Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Jul 1.
Published in final edited form as: J Abnorm Child Psychol. 2013 Jul;41(5):801–809. doi: 10.1007/s10802-013-9720-5

Bullies, Victims, and Antipathy: The Feeling is Mutual

Christopher A Hafen 1, Brett Laursen 2, Jari-Eri Nurmi 3, Katariina Salmela-Aro 4
PMCID: PMC3665741  NIHMSID: NIHMS438858  PMID: 23344703

Abstract

In this study we consider whether bullies and victims are disliked by most of their classmates, or whether antipathy is concentrated among the occupants of these roles. Antipathy nominations were collected from a community sample of 699 Finnish adolescents (14 to 17 years of age), who described their own bullying and victimization, as well as problem behaviors and school engagement. Victimization was associated with antipathy, but the strength of the association differed according to characteristics of the nominator. Victimization was related to antipathy when the nominator was high on bullying but not low. Similarly, bullying was related to antipathy when the nominator was high on victimization, but not low. The findings indicate that although bullies and victims have elevated mean levels of rejection, they are not disliked by most peers but rather by those who report themselves to be high on these attributes.

Keywords: Bullying, Victimization, Antipathy, Rejection

Conventional wisdom holds that bullies and victims are not well liked by classmates. The reasons are simple: Bullies score high on measures of relational or physical aggression and victims are often socially withdrawn, disruptive, or anxious (Salmivalli, 2010). Bullies and victims are thought to be widely disliked, because both receive a large number of antipathy or "liked least" nominations (Rodkin & Berger, 2008). But the claim that bullies and victims are disliked by most classmates assumes that the antipathy nominations directed toward them are distributed evenly across the peer population. In the present study we examine an alternative hypothesis, namely that antipathy for bullies is concentrated primarily among those who are victims and that antipathy for victims is concentrated primarily among those who bully. We test this hypothesis in a community sample of Finnish adolescents, using multilevel modeling to determine (a) whether the association between disliking a target and the target's self-reported victimization score varies as a function of the nominator's self-reported bullying score, and (b) whether the association between disliking a target and the target's self-reported bullying score varies as a function of the nominator's self-reported victimization score.

Bullying is a distinct form of aggressive behavior in which an individual repeatedly attacks, humiliates, or excludes another individual who has difficulty defending him- or herself (Olweus, 2002). A victim is the target of this bullying. Both bullying and victimization peak during the adolescent years, when up to 30% of youth report being either the agent or the victim of bullying behavior (Salmivalli & Peets, 2009). Adjustment problems accompany those who participate in bully-victim relationships, including school difficulties, depression, and substance use (Haynie et al., 2001). Bullying and victimization are usually assessed through either peer nominations (e.g., Espelage, Holt, & Henkel, 2003) or via self-reports (e.g., Olweus, 1989).

Antipathy refers to disliking or active avoidance of another person (Abecassis, 2003). Evidence suggests that antipathies are normative, with approximately 2/3 of adolescents disliking at least one agemate (Prinstein & Aikins, 2004). Mutual antipathies are less common. A recent meta-analysis reported that 35% of middle school and high school students are involved in at least one mutual antipathy (Card, 2010). Youth who participate in mutual antipathies tend to be less well adjusted than those who do not, reporting more internalizing problems and lower self-worth (Card & Hodges, 2007). The same cannot be said for those involved in unilateral antipathies (also known as unreciprocated antipathies); the adjustment outcomes of these youth tend to resemble those who claim to dislike no one (Erath, Petit, Dodge, & Bates, 2008). Antipathy is typically measured with a nomination procedure in which youth are asked to name someone they dislike (Erath et al., 2008), someone with whom they least like to play or work (Card & Hodges, 2003), or someone with whom they least like to spend time (Laursen et al., 2010), which was the method used in the present study.

Bullies and victims are assumed to be targets of antipathy for several reasons. Bullies tend to be high on perceived popularity, but this means that they are influential not that they are well liked (Sijtsema, Veenstra, Lindenberg, & Salmivalli, 2009). This is an important point as several studies support the assertion that some bullies can be popular and that bullies are not universally disliked, especially in adolescence (de Bruyn, Cillessen, & Wissink, 2010; Rodkin, Farmer, Pearl, & Van Acker, 2006) Both bullies and victims tend to score high on relational and physical aggression (Salmivalli & Peets, 2009). Bullies may be disliked because they exhibit callous-unemotional traits (Viding, Simmons, Petrides, & Frederickson, 2009). Victims may be disliked because they have poor social skills and difficulty interacting with peers (Guerra, Williams, & Sadek, 2011). Several previous studies have reported that bullies and victims are widely disliked by their classmates (e.g., Rodkin & Berger, 2008), but these studies are based on mean levels of antipathy nominations. Because frequency counts describe only the number of age-mates who dislike a target, not the identity of those who dislike the target, previous studies cannot rule out an alternative explanation, namely that bullies and victims tend to vigorously dislike one another (driving up the number of negative nominations each receives) but their antipathies are not widely shared by the rest of the peer population.

