Correlates of—and Consequences for—Bully-Victims in a Sample of Serious Adolescent Offenders

Abstract

Previous research on bullying has demonstrated that youth who engage in bullying and are also victims of bullying are at increased risk for maladjustment. Somewhat less investigated are the correlates of—and consequences associated with—this small group of persons who are simultaneously perpetrators and victims of bullying, commonly referred to as bully-victims. This article extends prior research by investigating bully-victims in a sample of serious adolescent offenders (n = 1,354) who were followed for 7 years after their adjudication for a serious juvenile offense. Analyses indicated that bully-victims had high levels of both symptomology and psychopathy as well as lower levels of temperance. Bully-victims were at increased risk of experiencing a higher number of arrests at the end of the 7-year follow-up period. Implications for future research are highlighted.

Keywords

bullying bully-victims serious adolescent offenders antisocial behavior

Bullying perpetration and especially bullying victimization have long been of interest to researchers in a variety of areas including psychology, work/organizations, criminology, and public health (Olweus, 1978). Most of the research on bullying has taken place within the school context, but more recent attention has been paid to workplace bullying and especially cyberbullying. Although a review of findings from the almost half a century of research is beyond the scope of this article, it is sufficient to conclude that both perpetrators and especially victims of bullying suffer from a range of psychosocial adjustment problems and experience adverse consequences and effects that permeate many different life domains (cf. Connell, Morris, & Piquero, 2015; Hong & Espelage, 2012; Hansen, Steenberg, Palic, & Elklit, 2012; Hawker & Boulton, 2000).¹

An important observation arose in the early 2000s that called attention to the small set of persons who were simultaneously perpetrators and victims of bullying, or “bully-victims” (Espelage & Horne, 2008; Haynie et al., 2001; Schwartz, 2000; Schwartz, Proctor, & Chien, 2001). The degree of overlap among this particular group, relative to victim-only, bully-only, or no-status comparison groups (neither bully nor victim) has been identified in a many studies, but the exact prevalence of this group has been difficult to pinpoint because of sample and methodological differences across studies (cf. Cook, Williams, Guerra, & Kim, 2009; Modecki, Minchin, Harbaugh, Guerra, & Runions, 2014). In fact, prevalence estimates have ranged from a low of 1% to as high as 30% (see Solberg, Olweus, & Endresen, 2007). Despite these variations, evidence is fairly consistent for correlates associated with—and adverse effects of—being a bully-victim. For example, bully-victims experience problems in psychosocial functioning that include poor social and emotional skills, school adjustment and educational success, interpersonal relationships and are also overrepresented in antisocial behavior (see Connell, Morris, & Piquero, 2016; Georgiou & Stavrinides, 2008; Haynie et al., 2001).

Collectively, these studies have provided a foundational knowledge base for this important social problem. Not surprisingly, evidence-based prevention and intervention programs have been developed for bullies, victims of bullying, and bully-victims that address the risk factors associated with membership in these groups as well as help such persons overcome the adverse consequences associated with bullying more generally (see review in Ttofi & Farrington, 2011). At the same time, several limitations with this line of research limit some of its generalizability, which the current study seeks to overcome.

In particular, most studies rely on general population samples from childhood and early adolescence. As a result, the extent to which the main findings from the extant research are generalizable to other samples, especially offender-based samples, is currently unknown. Understanding precursors and outcomes for bully-victims in an offender sample is important because these antisocial youth may be engaged in some of the highest rates of bullying and victimization and involved in the most serious episodes (see, e.g., Cook, Williams, Guerra, Kim, & Sadek, 2010). Studying bully-victims in an offender sample is also important because serious adolescent offenders are at a precipice in their development, especially during mid- to late adolescence—a point at which they may continue offending or desist from offending. Therefore, this study also extends current research by investigating official offending patterns among bully-victims, which provides an important compliment to the existing studies that rely on self-reported measures of offending. Only a few studies have examined official arrest records, but these have relied on community-based samples which do not often come to the attention of the formal justice system (see Klomek, Sourander, & Elonheimo, 2015; Piquero, Connell, Piquero, Farrington, & Jennings, 2013). Thus, whether prior study findings emerge among a sample of serious adolescent offenders remains in question.

