships between bullying perpetration and dating violence in boys (n = 268) were threatening OR = 3.40 (95% CI [0.90, 12.90]), verbal/emotional OR = 1.32 (95% CI [0.53, 3.28]), relational OR = 2.41 (95% CI [0.57, 10.21]), physical OR = 1.81 (95% CI [0.49, 6.62]), sexual OR = 2.34 (95% CI [0.74, 7.41]), and stalking OR = 0.81 (95% CI [0.14, 4.57]). Bullying victimization threatening OR = 1.19 (95% CI [0.19, 7.37]), verbal/emotional OR = 0.74 (95% CI [0.29, 1.89]), relational OR = 2.86 (95% CI [0.64, 12.83]), physical OR = 1.82 (95% CI [0.39, 8.59]), sexual OR = 1.53 (95% CI [0.39, 5.99]). Bully/victims threatening OR = 3.78 (95% CI [1.18, 12.13]), verbal/emotional OR = 3.41 (95% CI [1.5, 7.8]), relational OR = 3.22 (95% CI [1, 12.35]), physical 5.96 (95% CI [2.18, 16.35]), sexual OR = 3.59 (95% CI [1.41, 9.18]), stalking OR = 0.96 (95% CI [0.27, 3.37]). In girls (n = 413), coefficients between bullying perpetration and dating violence were threatening OR = 1.53 (95% CI [0.57, 4.1]), verbal/emotional OR = 10.02 (95% CI [2.13, 47.21]), relational OR = 1.76 (95% CI [0.53, 5.79]), physical OR = –2.41 (95% CI [1.03, 5.67]), sexual OR = 4.56 (95% CI [1.17, 17.67]). Bullying victimization: threatening OR = 0.93 (95% CI [0.33, 2.62]), verbal/emotional OR = 0.43 (95% CI [0.68, 3]), relational OR = 0.28 (95% CI [0.06, 1.19), physical OR = 1.7 (95% CI [0.77, 3.75]), sexual OR = 0.20 (95% CI [0.02, 2.07]), and stalking OR = 0.38 (95% CI [0.06, 2.42]). Bully/victims threatening OR = 1.71 (95% CI [0.75, 3.89]), verbal/emotional OR = 2.93 (95% CI [1.41, 6.08]), relational OR = 2.02 (95% CI [0.79, 5.13]), physical OR = 2.8 (95% CI [1.44, 5.44]), sexual OR = 2.02 (95% CI [0.58, 6.97]), and stalking OR = 1.29 (95% CI [0.41, 4.08]).
Pepler et al. (2006), Canada	1,896 students in Grades 6–8 (53% boys) and Grades 9–12 (49% boys)	Perpetration	Perpetration (direct and indirect)	In elementary school, 24% of boys and 22% of girls classified as bullies perpetrated social dating aggression (vs. 8% of nonbullies in boys and 7% in girls). In secondary schools, 25% of boys and 43% of girls classified as bullies perpetrated social dating aggression (vs. 5% of nonbullies among boys and 40% of nonbullies in girls). In elementary school, 10% of male and 9% of female bullies perpetrated physical dating aggression (vs. 4% of male and 1% of female nonbullies). In secondary school, 13% of male and 23% of female bullies perpetrated physical dating aggression (vs. 3% of male and 12% of female nonbullies). Adjusted coefficients were not available.
Peters, Hatzenbuehler, and Davidson (2017), USA	11,570 adolescents enrolled in 102 schools, Grades 9–12 (47% boys)	Victimization	Victimization	Victims of school bullying had higher odds of being dating violence victims (OR = 2.5, 95% CI [2.1, 2.9]). This relationship was still present after controlling for gender, race/ethnicity, sexual identity, and age (OR = 2.6, 95% CI [2.2, 3.1]).
Shamu et al. (2016), South Africa	2,245 dating adolescents (49% girls), 12–19 years old.	Perpetration	Victimization in girls and perpetration in boys (sexual and physical)	Bullying perpetration was related to dating violence victimization in girls (OR = 2.10, 95% CI [1.67, 2.66]) and dating violence perpetration in boys (OR = 1.85, 95% CI [1.49, 2.29]). After controlling for alcohol use, multiple partners, partner communication, school engagement, gender inequitable attitudes, experience of violence (school and home), childhood physical and sexual abuse, and bullying perpetration were related to more dating violence victimization in girls (OR = 0.93, 95% CI [0.88, 0.99]) and the relationship with dating violence perpetration in boys was not significant (no details).
Vivolo-Kantor et al. (2016), USA	9,889 students, Grades 9–12 (51% boys)	Victimization	Victimization (physical and sexual)	Among females, 44.3% of dating victims reported bullying victimization. Among boys, 29% of dating victims reported bullying victimization. Adjusted coefficients were not provided.
Yahner, Dank, Zweig, and Lachman (2015), USA	5,647 students enrolled in 10 schools, Grades 7–12 (49% boys)	Perpetration and victimization	Perpetration and victimization (physical, sexual, psychological, and cyber victimization)	Among adolescents not involved in bullying perpetration, 8.5% reported physical dating violence perpetration, 10.6% psychological dating abuse, 0.5% sexual coercion, and 3.9% cyber dating abuse. Among adolescents who reported bullying perpetration, 20.6% reported physical dating violence perpetration, 26.6% psychological dating perpetration, 3.3% sexual coercion, and 13.9% cyber dating perpetration. Among adolescents not involved in bullying victimization, 13.3% reported physical dating victimization, 21.5% psychological dating victimization, 5.2% sexual victimization, and 10.1% cyber dating victimization. Among adolescents who reported bullying victimization, 24.5% reported physical dating victimization, 39.7% psychological dating victimization, 11.8% sexual victimization, and 23% cyber dating victimization. No gender differences or adjusted coefficients were available.