In the present study, we used data from a community sample of Finnish adolescents to learn whether antipathies directed toward bullies and victims are restricted to these youth or are reported by most students. Results from a multilevel modeling technique known as hierarchical general linear logistic modeling (HGLLM; Ng, Carpenter, Goldstein, Rasbash, 2006; Raudenbush & Bryk, 2002) determined whether characteristics of the nominator moderated the likelihood that a bully or victim was disliked. This approach has the advantage of including everyone who makes (nominator) and receives (nominee) antipathy nominations in the same analysis. At level 1, nominee bullying and nominee victimization were the predictors of antipathy. At level 2, nominator bullying and nominator victimization were the predictors of these level 1 associations. Two alternative hypotheses were tested. The first hypothesis holds that bullies and victims are rejected because most peers dislike them (nominator characteristics do not predict the degree to which antipathy is associated with bullying and victimization). The second hypothesis holds that the elevated rejection scores of bullies and victims are driven primarily by nominations of one for the other (bullying and victimization on the part of the nominator predict the degree to which antipathy is associated with bullying and victimization).

Method

Participants

Participants included 359 boys and 340 girls enrolled in the 10th grade in 13 public schools in a small city in Finland. The average age of participants was 15.2 (SD = 0.19, range 14 to 17). The vast majority (99%) of participants were ethnic Finns who spoke Finnish at home. A total of 415 (59%) lived with both biological parents, 172 (25%) lived with one biological parent, 73 (10%) lived with a biological parent and a step-parent or significant other, and 39 (6%) had other living arrangements. Students were approximately equally distributed across homes with parents who held professional positions, other white collar positions, and blue collar positions. Of the 699 participants, 69% (n = 482) attended upper secondary school and 31% (n = 217) attended vocational schools. Schools ranged in size from 74 to 470 (M = 157.67, SD = 122.94).

Instruments

Participants completed items measuring bullying (“how often have you bullied other pupils this year?”) and victimization (“how often have you been bullied by other pupils this year?”) (Kaltiala-Heino, Rimpelä, Rantanen, & Rimpelä, 2000). The response scale ranged from 1 (not at all) to 4 (many times a week). This variable, like all other variables, was standardized within schools. One year test-retest correlations suggested that both bullying (r = .32) and victimization (r = .41) were stable over time, with stability scores similar to studies multiple item scales (e.g., Boulton & Smith, 1994; Salmivalli, Lappalainen, & Lagerspetz, 1998).

Participants completed a 4-item problem behavior scale (Kiuru, Aunola, Vuori, & Nurmi, 2007), adapted from the Finnish National School Health Survey (Koivusilta, Rimpelä, & Rimpelä, 1998). Two items addressed the frequency with which participants smoked cigarettes and drank alcohol over the course of a week on a scale ranging from 1 (daily smoking or weekly drinking) to 5 (never). Two items addressed the frequency with which participants committed illegal acts and consumed illicit drugs on a scale ranging from 1 (never) to 4 [at least 4 times (illegal acts) or at least 5 times (illicit drugs)]. Scores were standardized by item and then averaged. Internal reliability was adequate (α = .75).

Participants completed a 9-item measure of school burnout (Salmela-Aro, Kiuru, Leskinen, & Nurmi, 2009). Items were rated on a scale ranging from 1 (strongly disagree) to 6 (strongly agree). Items assess dissatisfaction with, disenfranchisement from, and exhaustion at school (e.g., I feel a lack of motivation in my schoolwork and I often think of giving up). Scores were standardized and averaged across items. Internal reliability was good (α = .84).

Participants reported their school grades during the previous term on a scale ranging from 4 (unacceptable) to 10 (highest passing grade). Test-retest stability across a 3-year period was high (r = .92). Unpublished data from a large, nearby school district suggests that self-reported school grades are highly correlated (r = .96) with actual school grades. North American samples have reported similar albeit smaller correlations (Dornbusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987). Scores were standardized.

Students completed a standard sociometric inventory (Coie, Dodge, & Coppotelli, 1982), nominating up to three same-grade schoolmates with whom “you most like to spend your time” (positive nominations) and up to three same-grade schoolmates with whom “you least like to spend your time” (antipathy nominations). Acceptance represents the total number of positive nominations an individual received. Rejection represents the total number of antipathy nominations an individual received. Mutual antipathies were identified as dyads with reciprocal antipathy nominations. Unilateral antipathies were identified as dyads with one unreciprocated antipathy nomination.