Accordingly, this article uses data from a longitudinal study of serious adolescent offenders who were followed for 7 years after their commission of a serious felony offense that brought them to the attention of the juvenile justice system in two large U.S. cities, Philadelphia and Phoenix (i.e., the Pathways to Desistance Study). Specifically, this study examines the prevalence of bully-victims in this large sample relative to three comparison groups: bully perpetrators only, bully victims only, and the no-status group. Secondly, this investigation explores some of the antecedents associated with membership in these four groups. Finally, this study examines the extent to which these groups differ in their criminal offending at the end of the study period, 7 years after entry into the Pathways Study. This last feature of our investigation is of particular interest, as we rely on a search of national arrests which is rare in longitudinal studies of offending.

Data

The Pathways to Desistance Study is a longitudinal study of 1,354 serious youthful offenders transitioning from adolescence to young adulthood in Philadelphia (Philadelphia County), PA (n = 700) and Phoenix (Maricopa County), AZ (n = 654; Mulvey, 2011). The sample of 1,354 youth represented about one in three adolescents adjudicated on the enumerated charges in each locale during the recruitment period (November 2000–January 2003). To be eligible for inclusion, a juvenile had to have a current referral to the court for either a felony or serious misdemeanor offense (the charges were overwhelmingly felony offenses). Among the 1,354 youth in the study, 244 were processed in adult court, while the remaining 1,110 were processed in juvenile court. About half of the youth were incarcerated at the baseline interview.

The sample was, on average, 16 years of age at the time of the baseline interview (SD = 1.14, range = 14–18), largely male (86.4%), and was ethnically diverse, with 44% of youth identifying as African American, 29% Latino, 25% Caucasian, and 2% Other.²At baseline, 71.79% of the sample was enrolled in school, with only 24 youth having graduated high school at the start of the study. By the end of the 7-year follow-up period, about 27% of the sample obtained a general education development, about 22% attained a high school diploma, and about 18% obtained either a college degree or more commonly a trade license of some sort.

After juveniles and parents/guardians provided informed consent, youth participated in a baseline interview followed by a series of time-point interviews at 6-month intervals for the first 3 years and yearly thereafter through Year 7. Retention rates were high, with an average of 90% of participants interviewed at each time point. Information regarding the rationale and overall design of the study can be found in Mulvey et al. (2004), while details regarding recruitment, a description of the full sample, and the study methodology may be found in Schubert et al. (2004).

Dependent Variables

The current study focuses on two outcomes. The first dependent variable, which will also be used as an independent variable in subsequent analyses, is a four-category indicator for bully-victim status. Two questions obtained at the baseline interview inquired about bullying: (1) “You ever bully anyone?” and (2) “Were you ever picked on by a bully?” Response options for both questions were no/yes. From these 2 items, a four-category variable indicating bully-victim status was created (see Figure 1): (1) no status, not bully/not victim; n = 753 (55.61%); (2) victim only (n = 293; 21.64%); (3) bully only (n = 149; 11%); and (4) bully-victim, bullied and was victim of bullying; n = 159 (11.74%). The second outcome variable is the total number of official arrests by the end of the 84-month follow-up period (i.e., 7 years after the baseline interview). Official records (based on petitions found in juvenile records in each jurisdiction prior to age 18 and arrests appearing in Federal Bureau of Investigation records thereafter) are assigned to a recall period based on the date of the petition/arrest as mapped onto the dates covered within a recall period. A binary indicator of a petition/arrest by wave was used, where, as noted earlier, the recall periods from Waves 1 through 6 correspond to 6-month recalls, while Waves 7 through 10 correspond to 12-month recalls. Note that for the second part of the analyses, the indicators were summed to have one overall measure of the total official number of petitions/arrests (M = 2.057, SD = 1.808, range = 0–9).³

Figure 1.

Prevalence of no-status, bully victims, bully perpetrators, and bully-victims.