Note. CI = confidence interval; OR = odds ratio.

Bullying Perpetration Versus Dating Violence Perpetration

A total of 12 studies reported the unadjusted relations and 8 studies reported the adjusted relations between bullying perpetration and dating violence perpetration (see Figure 2). It was found that bullies had higher odds of scoring high in dating violence perpetration (unadjusted OR _MVA = 1.98, 95% CI [1.48, 2.65], z = 4.55) and this relation held even after controlling for covariates (adjusted OR _MVA = 1.29, 95% CI [1.07, 1.56], z = 2.62). The fail-safe N test showed that around 1,031 missing studies with no effect would need to be added to make the unadjusted effect size nonsignificant. Around 90 studies with no effect would need to be added to make the adjusted effect nonsignificant.

Figure 2.

Unadjusted and adjusted effect sizes for the relation between bullying perpetration and dating violence perpetration.

Studies were significantly heterogeneous, (unadjusted Q(11) = 220.66, p < .01, I ² = 95.02; adjusted Q(7) = 34.72, p < .01, I ² = 79.84), and moderator analyses were performed (see Table 2). It was found that the relation between high bullying and dating violence perpetration was stronger for females. Cross-sectional studies yielded a stronger effect than longitudinal studies where the effect was nonsignificant using the MVA. Effects were stronger for electronic and emotional dating violence. Greater effect sizes were found in Canadian studies. These results should be interpreted with caution given that the number of studies differed greatly among the moderators (e.g., 10 cross-sectional vs. 3 longitudinal effect sizes).

Table 2.

Moderator Analyses for the Relation Between Bullying Perpetration and Dating Violence Perpetration.

	k	MVA OR	95% CI	Fixed OR	95% CI	Random OR	95% CI
	k	MVA OR	[Lower, Upper]	Fixed OR	[Lower, Upper]	Random OR	[Lower, Upper]
Unadjusted	12	1.98	[1.48, 2.65]	1.98	[1.85, 2.11]	2.12	[1.54, 2.11]
Adjusted	8	1.29	[1.07, 1.56]	1.29	[1.18, 1.40]	1.49	[1.19, 1.87]
Gender
Females	5	1.58	[1.34, 3.50]	1.58	[1.33, 1.89]	1.58	[1.33, 1.89]
Males	6	1.39	[1.05, 1.84]	1.36	[1.22, 1.52]	1.39	[1.14, 1.71]
Design
Cross sectional	10	2.11	[1.41, 2.78]	2.11	[1.97, 2.25]	2.23	[1.60, 3.11]
Longitudinal	3	1.20	[0.98, 1.47]	1.20	[1.10, 1.30]	1.35	[1.03, 1.76]
Type of dating violence
Electronic	2	3.08	[1.37, 6.91]	3.58	[2.57, 3.69]	2.53	[1.02, 6.25]
Emotional	5	2.16	[1.50, 3.11]	2.16	[1.98, 2.35]	1.98	[1.35, 2.90]
Physical	8	1.42	[1.11, 1.82]	1.42	[1.34, 1.52]	1.64	[1.23, 2.17]
Relational	2	1.10	[0.66, 1.83]	1.10	[0.93, 1.30]	1.59	[0.62, 4.12]
Sexual	3	1.78	[1.11, 2.86]	1.78	[1.56, 2.03]	2.44	[1.41, 4.20]
Stalking	1	0.92	[0.77, 1.09]	0.92	[0.77, 1.09]	0.92	[0.77, 1.09]
Country
USA	8	1.70	[1.18, 2.46]	1.70	[1.60, 1.80]	1.87	[1.25, 2.79]
Canada	4	2.24	[1.66, 3.03]	2.24	[1.87, 2.68]	2.56	[1.60, 4.11]
Other	1	1.40	[1.22, 1.60]	1.40	[1.22, 1.60]	1.40	[1.22, 1.60]

Note. CI = confidence interval; OR = odds ratio; MVA = multiplicative variance adjustment.

Bullying Perpetration Versus Dating Violence Victimization

Seven studies provided primary unadjusted results and three studies provided adjusted results about the relation between bullying perpetration and dating violence victimization. Bullying perpetration was related to dating violence victimization (MVA OR = 2.59, 95% CI [2.14, 3.13], z = 9.81), and this relation held after controlling for confounding variables (MVA OR = 1.09, 95% CI [1.01, 1.12], z = 2.27). Nevertheless, the latter OR should be interpreted with caution because only three studies were used for its calculation and the relation is nonsignificant using the random effects calculation (see Figure 3). The fail-safe N test showed that 1,030 missing studies with nonsignificant effect sizes would need to be added to make the unadjusted effect nonsignificant. The adjusted effect size would be nonsignificant if four primary studies with nonsignificant effect were added.

Figure 3.