Of the 699 participants, 56 participated in at least one mutual antipathy (21 boys and 35 girls), 496 participated in at least one unilateral antipathy but no mutual antipathies (259 boys and 237 girls), and 147 participated in neither mutual nor unilateral antipathies (79 boys and 68 girls). Analyses of variance (ANOVA) indicated that those who participated in at least one mutual antipathy reported more victimization (M = 0.87, SD = 0.87) than (a) those who participated in at least one unilateral antipathy but no mutual antipathies (M = −0.09, SD = 0.99) and (b) those who participated in neither mutual nor unilateral antipathies (M = 0.04, SD = 1.03).

Procedure

In keeping with Finnish law and under the guidance and approval of school officials, passive parent consent procedures were employed. Parents were notified of the study and could withdraw their child at any time. Only students who provided individual assent participated in the study. Participants completed surveys in groups during regular school hours. Noninstructional school staff read the directions aloud to students and remained in the classroom to answer questions during data collection sessions. Students were informed that questionnaires would not be viewed by teachers or parents. To ensure confidentiality, students returned questionnaires in sealed envelopes. All 10th grade students in 13 senior high schools and vocational schools were invited to participate (n = 964). The overall initial participation rate was 72.5%. Participation rates ranged from 68% to 85% across schools. None of the schools had any bullying prevention programs or special curriculum about bullying at the time of data collection.

Plan of Analysis

Preliminary analyses tested for skew in all study variables, as well as for univariate and multivariate outliers. No variables required correction for skew. The few univariate outliers on acceptance (n = 7) and rejection (n = 9) were brought back to 3 standard deviations from the mean, as recommended for multivariate analyses (Tabachnick, Fidell, & Osterlind, 2001).

Interclass correlations described associations between all study variables. A chi-square analysis described the proportion of antipathy nominations that bullies, victims, and other children direct toward and receive from one another. To this end, participants were divided into three groups: (1) those 1 SD above the mean on victimization; (2) those 1 SD and above the mean on bullying; (3) those 1 SD and above the mean on both victimization and bullying; and (4) everyone else. Adjusted residual scores were used to identify nominator/nominate combinations that occurred at levels greater than chance.

The main analyses explored who dislikes bullies and victims. In these analyses, all individuals served as nominators (actors) and nominees (targets) of antipathy. The term “nominator” refers to the individual making an antipathy nomination (yes or no) and the term “nominee” refers to the individual receiving an antipathy nomination (yes or no). Hierarchical generalized logistic linear modeling (HGLLM) with HLM 6 (Raudenbush, Bryk, & Congdon, 2008) described the extent to which nominator bullying and victimization and nominee bullying and victimization predict the antipathy of a nominator for a nominee.

The Level 1 file included each nominator paired with each nominee within a school, on a separate line of data that described characteristics of the nominee (victimization and bullying) and whether the nominee received an antipathy nomination from the nominator. The Level 2 file described the characteristics of the nominators (victimization and bullying) on a separate line of data for each individual. Four demographic control variables were included in the Level 1 and Level 2 models. The nominee’s rejection score was included to ensure that the effects were not being driven by a few disliked individuals. The nominee’s acceptance score was included to disentangle liking and peer status from disliking and antipathy. The sex and school track of the nominee were also included.

Are some youth more likely to make antipathy nominations than others? The unconditional model determined whether the probability of an antipathy nomination differed across nominators. Are some youth more likely to receive antipathy nominations? The Level 1 random effects model determined the degree of association between antipathy nominations and the Level 1 predictor variables (nominee victimization and bullying). Does the association between a nominee’s characteristics and nominator antipathy for the nominee differ as a function of nominator characteristics? The Level 2 model explores whether the characteristics of the nominator (victimization and bullying) are related to the likelihood of disliking a nominee based on the nominee’s victimization and bullying. Equations for each model are given in the appendix.

A follow-up chi-square analysis described the proportion of antipathy nominations that bullies, victims, and other children directed toward and receive from one another. Adjusted residuals identified antipathy nomination combinations that occurred at levels greater and less than chance. These analyses illustrate the proportion of victim antipathy nominations directed to bullies and the proportion of bully nominations directed to victims. These analyses also test the hypothesis that others (those who were neither victims nor bullies) tend not to direct antipathy nominations disproportionately to victims or bullies. To this end, participants were divided into three groups: (1) those 1 SD above the mean on victimization (n = 27); (2) those 1 SD and above the mean on bullying (n = 46); and (3) those between 1SD above the mean and 1SD below the mean on bullying and victimization (n = 605). A small number of youth (n = 21) were classified as both victims and bullies. These children were excluded from the chi-square analysis only.