Independent Variables

Three demographic variables were also included. Gender was coded as male (= 1) or female (= 2). We also control for two race/ethnicity variables: Black (= 1, else = 0) and Hispanic (= 1, else = 0), with Whites serving as the reference category.

For the investigation of the correlates of bully-victim status, analyses investigated 11 correlates that prior research has found to be related to bullying perpetration and bullying victimization, many of which are social–emotional or psychosocial variables.

Temperance is a combination of the impulse control (e.g., “I say the first thing that comes into my mind without thinking enough about it.”) and suppression of aggression (e.g., “People who get me angry better watch out.”) subscales from the Weinberger Adjustment Inventory, which assesses an individual’s social–emotional adjustment within the context of external constraints (Weinberger & Schwartz, 1990). All of the items ask participants to rank how much (1 = false to 5 = true) their behavior in the past 6 months matches a series of statements. Higher scores on each of the subscales indicate more positive behavior, that is, more temperance, after reverse-coding necessary items (mean = 2.874, SD = 0.854, range = 1–5). It is important to note that the temperance measure was fit using a second-order confirmatory factor analysis model, where temperance was the second-order factor and impulse control and suppression of aggression were the first-order factors. The model showed acceptable fit (comparative fit index = .93, root mean square error of approximation = .06).

The Psychopathy Checklist: Youth Version (Forth, Kosson, & Hare, 2003) assesses psychopathic characteristics among youth. Procedures advocated by the authors suggest a 60- to 90-min, semistructured interview to assess the youth’s interpersonal style, obtain information on a variety of aspects of his or her history and current functioning, and assess the credibility of his or her statements. This study uses the total score (mean = 15.906, SD = 7.733, range = 0–39).

The Brief Symptom Inventory (BSI) GLOBAL is a 53-item self-report inventory on which participants rated the extent to which they have been bothered (0 = not at all to 4 = extremely) in the past week by various symptoms (see Derogatis & Melisaratos, 1983). The BSI has nine subscales designed to assess individual symptom groups: somatization (SOM, e.g., “Faintness or dizziness”), obsessive–compulsive (OC, e.g., “Having to check and double-check what you do”), interpersonal sensitivity (IS, e.g., “Feeling inferior to others”), depression (DEP, e.g., “Feeling no interest in things”), anxiety (ANX, e.g., “Feeling tense or keyed up”), hostility (HOS, e.g., “Having urges to break or smash things”), phobic anxiety (PHB, e.g., “Feeling uneasy in crowds, such as shopping or at a movie”), paranoid ideation (PAR, e.g., “Others not giving you proper credit for your achievements”), and psychoticism (PSY, e.g., “The idea that something is wrong with your mind”; mean = 0.559, SD = 0.532, range = 0.02–3.25).

The Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler, 1999) produces an estimate of general intellectual ability based on two subtests: Vocabulary (42 total items that require the subject to orally define four images and 37 words presented both orally and visually) and Matrix Reasoning (35 incomplete grid patterns that require the participant to select the correct response from five possible choices). Administered in approximately 15 min, the WASI is a quick estimate of an individual’s level of intellectual functioning, with higher scores indicating greater intellectual ability. The WASI is linked to both the Wechsler Intelligence Scale for Children (WISC-III) and the Wechsler Adult Intelligence Scale (WAIS-III) and has been normed for individuals’ ages 6–89 years (mean = 84.524, SD = 13.029, range = 55–128).

The Exposure to Violence Inventory (ETV; Selner-O’Hagan, Kindlon, Buka, Raudenbush, & Earls, 1998) was modified for this study to assess the frequency of exposure to violent events. Items from the ETV document the types of violence the adolescent has both experienced (i.e., Victim—6 items, e.g., “Have you ever been chased where you thought you might be seriously hurt?”) and observed (i.e., Witnessed—7 items, e.g., “Have you ever seen someone else being raped, an attempt made to rape someone or any other type of sexual attack?”). The total measure was used, which includes the count of items where the participant was a victim and the count of items where the participant witnessed violence (mean = 5.343, SD = 2.993, range = 0–13).