Unadjusted and adjusted effect sizes for the relation between bullying perpetration and dating violence victimization.

Unadjusted effects were significantly heterogeneous (Q = 26.28, df = 6, p < .01, I ² = 77.17). The adjusted effects were homogeneous (Q = 2.78, df = 2, p = .25, I ² =28.05), although this needs to be interpreted with caution because only three studies were included. Moderator analyses (see Table 3) revealed that the relation between bullying perpetration and dating victimization was stronger for males, longitudinal studies, emotional dating violence, and the studies conducted in the United States. Again, numbers of primary effects varied among the moderators and results should be interpreted with caution.

Table 3.

Moderator Analyses for the Relation Between Bullying Perpetration and Dating Violence Victimization.

	k		95% CI	Fixed OR	95% CI	Random OR	95% CI
	k	MVA OR	[Lower, Upper]	Fixed OR	[Lower, Upper]	Random OR	[Lower, Upper]
Unadjusted	7	2.59	[2.14, 3.13]	2.59	[2.36, 2.83]	2.28	[1.78, 2.92]
Adjusted	3	1.09	[1.01, 1.17]	1.09	[1.02, 1.16]	1.14	[0.98, 1.33]
Gender
Females	2	1.13	[1.12, 1.14]	1.13	[1.06, 1.20]	1.13	[1.06, 1.20]
Males	1	2.09	[1.28, 3.41]	2.09	[1.28, 3.41]	2.09	[1.28, 3.41]
Design
Cross sectional	6	1.43	[1.00, 1.14]	1.43	[1.36, 1.51]	2.08	[1.26, 3.43]
Longitudinal	1	1.51	[1.07, 2.13]	1.51	[1.07, 2.13]	1.51	[1.07, 2.13]
Type of dating violence
Electronic	1	1.60	[1.13, 2.26]	1.60	[1.13, 2.26]	1.60	[1.13, 2.26]
Emotional	3	1.83	[1.70, 1.97]	1.83	[1.56, 2.14]	1.83	[1.56, 2.14]
Physical	3	1.60	[1.14, 2.24]	1.60	[1.37, 1.88]	1.67	[1.14, 2.44]
Country
USA	3	2.78	[2.07, 3.73]	2.78	[2.49, 3.09]	2.25	[1.44, 3.50]
Canada	3	1.75	[1.22, 2.50]	1.75	[1.46, 2.10]	2.12	[1.21, 3.70]
Other	1	1.13	[1.06, 1.20]	1.13	[1.06, 1.20]	1.13	[1.06, 1.20]

Note. CI = confidence interval; OR = odds ratio; MVA = multiplicative variance adjustment.

Bullying Victimization Versus Dating Violence Victimization

There was a strong unadjusted relation between high bullying victimization and high dating victimization calculated from 12 primary studies (MVA OR = 2.51, 95% CI [2.17, 2.90], z = 12.47). This relation held after controlling for confounding variables in five primary studies (MVA OR = 1.96, 95% CI [1.68, 2.29], z = 8.45); see Figure 4 for more details. According to the fail-safe N test, 1,900 studies with nonsignificant results would be needed to make the unadjusted effect nonsignificant and 385 to make the adjusted effect nonsignificant.

Figure 4.

Unadjusted and adjusted effect sizes for the relation between bullying victimization and dating violence victimization.

The effect sizes of the included studies were significantly heterogeneous (unadjusted Q = 69.11, df = 11, p < .01, I ² = 84.08; adjusted Q = 28.13, df = 4, p < .01, I ² = 85.78). Moderator analyses showed that the relation between bullying victimization and dating victimization was stronger for females, cross-sectional studies, sexual dating violence, and the U.S. studies (see Table 4).

Table 4.

Moderator Analyses for the Relation Between Bullying Victimization and Dating Violence Victimization.

	k		95% CI	Fixed OR	95% CI	Random OR	95% CI
	k	MVA OR	[Lower, Upper]	Fixed OR	[Lower, Upper]	Random OR	[Lower, Upper]
Unadjusted	12	2.51	[2.17, 2.90]	2.51	[2.37, 2.66]	2.18	[1.85, 2.58]
Adjusted	5	1.96	[1.68, 2.29]	1.96	[1.84, 2.07]	1.80	[1.45, 2.24]
Gender
Females	4	2.44	[1.99, 3.00]	2.44	[2.25, 2.64]	2.22	[1.69, 2.92]
Males	4	1.86	[1.55, 2.24]	1.86	[1.70, 2.03]	1.94	[1.49, 2.54]
Design
Cross sectional	13	2.23	[1.95, 2.55]	2.23	[2.14, 2.33]	2.10	[1.81, 2.45]
Longitudinal	1	1.55	[1.10, 2.18]	1.55	[1.10, 2.18]	1.55	[1.10, 2.18]
Type of dating violence
Electronic	3	2.35	[1.82, 3.04]	2.35	[2.07, 2.68]	2.01	[1.55, 2.85]
Emotional	7	1.94	[1.63, 2.34]	1.94	[1.80, 2.10]	1.85	[1.50, 2.27]
Physical	8	1.79	[1.58, 2.03]	1.79	[1.65, 1.95]	1.76	[1.54, 2.03]
Sexual	2	2.44	[2.40, 2.48]	2.44	[2.04, 2.92]	2.44	[2.04, 2.92]
Country
USA	11	2.29	[2.03, 2.58]	2.29	[2.20, 2.40]	2.21	[1.93, 2.54]
Canada	3	1.46	[0.88, 2.43]	1.46	[1.24, 1.70]	1.70	[0.95, 3.03]
Other	1	2.27	[1.57, 3.27]	2.27	[1.57, 3.27]	2.27	[1.57, 3.27]

Note. CI = confidence interval; OR = odds ratio; MVA = multiplicative variance adjustment.