Four sets of supplemental HGLLM analyses were conducted. First, adolescents involved in mutual antipathies (n = 56) were excluded to ensure the findings were not a product of a few enemy relationships. Second, adolescents who did not make or receive any antipathy nominations (n = 147) were excluded to ensure the findings did not reflect the characteristics of those who did not contribute antipathy data. Third, adolescents without reciprocated friends (n = 71) were excluded to ensure that the findings were not driven by a few individuals who had problems establishing and maintaining close ties to other peers. Fourth, models were re-run with the addition of three new control variables (school grades, school burnout, and problem behavior of the nominator). In each of these four cases, the same pattern of statistically significant results emerged. Finally, moderator analyses evaluated all two-way interactions for Level 2 associations between nominee victimization or bullying and nominee demographic control variables (sex, school track, rejection, and acceptance) and nominee characteristics (victimization and bullying). There were no statistically significant two-way interactions.

Results

Preliminary Analyses

Intercorrelations are presented in Table 1. Positive statistically significant associations emerged between acceptance and school grades, between rejection, victimization, and bullying, and between victimization, bullying, school burnout, and problem behavior. Problem behavior was negatively associated with acceptance and school grades.

Table 1.

Intercorrelations between Study Variables

Variable 1 2 3 4 5 6
1. Rejection -
2. Acceptance −.05 -
3. Victimization   .15**   .01 -
4. Bullying   .12** −.03   .36** -
5. School Grades −.03   .21** −.05 −.07 -
6. Problem Behavior   .02 −.36**   .11**   .19** −.31** -
7. School Burnout −.05   .03   .13*   .11** −.05   .08*
Open in a new tab

Note. N = 699.

*

p < .05.

**

p < .01.

Describing the Disliked and Those Who Dislike Them

Unconditional Model

An unconditional model determined whether the probability of an antipathy nomination differed across nominators. The intercept was statistically significant (coefficient effect: 0.2987, SE = 0.0211, t = 14.16, p < .001).

Level 1 Model

The Level 1 random effects model determined whether nominee characteristics predicted nominator antipathy for the nominee. Statistically significant findings emerged for rejection (coefficient effect = 0.0422, SE = 0.0038, p < .001), victimization (coefficient effect = 0.0209, SE = 0.0103, p < .05), and bullying (coefficient effect = 0.0283, SE = 0.0121, p < .05). As nominee rejection, victimization, and bullying scores increased, the probability of a negative nomination from the nominator increased.

Level 2 Model

The Level 2 model determined whether nominator characteristics predicted the association between nominee characteristics and nominator antipathy for the nominee. The model included nominator victimization and bullying as predictors, as well as four demographic control variables (rejection, sex, school track, and acceptance). The dependent variables were (a) the Level 1 slope association between nominee victimization and nominator antipathy for the nominee, and (b) the Level 1 slope association between nominee bullying and nominator antipathy for the nominee. Table 2 summarizes the results. The same pattern of results emerged when the control variables were omitted from the model.

Table 2.

Hierarchical Generalized Linear Logistic Model Predicting Nominator Antipathy for Nominee

Variable Effect SE t test
Intercept   0.5161 0.0193 20.63**
Slope of Nominee Rejection   0.0431 0.0068   6.34**
Slope of Nominee Sex −0.0253 0.0262 −0.96
Slope of Nominee Acceptance   0.0036 0.0026   1.11
Slope of Nominee Victimization   0.0209 0.0103   1.99*
   Nominator Victimization   0.0041 0.0046   0.84
   Nominator Bullying   0.0048 0.0012   3.63**
Slope of Nominee Bullying   0.0283 0.0121   2.75*
   Nominator Victimization   0.0046 0.0009   5.61**
   Nominator Bullying   0.0017 0.0036   0.42
Open in a new tab

Note. N = 699.

*

p < .05.

**

p < .01.

Nominee victimization

Nominator bullying predicted the association between the nominee’s victimization and the nominator’s antipathy for the nominee. As the nominator’s bullying increased, the strength of the association between nominee victimization and nominator antipathy for the nominee increased. Figure 1 depicts this result. Compared to nominators low on bullying (−1 SD), nominators high on bullying (+1 SD) were more apt to dislike the nominee as a function of the nominee’s victimization scores. Nominator victimization did not predict the association between nominee victimization and the nominator’s antipathy for the nominee.

Figure 1.

Figure 1

Open in a new tab

Slope of Association between Nominee Victimization and Nominator Antipathy for Nominee as a Function of Nominator Bullying.

Nominee bullying

Nominator victimization predicted the association between the nominee’s bullying and the nominator’s antipathy for the nominee. As the nominator’s victimization increased, the strength of the association between nominee bullying and nominator antipathy for the nominee increased. Figure 3 depicts this result. Compared to nominators low on victimization (−1 SD), nominators high on victimization (+1 SD) were more apt to dislike the nominee as a function of the nominee’s bullying. Nominator bullying did not predict the association between nominee bullying and the nominator’s antipathy for the nominee.