Mental health/substance use problems were assessed with a variety of instruments. Specifically, eight modules from the Composite International Diagnostic Interview (CIDI) were used to assess the lifetime presence of major depression, dysthymia, mania, alcohol abuse and dependence, drug abuse and dependence, and posttraumatic stress disorder (PTSD). The CIDI is a comprehensive, fully structured clinical assessment of psychiatric disorders (based on Diagnostic and Statistical Manual of Mental Disorders (DSM) IV criteria). Also, the Revised Children’s Manifest Anxiety Scale was used to identify youth with high anxiety symptoms, and a total anxiety score was computed based on 28 items across three subscales (Physiological Anxiety, Worry/Oversensitivity, and Social Concerns/Concentration). Following Schubert, Mulvey, and Glasheen (2011, p. 928), scores at or above 1 SD of the sample mean, were taken as indicators of “high anxiety” for the current analyses. Attention and hyperactivity problems (ADHD) were assessed at baseline using the Disruptive Behavior Disorders Scale, which comprises the DSM-III-R diagnostic criteria for disruptive behavior disorders. All of the aforementioned assessments used youth self-report, except for the assessment of ADHD symptoms, which was based on the parent/guardian report, described as more valid in this context (see Jensen et al., 1999). This study focused on four clusters of mental health problems indicating the presence or absence of each cluster: affective disorders (depression, dysthymia, or mania; 5.26%), anxiety disorders (PTSD or high anxiety, 3.91%), alcohol abuse/dependence (8.14%), and drug abuse/dependence (19.20%).⁴

Number of prior petitions/arrests reflects the participant’s history of legal involvement based on the total number of prior petitions to court before the baseline assessment. This count included the study initial referring petition as well as petitions prior to the initial referring petition (mean = 2.919, SD = 2.144; range = 1–15). The link between prior and future criminal behavior is well established (see Piquero, Farrington, & Blumstein, 2003).

Family member criminal history is an indicator obtained from the youth regarding whether anyone in the respondent’s family had ever been arrested (mean = 0.783, SD = 0.411, range = 0–1). Youth who have family members with previous antisocial behavior are at risk for future offending themselves (see Farrington, Coid, & Murray, 2009).

Analysis Plan

As noted earlier, two variables serve as outcome variables, both of which necessitate separate analytic methods. In the first part of the analysis predicting bully-victim status, a multinomial logistic regression is used, given the nature of the categorical outcome, bully-victim status. In this analysis, the no-status group serves as the reference group and all of the independent variables noted earlier will be used as covariates. For this analysis, multinomial logistic regression estimates were converted into relative risk ratios. In the second phase of the analysis predicting the frequency of rearrests at the 84-month follow-up, a negative binomial regression is utilized, which is appropriate, given the count-oriented nature of the rearrest outcome variable. For this second analysis, the focus is on how the bully-victim status groups demarcated at the baseline interview relate to the total number of rearrests. As well, the negative binomial regression coefficients were converted into incidence rate ratios.

Results

Descriptive Statistics

Before presenting the substantive findings that emerge from the regression analyses, it is worthwhile to provide some baseline comparisons across the four bully-victim status groups across the entire range of variables considered in this analysis. Although space precludes a detailed overview of all of these findings, it is important to highlight the fact that 13 of the 15 overall comparisons evince significant differences (neither gender nor Hispanic were significantly different across groups). As can be seen from Table 1, the bully-victim group tended to have the highest value on the majority of risk factors and also had the highest number of rearrests compared to the other groups. With respect to rearrests in particular, post hoc tests indicated that the bully-victim group differs significantly from both the no-bully/no-victim group and the victim-only group but does not significantly differ from the bully-only group.⁵

Table 1.

Bivariate Comparisons of All Variables by Bully-Victim Status Indicator.