Bullying Victimization Versus Dating Violence Perpetration

The relation between bullying victimization and dating violence perpetration (see Figure 5) was nonsignificant (unadjusted MVA OR = 1.43, 95% CI [0.92, 2.22], z = 1.59; adjusted MVA OR = 1.01, 95% CI [0.99, 1.03], z = 0.61). These results should be interpreted with caution because only five studies reported unadjusted effects and three studies adjusted effects. Fail-safe N tests were not performed because the overall effect sizes were not significant. Unadjusted studies were significantly heterogeneous (Q = 144.37, df = 4, p < .01, I ² = 97.23), whereas adjusted studies were not (Q = 0.07, df = 2, p = .97, I ² = 0). Moderator analyses (Table 5) showed nonsignificant results for all the analyzed relations except for the relation between high bullying victimization and high electronic dating violence perpetration.

Figure 5.

Unadjusted and adjusted effect sizes for the relation between bullying victimization and dating violence perpetration.

Table 5.

Moderator Analyses for the Relation Between Bullying Victimization and Dating Violence Perpetration.

	k	MVA OR	95% CI	Fixed OR	95% CI	Random OR	95% CI
	k	MVA OR	[Lower, Upper]	Fixed OR	[Lower, Upper]	Random OR	[Lower, Upper]
Unadjusted	5	1.43	[0.92, 2.22]	1.43	[1.33, 1.54]	1.29	[0.80, 2.08]
Adjusted	3	1.01	0.99, 1.03	1.01	[0.89, 1.14]	1.01	[0.89, 1.14]
Gender
Females	2	0.77	0.51, 1.15	0.77	[0.62, 0.96]	0.86	[0.53, 1.40]
Males	2	0.94	0.50, 1.78	0.94	[0.75, 1.17]	1.15	[0.52, 2.52]
Design
Cross sectional	3	1.56	0.81, 2.99	1.56	[1.44, 1.70]	1.28	[0.62, 2.61]
Longitudinal	2	1.05	0.88, 1.25	1.05	[0.95, 1.15]	1.16	[0.84, 1.59]
Type of dating violence
Electronic	1	1.46	1.03, 2.05	1.46	[1.03, 2.05]	1.46	[1.03, 2.05]
Emotional	3	0.99	0.70, 1.40	0.99	[0.88, 1.12]	1.08	[0.74, 1.58]
Physical	3	0.96	0.76, 1.21	0.96	[0.88, 1.04]	0.98	[0.73, 1.31]
Relational	1	0.79	0.66, 0.94	0.79	[0.66, 0.94]	0.79	[0.66, 0.94]
Sexual	1	0.67	0.56, 0.80	0.67	[0.56, 0.80]	0.67	[0.56, 0.80]
Stalking	1	0.76	0.63, 0.90	0.76	[0.63, 0.90]	0.76	[0.63, 0.90]
Country
USA	4	1.32	0.80, 2.19	1.32	[1.23, 1.41]	1.25	[0.73, 2.16]
Canada	1	1.21	1, 1.47	1.21	[1, 1.47]	1.21	[1, 1.47]

Note. CI = confidence interval; OR = odds ratio; MVA = multiplicative variance adjustment.

Discussion

Bullying and dating violence are two antisocial behaviors present in young people. Although many anti-bullying programs are being conducted around the world (Gaffney et al., 2018) together with many interventions against dating violence (De La Rue et al., 2017; Jennings et al., 2017), these two types of aggressive behaviors are rarely studied together. Given that the behaviors and attitudes of the peer groups are crucial in bullying (Salmivalli, 2010) and in dating violence (Leen et al., 2013) and that many antisocial behaviors share risk and protective factors (Catalano et al., 2012), understanding the relation between bullying and dating violence could be a landmark in the developmental prevention of antisocial behavior. New prevention programs could become more comprehensive and more efficient, tackling these problem behaviors through similar program components. Thus, the current research synthesis was conducted to describe the relation of bullying perpetration and victimization with dating violence perpetration and victimization.

After systematic searches and the application of the inclusion and exclusion criteria, this meta-analysis included longitudinal and cross-sectional projects. Overall effect sizes were calculated for direct relations and also after adjusting for confounders and moderators. It was found that the number of studies on the relation between bullying and dating violence is still low and that most of the projects are cross sectional. Given the importance of the topic, more research is needed to draw firm conclusions regarding the relation between these two antisocial behaviors. The cross-sectional nature of these studies does not make it possible to establish causal relations, and tentative conclusions are only drawn on a theoretical basis.