Follow-up Analyses

Table 3 describes the results of a 3 (nominator) X 3 (nominatee) chi-square that describes the proportion of nominations directed toward and received from bullies, victims, and other peers. The statistically significant chi-square indicates that antipathy nominations were not randomly assigned, χ2(4) = 233.26, p < .001. Statistically significant adjusted residuals identified three nomination combinations that occurred at above chance levels: Victims nominations directed toward bullies (SR = 9.7), bullies nominations directed toward victims (SR = 11.4), and others (adolescents who were neither bullies nor victims) nominations directed toward others (SR = 11.7). Statistically significant adjusted residuals also identified four nomination combinations that occurred at below chance levels: Others nominations directed toward bullies (SR = −6.2) and victims (SR = −9.4), bullies nominations directed toward others (SR = −7.9), and victims nominations directed toward others (SR = −8.1).

Table 3.

Antipathy nominations made by and directed toward bullies, victims, and others

Nominators
Nominees Bullies Victims Others Total Nominations Received
Bullies (n = 46)
   n (expected n) 12 (9.4) 24 (4.8) 49 (70.7) 85
   Adjusted residual 1.0 9.7* −6.2*
   % of nominations made by nominator directed to bullies 17.6% 68.6% 9.6%
Victims (n = 27)
   n (expected n) 27 (4.7) 2 (2.4) 13 (34.9) 42
   Adjusted residual 11.4* −0.3 −9.4*
   % of nominations made by nominator directed to victims 39.7% 5.7% 2.5%
Others (n = 605)
   n (expected n) 29 (53.9) 9 (27.8) 449 (405.3) 487
   Adjusted residual −7.9* −8.1* 11.7*
   % of nominations made by nominator directed to others 42.6% 25.7% 87.9%
Total Nominations Made 68 35 511 614
Open in a new tab

Note.

*

p < .01. N = 678. Bullies were 1 SD above the mean on self-reports of bullying. Victims were 1 SD above the mean on self-reports of victimization.

Discussion

That bullies and victims dislike one another is perhaps not newsworthy. What is newsworthy is that neither, particularly victims, is especially disliked by the rest of their age-mates. In the present study we employed a novel hierarchical generalized logistic linear modeling approach to identify whether disliking for bullies and victims is driven by the degree to which the nominator reported being victimized or being a bully. Like others (Perry, Kusel, & Perry, 1988), we found that rejection was positively associated with bullying and victimization. Our HGLLM analyses demonstrated that the elevated rejection scores of bullies and victims are disproportionately a function of the dislike that bullies have for victims and that victims have for bullies. Nominators low on bullying and low on victimization are similarly unlikely to direct antipathies toward others on the basis of these attributes.

This is not to say that bullies and victims are well liked. Some bullies who are perceived to be popular are among the least liked members of a peer group (De Bruyn, Cillessen, & Wissink, 2010). Victims do not fare any better with peers; they are neither popular nor accepted (Asher & McDonald, 2009). The fact that most classmates do not prefer the company of bullies and victims, however, does not necessarily mean that bullies and victims are actively avoided. Results from the present study suggest that animosity for bullies and victims is a characteristic of those who claim these attributes. The results are consistent with a goal-framing approach view, which suggests that status and threat motivate antipathy (Baumeister & Leary, 1995; Veenstra, Lindenberg, Munniksma, & Dijkstra, 2010).

The findings are not an artifact of friendlessness or enemy relationships. Although the co-occurrence of friendlessness, victimization, and rejection is quite high (Scholte et al., 2009), supplemental analyses excluding the friendless produced the same pattern of results. Because mutual antipathies often involve bully-victim dyads (Card & Hodges, 2007), supplemental analyses also excluded mutual antipathies. The results remained the same, indicating that isolates or a few pairs of enemies did not drive the findings. Finally, it is worth noting that characteristics of the nominator, such as school track and school grades, did not predict antipathy for the nominee; nor did they predict the association between antipathy for the nominee and the nominee’s reports of victimization or bullying. These results, combined with the chi-square descriptive statistics on who nominates whom, bolster confidence in our conclusion that antipathy for victims and bullies is not widely shared by most members of a peer network.

Our concurrent findings do not speak to the causal mechanisms associations between antipathy, victimization, and bullying. Indeed, our analyses describe associations between variables, not characteristics of dyadic relationships. Given that we had neither class play style nominations nor information on who bullied whom, we can only speculate as to the origins of specific antipathies. Bullies may victimize those whom they dislike and victims may dislike those who bully them. But other processes may also be at play. Bullying and antipathy may follow from rejection or the dissolution of a close relationship. Bullying may be a response to perceived antipathy. Antipathy may arise because of antecedent differences in salient features that characterize bullies and victims (Laursen et al., 2010). Status differences may give rise to both bullying and antipathy (Salmivalli, 2010; Hawley, Little, & Card, 2007). The current results do suggest that conceptual models may need to be reconsidered to account for the fact that most students do not direct antipathies toward bullies or victims.