The current meta-analysis discovered that bullies are at high risk of being dating violence perpetrators and victims of bullying are at high risk of being dating violence victims. Specifically, there was a direct relation between bullying perpetration and dating violence perpetration (OR = 1.98). If a 20% prevalence of dating violence is assumed in nonbullies, this effect size corresponds roughly to 33% of bullies being dating violence perpetrators. This effect holds after adjusting for confounding variables (OR = 1.29), which corresponds roughly to 25% of bullies (assuming 20% of nonbullies) becoming dating violence perpetrators. There was also a direct relation between bullying victimization and dating violence victimization (OR = 2.51) that corresponds roughly to 40% of victims of bullying (assuming 20% of nonvictims) being victims of dating violence. After controlling for confounders (OR = 1.96), the effect size corresponds roughly to 33% of victims of bullying becoming victims of dating violence (assuming 20% of nonvictims).

As suggested by some authors based on the social learning framework, it is possible that behaviors are learned in one social context, such as peer relationships, and continue in a different context, such as romantic relationships (Malik, Sorenson, & Aneshensel, 1997). Some children might learn that bullying others might improve the social status of the bullies (Salmivalli, 2010) and they might apply this learning about aggressive behaviors that are perceived as desirable also to romantic relationships. The current study opens up new research horizons where this possible causal relations should be confirmed and early risk factors for both problem behaviors discovered.

Moderator analyses showed that the relation between bullying perpetration and dating violence perpetration and the relation between bullying victimization and dating violence victimization were stronger for females, in cross-sectional designs, and for electronic and emotional dating violence. It is possible that females are more vulnerable than males and that the co-occurrence measured in cross-sectional studies is more frequent than the long-term persistence in perpetration or victimization across contexts. Nevertheless, gender differences need to be explored further in future research taking into account dating in sexual majority and also minority groups. It is also possible that certain types of perpetration and victimization co-occur more frequently across contexts than other types of perpetration and victimization. The small number of studies makes the moderator analyses preliminary and these relations need to be further explored and explained in future projects.

The results of this meta-analysis showed that bullying perpetration was related to dating violence victimization (OR = 2.59), corresponding roughly to 40% of bullies becoming dating violence victims (assuming 20% of nonvictims). This relation was much weaker, but still just significant, after controlling for covariates. Previous research found that bullies are at high risk of becoming offenders later in life (Ttofi et al., 2011). A literature review by Jennings, Piquero, and Reingle (2012) showed that, in general, there is an overlap between victimization and offending. Thus, it is possible that bullies become a part of a violent social context where both victimization and perpetration are common. Most of the reviewed studies focused on perpetration or victimization separately and bully/victims or perpetrators/victims in dating violence were rarely considered. It is possible that many bullies were simultaneously perpetrators and victims of dating violence. Future research should focus on the overlap between victimization and perpetration.

Although the findings of this systematic review and meta-analysis advance knowledge about the relation between bullying and dating violence, it also has some limitations. In general, the number of studies was low and almost all of them were cross sectional. Thus, the causal order of relations between bullying and dating violence cannot be firmly established. Based on theory, we concluded that it is possible that adolescents who are involved in bullying later become involved in dating violence. Nevertheless, it is also possible that adolescents become involved in both problem behaviors at the same time or that they get involved in dating violence first and then become involved in bullying. These relations should be clarified in future.

Implications for research, policy, and practice

Notwithstanding its limitations, this study has important implications for research, policy, and practice. It seems reasonable to suggest that bullying should be studied within the context of antisocial behaviors in general. Most studies on bullying do not focus on other problem behaviors, which might not make it possible to see a person from a comprehensive perspective. It could be desirable to design and evaluate interventions that target bullying, dating violence, and possibly other antisocial behaviors at the same time. Both bullying and dating violence should be tackled in schools as two related behaviors. If anti-bullying programs are effective and the relation between bullying and teen dating violence is not coincidental, it could be possible to tackle both problem behaviors in comprehensive interventions. Taking into account the results of the moderator analysis, specific components related to gender, age, and type of dating violence could be included. Also, new studies could explore relations among bullying, dating violence, and other antisocial behaviors. Thus, the current study opens up new horizons regarding future research and practice focused on bullying and dating violence.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Universidad de Córdoba (Mobility Grant).

ORCID iD

Izabela Zych

References

Analitis

Velderman

M. K.

Ravens-Sieberer

Detmar

Erhart

Herdman

… Group

E. K.

(2009). Being bullied: Associated factors in children and adolescents 8 to 18 years old in 11 European countries. Pediatrics, 123, 569–577. doi:10.1542/peds.2008-0323

Antônio

Koller

S. H.

Hokoda

(2012). Peer influences on the dating aggression process among Brazilian street youth: A brief report. Journal of Interpersonal Violence, 27, 1579–1592. doi:10.1177/0886260511425794

Archer

Ireland

I. J.

Monks

Gilling

(2000). Sex differences in aggression between heterosexual partners: A meta-analytic review. Psychological Bulletin, 126, 651–680. doi:10.1037/0033-2909.126.5.651

Catalano

R. F.

Fagan

A. A.

Gavin

L. E.

Greenberg

M. T.

Irwin

C. E.

Jr Ross

D. A.

Shek

D. T.

(2012). Worldwide application of prevention science in adolescent health. The Lancet, 379, 1653–1664. doi:10.1016/S0140-6736(12)60238-4

Cava

M. J.

Buelga

Carrascosa

(2015). Violencia física y psicológica ejercida en parejas adolescentes: relación con el autoconcepto y la violencia entre iguales [Physical and psychological violence in adolescent couples: relation to self-concept and violence among peers]. Psicología Conductual, 23, 429–446.

*Coker

A. L.

Clear

E. R.

Garcia

L. S.

Asaolu

I. O.

Cook-Craig

P. G.

Brancato

C. J.

… Fisher

B. S

. (2014). Dating violence victimization and perpetration rates among high school students. Violence Against Women, 20, 1220–1238. doi:10.1177/1077801214551289

*Connolly

Josephson

Schnoll

Simkins-Strong

Pepler

MacPherson

… Jiang

. (2015). Evaluation of a youth-led program for preventing bullying, sexual harassment, and dating aggression in middle schools. The Journal of Early Adolescence, 35, 403–434. doi:10.1177/0272431614535090

*Connolly

Pepler

Craig

Taradash

(2000). Dating experiences of bullies in early adolescence. Child Maltreatment, 5, 299–310.

Craig

Harel-Fisch

Fogel-Grinvald

Dostaler

Hetland

Simons-Morton

… Pickett

(2009). A cross-national profile of bullying and victimization among adolescents in 40 countries. International Journal of Public Health, 54, 216–224. doi:10.1007/s00038-009-5413-9

10.

*Cutbush

Williams

(2016). Teen dating violence, sexual harassment, and bullying among middle school youth: Examining measurement invariance by gender. Journal of Research on Adolescence, 26, 918–926. doi:10.1111/jora.12244

11.

*Cutbush

Williams

Miller

(2016). Teen dating violence, sexual harassment, and bullying among middle school students: Examining mediation and moderated mediation by gender. Prevention Science, 17, 1024–1033. doi:10.1007/s11121-016-0668-x

12.

*Debnam

K. J.

Waasdorp

T. E.

Bradshaw

C. P.

(2016). Examining the contemporaneous occurrence of bullying and teen dating violence victimization. School Psychology Quarterly, 31, 76–90. doi:10.1037/spq0000124

13.

De La Rue

Polanin

J. R.

Espelage

D. L.

Pigott

T. D.

(2017). A meta-analysis of school-based interventions aimed to prevent or reduce violence in teen dating relationships. Review of Educational Research, 87, 7–34. doi:10.3102/0034654316632061

14.

Dunphy

D. C.

(1963). The social structure of urban adolescent peer groups. Sociometry, 26, 230–246. doi:10.2307/2785909

15.

Eisner

M. P.

Malti

(2015). Aggressive and violent behavior. In Lamb

M. E.

(Vol. Ed.) Lerner

R. M.

(Series Ed.), Handbook of child psychology and developmental science, : Social, emotional and personality development (Vol. 3, 7th ed., pp. 795–884). New York, NY: Wiley.

16.

*Ellis

W. E.

Wolfe

D. A.

(2015). Bullying predicts reported dating violence and observed qualities in adolescent dating relationships. Journal of Interpersonal Violence, 30, 3043–3064. doi:10.1177/0886260514554428

17.

*Espelage

D. L.

Holt

M. K.

(2007). Dating violence & sexual harassment across the bully-victim continuum among middle and high school students. Journal of Youth and Adolescence, 36, 799–811. doi:10.1007/s10964-006-9109-7

18.

*Espelage

D. L.

Low

S. K.

Anderson

De La Rue

(2014). Bullying, sexual, and dating violence trajectories from early to late adolescence. Washington, DC: The U.S. Department of Justice.

19.

Farrington

D. P.

Welsh

B. C.

(2013). Measuring effect size in meta-analysis, with special reference to area-based crime prevention programmes and the effects of closed-circuit television on crime. In Kuhn

Schwarzenegger

Margot

Donatsch

Aebi

Jositsch

(Eds.), Criminology, criminal policy and criminal law from an international perspective (pp. 75–89). Berne, Switzerland: Stampfli.

20.

*Foshee

V. A.

Benefield

T. S.

McNaughton Reyes

H. L.

Eastman

Vivolo-Kantor

A. M.

Basile

K. C.

… Faris

(2016). Examining explanations for the link between bullying perpetration and physical dating violence perpetration: Do they vary by bullying victimization? Aggressive Behavior, 42, 66–81. doi:10.1002/ab.21606

21.

*Foshee

V. A.

McNaughton

H. L.

Chen

M. S.

Ennett

S. T.

Basile

K. C.

DeGue

… Bowling

J. M.

(2016). Shared risk factors for the perpetration of physical dating violence, bullying, and sexual harassment among adolescents exposed to domestic violence. Journal of Youth and Adolescence, 45, 672–686. doi:10.1007/s10964-015-0404-z

22.

*Foshee

V. A.

McNaughton

H. L.

Vivolo-Kantor

A. M.

Basile

K. C.

Chang

L. Y.

Faris

Ennett

S. T.

(2014). Bullying as a longitudinal predictor of adolescent dating violence. Journal of Adolescent Health, 55, 439–444. doi:10.1016/j.jadohealth.2014.03.004

23.

Gaffney

Ttofi

M. M.

Farrington

D. P.

(2019). Evaluating the effectiveness of school-bullying prevention programs: An updated meta-analytical review. Aggression and Violent Behavior, 45, 111–133. doi:10.1016/j.avb.2018.07.001

24.