The results have important implications for parents, professionals, and scholars. For parents and teachers, the findings suggest that many victims have problems that stem from experiences with bullies, which can be monitored, rather than problems with the peer group as a whole, which can be difficult to follow. For clinicians, the findings suggest that bullies and victims may not be universally despised, a message that can be communicated in ways that alleviate interpersonal concerns and boost self-worth. This finding implies that interventions aimed at pairing victims (or bullies) with children who are not involved in bullying and victimization may be particularly effective at building their social skills and reducing the likelihood of a detrimental developmental pattern. For scholars, the findings should provide impetus for new research efforts that take into account who dislikes whom in order to better understand how complex interpersonal relationships with friends and peer groups may set the stage for antipathy. Already some have taken some initial steps in this direction with an examination of sex differences in antipathy targets and nominators (Veenstra, Lindenberg, Munniksma, & Dijsktra, 2010). This work, along with the current study, suggests that there is value in looking beyond the rejection of bullies and victims, to explore the contexts and mechanisms of the peer culture that underlie rejection.

The study has limitations to consider. First, the negative nominations used in the present study were indicators of someone “with whom you least liked to spend time,” which is an indirect assessment of antipathy. It is unlikely that this strategy failed to capture disliked targets, although it may have included as targets some who are avoided but not disliked. Second, single-item self-report measures of victimization and bullying during the previous year are suboptimal. We think that behavioral self-perceptions are an important tool for understanding how an individual makes decisions (Pelham, 1991), but until the findings are replicated with peer reports and measures that span shorter timeframes and more specific assessments bullying and victimization behaviors, caution should be exercised in interpreting the results. Third, the use of a limited nomination protocol for antipathy may mean that some individuals were not able to list all of the individuals they disliked. Given that very few individuals actually made three antipathy nominations, this concern may not have mattered much in this sample. Fourth, we do not claim to capture all sources or targets of antipathy. Low agreeable youth, for instance, represent a distinct category of children who are probably disliked by broad swaths of the peer group (Laursen, Hafen, Rubin, Booth-LaForce, & Rose-Krasnor, 2010). Finally, the data comes from a single cohort of 10th graders living in a stable, homogeneous, prosperous region of Northern Europe. Generalizations to youth of other ages who live in different circumstances may not be warranted.

The fact that victims may not be disliked by most age-mates does not in any way diminish the discomfort and pain brought on by exposure to bullying. But it can offer some hope. We know that children with friends are less likely to be bullied than children without friends, even among those who are at risk for peer rejection (Hodges, Boivin, Vitaro, & Bukowski, 1999). Children who are bullied need to know that they are not universally disliked and that there are considerable benefits to resisting the temptation to withdraw from all forms of peer contact.

Figure 2.

Figure 2

Open in a new tab

Slope of Association between Nominee Bullying and Nominator Antipathy for Nominee as a Function of Nominator Victimization.

Acknowledgments

Support for the FinEdu study was provided by the Academy of Finland (134931, 139168, 7213486) and the Jacobs Foundation. Brett Laursen received support from the U.S. National Institute of Child Health and Human Development (HD068421) and the U.S. National Science Foundation (0923745, 0909733).

Appendix

Unconditional Model

Nominator Antipathy for Nominee = β0 (intercept) + r

Level 1 Model

Association between Nominee Characteristics and Nominator Antipathy for Nominee = β0 (intercept) + β1 (rejection score of Nominee) + β2 (Nominee sex) + β3 (Nominee school track) + β4 (acceptance score of Nominee) + β5 (victimization of Nominee) + β6 (bullying of Nominee) + r

Level 2 model

Level 1

Association between Nominee Characteristics and Nominator Antipathy for Nominee = β0 (intercept) + β1 (rejection score of nominee) + β2 (nominee sex) + β3 (nominee school track) + β4 (acceptance score of nominee) + β5 (victimization of nominee) + β6 (bullying of nominee) + r

Level 2

β0 = γ00 + μ0

β1 = γ10 (rejection score of nominee) + μ1

β2 = γ20 (nominee sex) + μ2

β3 = γ30 (nominee school track) + μ3

β4 = γ40 (acceptance score of nominee) + μ4

β5 = γ50 (victimization of nominee) + γ51 (victimization of nominator) + γ52 (bullying of nominator) + μ5

β6 = γ60 (bullying of nominee) + γ51 (bullying of nominator) + γ62 (victimization of nominator) + μ6

Contributor Information

Christopher A. Hafen, University of Virginia

Brett Laursen, Florida Atlantic University.

Jari-Eri Nurmi, University of Jyväskylä.

Katariina Salmela-Aro, University of Helsinki.

References

  1. Abecassis M. I hate you just the way you are: Exploring the formation, maintenance, and need for enemies. In: Hodges EVE, Card NA, editors. New Directions for Child and Adolescent Development: Vol. 102, Enemies and the Darker Side of Peer Relations. San Francisco: Jossey-Bass; 2003. pp. 5–22. [DOI] [PubMed] [Google Scholar]
  2. Asher SR, McDonald KL. The behavioral basis of acceptance, rejection, and perceived popularity. In: Rubin KH, Bukowski WM, Laursen B, editors. Handbook of peer interactions, relationships, and groups. New York: Guilford Press; 2009. pp. 232–248. [Google Scholar]
  3. Baumeister RF, Leary MR. The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin. 1995;117:497–529. [PubMed] [Google Scholar]
  4. Boulton MJ, Smith PK. Bully/victim problems in middle-school children: Stability, self-perceived competence, peer perceptions and peer acceptance. British Journal of Developmental Psychology. 1994;12:315–329. [Google Scholar]
  5. Card NA. Antipathetic relationships in child and adolescent development: A metaanalytic review and recommendations for an emerging area of study. Developmental Psychology. 2010;46:516–529. doi: 10.1037/a0017199. [DOI] [PubMed] [Google Scholar]
  6. Card NA, Hodges EVE. Parent-child relationships and enmity with peers: The role of avoidant and preoccupied attachment. In: Hodges EVE, Card NA, editors. New Directions for Child and Adolescent Development: Vol. 102, Enemies and the Darker Side of Peer Relations. San Francisco: Jossey-Bass; 2003. pp. 23–37. [DOI] [PubMed] [Google Scholar]
  7. Card NA, Hodges EVE. Victimization within mutually antipathetic peer relationships. Social Development. 2007;16:479–496. [Google Scholar]
  8. Coie JD, Dodge KA, Coppotelli H. Dimensions and types of social status: A crossage perspective. Developmental Psychology. 1982;18(4):557–570. [Google Scholar]
  9. De Bruyn EH, Cillessen AHN, Wissink I. Associations of peer acceptance and perceived popularity with bullying and victimization in early adolescence. Journal of Early Adolescence. 2010;30:543–566. [Google Scholar]
  10. Dornbusch SM, Ritter PL, Leiderman PH, Roberts DF, Fraleigh MJ. The relation of parenting style to adolescent school performance. Child Development. 1987;58:1244–1257. doi: 10.1111/j.1467-8624.1987.tb01455.x. [DOI] [PubMed] [Google Scholar]
  11. Erath SA, Petit GS, Dodge KA, Bates JE. Who dislikes whom, and for whom does it matter: Predicting aggression in middle childhood. Social Development. 2008;18:577–596. doi: 10.1111/j.1467-9507.2008.00497.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Espelage DL, Holt MK, Henkel RR. Examination of peer-group contextual effects on aggression during early adolescence. Child Development. 2003;74:205–220. doi: 10.1111/1467-8624.00531. [DOI] [PubMed] [Google Scholar]
  13. Guerra NG, Williams KR, Sadek S. Understanding bullying and victimization during childhood and adolescence: A mixed methods study. Child Development. 2011;82:295–310. doi: 10.1111/j.1467-8624.2010.01556.x. [DOI] [PubMed] [Google Scholar]
  14. Hawley PH, Little TD, Card NA. The allure of a mean friend: Relationship quality and processes of aggressive adolescents with prosocial skills. International Journal of Behavioral Development. 2007;31:170–180. [Google Scholar]
  15. Haynie D, Nansel T, Eitel P, et al. Bullies, victims, and bully/victims: distinct groups of at risk youth. Journal of Early Adolescence. 2001;21:29–49. [Google Scholar]
  16. Hodges EVE, Boivin M, Vitaro F, Bukowski WM. The power of friendship: Protection against an escalating cycle of peer victimization. Developmental Psychology. 1999;35:94–101. doi: 10.1037//0012-1649.35.1.94. [DOI] [PubMed] [Google Scholar]
  17. Kaltiala-Heino R, Rimpelä M, Rantanen P, Rimpelä A. Bullying at school—an indicator of adolescents at risk for mental disorders. Journal of Adolescence. 2000;23:661–674. doi: 10.1006/jado.2000.0351. [DOI] [PubMed] [Google Scholar]
  18. Kiuru N, Aunola K, Vuori J, Nurmi J-E. The role of peer groups in adolescents’ educational expectations and adjustment. Journal of Youth and Adolescence. 2007;36:995–1009. [Google Scholar]