*Hellevik

Øverlien

(2016). Teenage intimate partner violence: Factors associated with victimization among Norwegian youths. Scandinavian Journal of Public Health, 44, 702–708. doi:10.1177/1403494816657264

25.

*Holt

M. K.

Espelage

D. L.

(2003). A cluster analytic investigation of victimization among high school students. Journal of Applied School Psychology, 19, 81–98. doi:10.1300/J008v19n02_06

26.

Jennings

W. G.

Piquero

A. R.

Reingle

J. M.

(2012). On the overlap between victimization and offending: A review of the literature. Aggression and Violent Behavior, 17, 16–26. doi: 10.1016/j.avb.2011.09.003

27.

Jennings

W. G.

Okeem

Piquero

A. R.

Sellers

C. S.

Theobald

Farrington

D. P.

(2017). Dating and intimate partner violence among young persons ages 15–30: Evidence from a systematic review. Aggression and Violent Behavior, 33, 107–125. doi:10.1016/j.avb.2017.01.007

28.

*Leadbeater

B. J.

Banister

E. M.

Ellis

W. E.

Yeung

(2008). Victimization and relational aggression in adolescent romantic relationships: The influence of parental and peer behaviors, and individual adjustment. Journal of Youth and Adolescence, 37, 359–372. doi:10.1007/s10964-007-9269-0

29.

Leen

Sorbring

Mawer

Holdsworth

Helsing

Bowen

(2013). Prevalence, dynamic risk factors and the efficacy of primary interventions for adolescent dating violence: An international review. Aggression and Violent Behavior, 18, 159–174. doi:10.1016/j.avb.2012.11.015

30.

Lipsey

M. W.

Wilson

D. B.

(2000). Practical meta-analysis. London, England: Sage.

31.

*Lozano

M. G.

(2012). Teen dating violence: Relationship to bullying, sexual behaviors and school safety (Doctoral dissertation). California State University, Long Beach. Retrieved from Proquest.

32.

Malik

Sorenson

S. B.

Aneshensel

C. S.

(1997). Community and dating violence among adlescents: Perpetration and victimization. Journal of Adolescent Health, 21, 291–302. doi:10.1016/S1054-139X(97)00143-2

33.

*Martin-Storey

(2015). Prevalence of dating violence among sexual minority youth: Variation across gender, sexual minority identity and gender of sexual partners. Journal of Youth and Adolescence, 44, 211–224. doi:10.1007/s10964-013-0089-0

34.

*Melancon

Gagne

M. H.

(2011). Father’s and mother’s psychological violence and adolescent behavioral adjustment. Journal of Interpersonal Violence, 26, 991–1011. doi:10.1177/0886260510365863

35.

Menesini

Nocentini

Ortega-Rivera

F. J.

Sanchez

Ortega

(2011). Reciprocal involvement in adolescent dating aggression: An Italian–Spanish study. European Journal of Developmental Psychology, 8, 437–451. doi:10.1080/17405629.2010.549011

36.

*Miller

Williams

Cutbush

Gibbs

Clinton-Sherrod

Jones

(2013). Dating violence, bullying, and sexual harassment: Longitudinal profiles and transitions over time. Journal of Youth and Adolescence, 42, 607–618. doi:10.1007/s10964-013-9914-8

37.

Modecki

K. L.

Minchin

Harbaugh

A. G.

Guerra

N. G.

Runions

K. C.

(2014). Bullying prevalence across contexts: A meta-analysis measuring cyber and traditional bullying. Journal of Adolescent Health, 55, 602–611. doi:10.1016/j.jadohealth.2014.06.007

38.

*Niolon

P. H.

Vivolo-Kantor

A. M.

Latzman

N. E.

Valle

L. A.

Kuoh

Burton

… Tharp

A. T.

(2015). Prevalence of teen dating violence and co-occurring risk factors among middle school youth in high-risk urban communities. Journal of Adolescent Health, 56, S5–S13. doi:10.1016/j.jadohealth.2014.07.019

39.

*Pepler

D. J.

Craig

W. M.

Connolly

J. A.

Yuile

McMaster

Jiang

(2006). A developmental perspective on bullying. Aggressive Behavior, 32, 376–384. doi:10.1002/ab.20136

40.

*Peters

Z. J.

Hatzenbuehler

M. L.

Davidson

L. L.

(2017). Examining the intersection of bullying and physical relationship violence among New York City high school students. Journal of Interpersonal Violence, 32, 49–75. doi:10.1177/0886260515585532

41.

Olweus

(1999). Sweden. In Smith

P. K.

Morita

Junger-Tas

Olweus

Catalano

Slee

(Eds.), The nature of school bullying: A cross-national perspective (pp.7–27). London, England: Routledge.

42.

Rodríguez Franco

Antuña Bellerín

M. A.

López-Cepero Borrego

Rodríguez Díaz

F. J.

Bringas Molleda

(2012). Tolerance towards dating violence in Spanish adolescents. Psicothema, 24, 236–242.

43.

Sánchez

Ortega-Rivera

Ortega-Ruiz

Viejo

(2008). Las relaciones sentimentales en la adolescencia: satisfacción, conflictos y violencia [Romantic relationships in adolescence: Satisfaction, conflicts and dating violence]. Escritos de Psicología, 2, 97–109.

44.

Salmivalli

(2010). Bullying and the peer group: A review. Aggression and Violent Behavior, 15, 112–120. doi:10.1016/j.avb.2009.08.007

45.

Salmivalli

Lagerspetz

Björkqvist

Österman

Kaukiainen

(1996). Bullying as a group process: Participant roles and their relations to social status within the group. Aggressive Behavior, 22, 1–15. doi:10.1002/(SICI)1098-2337(1996)22:1<1::AID-AB1>3.0.CO;2-T

46.

*Shamu

Gevers

Mahlangu

B. P.

Jama Shai

P. N.

Chirwa

E. D.

Jewkes

R. K.

(2016). Prevalence and risk factors for intimate partner violence among Grade 8 learners in urban South Africa: Baseline analysis from the Skhokho Supporting Success cluster randomised controlled trial. International Health, 8, 18–26. doi:10.1093/inthealth/ihv068

47.

Shore

R. E.

Gardner

M. J.

Pannett

(1993). Ethylene oxide: An assessment of the epidemiological evidence on carcinogenicity. Occupational and Environmental Medicine, 50, 971–997. doi:10.1136/oem.50.11.971

48.

Smith

P. K.

Cowie

Olafsson

R. F.

Liefooghe

A. P.

(2002). Definitions of bullying: A comparison of terms used, and age and gender differences, in a Fourteen–Country international comparison. Child Development, 73, 1119–1133. doi:10.1111/1467-8624.00461

49.

Smith

P. K.

López-Castro

Robinson

Görzig

. (2018). Consistency of gender differences in bullying in cross-cultural surveys. Aggression and Violent Behavior, 45, 33–40. doi:10.1016/j.avb.2018.04.006

50.

Ttofi

M. M.

Farrington

D. P.

Lösel

Loeber

(2011). The predictive efficiency of school bullying versus later offending: A systematic/meta-analytic review of longitudinal studies. Criminal Behaviour and Mental Health, 21, 80–89. doi: 10.1002/cbm.808

51.

Ttofi

M. M.

Farrington

D. P.

Lösel

(2012). School bullying as a predictor of violence later in life: A systematic review and meta-analysis of prospective longitudinal studies. Aggression and Violent Behavior, 17, 405–418. doi:10.1016/j.avb.2012.05.002

52.

Viejo

Ortega-Ruiz

Sánchez

(2015). Adolescent love and well-being: The role of dating relationships for psychological adjustment. Journal of Youth Studies, 18, 1219–1236. doi:10.1080/13676261.2015.1039967

53.

*Vivolo-Kantor

A. M.

Olsen

E. O. M.

Bacon

(2016). Associations of teen dating violence victimization with school violence and bullying among US high school students. Journal of School Health, 86, 620–627.

54.

Wincentak

Connolly

Card

(2016). Teen dating violence: A meta-analytic review of prevalence rates. Psychology of Violence, 7, 224–241. doi:10.1037/a0040194

55.

Wilson

D. B.

Gottfredson

D. C.

Najaka

S. S.

(2001). School-based prevention of problem behaviors: A meta-analysis. Journal of Quantitative Criminology, 17, 247–272.

56.

Wolitzky-Taylor

K. B.

Ruggiero

K. J.

Danielson

C. K.

Resnick

H. S.

Hanson

R. F.

Smith

D. W.

… Kilpatrick

D. G.

(2008). Prevalence and correlates of dating violence in a national sample of adolescents. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 755–762. doi:10.1097/CHI.0b013e318172ef5f

57.

Wolfe

D. A.

Crooks

Jaffe

Chiodo

Hughes

Ellis

… Donner

(2009). A school-based program to prevent adolescent dating violence: A cluster randomized trial. Archives of Pediatrics & Adolescent Medicine, 163, 692–699. doi:10.1001/archpediatrics.2009.69

58.

Wolke

Copeland

W. E.

Angold

Costello

E. J.

(2013). Impact of bullying in childhood on adult health, wealth, crime, and social outcomes. Psychological Science, 24, 1958–1970. doi:10.1177/0956797613481608

59.

*Yahner

Dank

Zweig

J. M.

Lachman

(2015). The co-occurrence of physical and cyber dating violence and bullying among teens. Journal of Interpersonal Violence, 30, 1079–1089. doi:10.1177/0886260514540324

60.

Zych

Baldry

A. C.

Farrington

D. P.

Llorent

V. J.

(2018). Are children involved in cyberbullying low on empathy? A systematic review and meta-analysis of research on empathy versus different cyberbullying roles. Aggression and Violent Behavior, 45, 83–97. doi:10.1016/j.avb.2018.03.004

61.

Zych

Ortega-Ruiz

Del Rey

(2015a). Scientific research on bullying and cyberbullying: Where have we been and where are we going. Aggression and Violent Behavior, 24, 188–198. doi:10.1016/j.avb.2015.05.015

62.

Zych

Ortega-Ruiz

Del Rey

(2015b). Systematic review of theoretical studies on bullying and cyberbullying: Facts, knowledge, prevention and intervention. Aggression and Violent Behavior, 23, 1–21. doi:10.1016/j.avb.2015.10.001

63.

Zych

Ttofi

M. M.

Llorent

V. J.

Farrington

D. P.

Ribeaud

Eisner

M. P

. (2018). A longitudinal study on stability and transitions among bullying roles. Child Development, 91, 527–545. doi:10.1111/cdev.13195