  19. Koivusilta L, Rimpelä A, Rimpelä M. Health related lifestyle in adolescence predicts adult educational level: A longitudinal study from Finland. Journal of Epidemiology and Community Health. 1998;52:794–801. doi: 10.1136/jech.52.12.794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Laursen B, Bukowski WM, Nurmi J-E, Marion D, Salmela-Aro K, Kiuru N. Opposites detract: Middle school peer group antipathies. Journal of Experimental Child Psychology. 2010;106:240–256. doi: 10.1016/j.jecp.2010.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Laursen B, Hafen CA, Rubin KH, Booth-LaForce C, Rose-Krasnor L. The distinctive difficulties of disagreeable youth. Merrill-Palmer Quarterly. 2010;56:80–103. doi: 10.1353/mpq.0.0040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Ng ESW, Carpenter JR, Goldstein H, Rasbash J. Estimation in generalised linear mixed models with binary outcomes by simulated maximum likelihood. Statistical Modeling. 2006;6:23–42. [Google Scholar]
  23. Olweus D. Bully/Victim problems among school children: Basic facts and effects of a school based intervention program. In: Rubin K, Pepler D, editors. The development and treatment of childhood aggression. Hillsdale, NJ: Erlbaum; 1989. [Google Scholar]
  24. Olweus D. Bully/victim problems in school. In: Eienarsen S, Hoel H, Zapf D, Cooper C, editors. Bullying and Emotional Abuse in the Workplace. New York: Taylor & Francis; 2002. pp. 62–78. [Google Scholar]
  25. Pelham BW. On confidence and consequence: The certainty and importance of selfknowledge. Journal of Personality and Social Psychology. 1991;60:518–530. doi: 10.1037//0022-3514.60.4.518. [DOI] [PubMed] [Google Scholar]
  26. Perry DG, Kusel SJ, Perry LC. Victims of peer aggression. Developmental Psychology. 1988;24:807–814. [Google Scholar]
  27. Prinstein MJ, Aikins JW. Cognitive moderators of the longitudinal association between peer rejection and adolescent depressive symptoms. Journal of Abnormal Child Psychology. 2004;32:147–158. doi: 10.1023/b:jacp.0000019767.55592.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Raudenbush SW, Bryk AS. Hierarchical linear models. Applications and data analysis methods. New York: Sage; 2002. [Google Scholar]
  29. Raudenbush S, Bryk A, Congdon R. Hierarchical linear and nonlinear modeling: Version 6.06. Lincolnwood IL: Scientific Software International; 2008. [Google Scholar]
  30. Rodkin PC, Berger C. Who bullies whom? Social status asymmetries by victim gender. International Journal of Behavioral Development. 2008;32:473–485. [Google Scholar]
  31. Rodkin PC, Farmer TW, Pearl R, Van Acker R. They’re cool: Social status and peer group supports for aggressive boys and girls. Social Development. 2006;15:175–204. [Google Scholar]
  32. Salmela-Aro K, Kiuru N, Leskinen E, Nurmi J-E. School-Burnout Inventory (SBI): Reliability and validity. European Journal of Psychological Assessment. 2009;25:48–57. [Google Scholar]
  33. Salmivalli C. Bullying and the peer group: A review. Aggression and Violent Behavior. 2010;15:112–120. [Google Scholar]
  34. Salmivalli C, Lappalainen M, Lagerspetz KMJ. Stability and change of behavior in connection with bullying in schools: A two-year follow-up. Aggressive Behavior. 1998;24:205–218. [Google Scholar]
  35. Salmivalli C, Peets K. Bullies, victims, and bully-victim relationships in middle childhood and adolescence. In: Rubin K, Bukowski WM, Laursen B, editors. Handbook of Peer Interactions, Relationships, and Groups. New York: Guilford Press; 2009. pp. 322–340. [Google Scholar]
  36. Scholte RHJ, Overbeek G, ten Brink G, et al. The significance of reciprocal and unilateral friendships for peer victimization in adolescence. Journal of Youth and Adolescence. 2009;38:89–100. doi: 10.1007/s10964-008-9287-6. [DOI] [PubMed] [Google Scholar]
  37. Sijtsema R, Veenstra R, Lindenberg S, Salmivalli C. Empirical test of bullies' status goals: Assessing direct goals, aggression, and prestige. Aggressive Behavior. 2009;35:57–67. doi: 10.1002/ab.20282. [DOI] [PubMed] [Google Scholar]
  38. Tabachnick BG, Fidell LS, Osterlind SJ. Using Multivariate Statistics. New York: Harper Collins; 2001. [Google Scholar]
  39. Veenstra R, Lindenberg S, Munniksma A, Dijkstra JK. The complex relation between bullying, victimization, acceptance, and rejection: Giving special attention to status, affection, and sex differences. Child Development. 2010;81:510–516. doi: 10.1111/j.1467-8624.2009.01411.x. [DOI] [PubMed] [Google Scholar]
  40. Viding E, Simmonds E, Petrides KV, Frederickson N. The contribution of callous-unemotional traits and conduct problems to bullying in early adolescence. Journal of Child Psychology and Psychiatry. 2009;50:471–481. doi: 10.1111/j.1469-7610.2008.02012.x. [DOI] [PubMed] [Google Scholar]

RESOURCES