Variable	No Victim-No Bully	Victim Only	Bully Only	Bully-Victim	F Test
Rearrests	3.26 (3.27)	2.92 (3.06)	3.13 (3.34)	3.96 (4.06)	3.42*
Gender	1.13 (0.33)	1.15 (0.35)	1.17 (0.38)	1.13 (0.33)	0.98
Black	0.47 (0.50)	0.35 (0.48)	0.40 (0.49)	0.28 (0.45)	9.68*
Hispanic	0.32 (0.47)	0.34 (0.47)	0.33 (0.47)	0.41 (0.49)	1.62
Temperance	3.05 (0.82)	2.88 (0.84)	2.46 (0.78)	2.42 (0.82)	40.90*
Exposure to violence	4.91 (2.89)	5.48 (3.10)	6.14 (2.80)	6.41 (3.04)	16.20*
IQ	83.49 (12.68)	86.32 (13.46)	84.92 (13.56)	85.78 (12.97)	3.97*
Family arrest	0.77 (0.42)	0.76 (0.43)	0.87 (0.33)	0.83 (0.38)	3.82*
Depression	0.04 (0.18)	0.08 (0.28)	0.05 (0.23)	0.08 (0.27)	4.03*
PTSD	0.03 (0.16)	0.06 (0.24)	0.04 (0.20)	0.06 (0.23)	3.12*
Alcohol abuse	0.07 (0.25)	0.07 (0.26)	0.11 (0.32)	0.14 (0.35)	4.00*
Drug abuse	0.19 (0.39)	0.14 (0.35)	0.26 (0.44)	0.25 (0.44)	4.17*
Priors	2.88 (2.03)	2.66 (2.07)	3.37 (2.51)	3.16 (2.34)	4.43*
BSI	0.47 (0.50)	0.68 (0.58)	0.58 (0.46)	0.74 (0.57)	17.32*
PCL:YV	14.76 (7.56)	15.47 (7.41)	19.13 (7.75)	19.17 (7.451)	24.15*

Note. BSI = Brief Symptom Inventory; IQ = intelligence quotient; PCL: YV = Psychopathy Checklist: Youth Version; PTSD = posttraumatic stress disorder.

*p < .05.

Distinguishing Between Bully-Victim Status Groups

Table 2 presents the results for predicting bully-victim status. For ease of presentation, each set of comparisons is discussed starting with the comparison between victim only and no status. As can be seen, five variables significantly distinguish between the two groups. Here, Blacks, Hispanics, and youth with drug abuse/dependence are less likely to be victims-only and more likely to be in the no-status group, whereas youth who report exposure to violence and youth scoring higher on the BSI are more likely to be victims-only compared to being in the no-status group.

Table 2.

Multinomial Logistic Regression Distinguishing Between Bully-Victim Categorizations.

	Victim Only Versus No Status			Bully Only Versus No Status			Bully Victim Versus No Status
Variable	RRR	SE	95% CI	RRR	SE	95% CI	RRR	SE	95% CI
Gender	0.88	.20	[0.56, 1.39]	1.73	.49*	[0.99, 3.03]	0.96	.30	[0.52, 1.76]
Black	0.57	.12*	[0.38, 0.85]	0.85	.23	[0.50, 1.43]	0.51	.14*	[0.30, 0.86]
Hispanic	0.68	.14*	[0.46, 1.02]	0.80	.22	[0.47, 1.35]	0.80	.19	[0.49, 1.30]
Temperance	0.94	.10	[0.76, 1.56]	0.47	.07*	[0.35, 0.62]	0.52	.08*	[0.39, 0.69]
Exposure to Violence	1.06	.03*	[0.99, 1.12]	1.03	.04	[0.96, 1.12]	1.06	.04	[0.98, 1.14]
WASI IQ	1.01	.01	[0.99, 1.02]	1.01	.008	[0.99, 1.03]	1.01	.01	[0.99, 1.03]
Family arrest	0.87	.16	[0.60, 1.25]	1.54	.45	[0.87, 2.73]	1.14	.31	[0.68, 1.94]
Depression	1.22	.41	[0.63, 2.37]	0.83	.40	[0.32, 2.12]	1.18	.48	[0.53, 2.64]
PTSD	1.30	.49	[0.62, 2.73]	1.14	.59	[0.41, 3.14]	1.31	.60	[0.53, 3.22]
Alcohol abuse	0.91	.27	[0.50, 1.63]	0.99	.34	[0.51, 1.94]	1.44	.44	[0.79, 2.62]
Drug abuse	0.68	.14*	[0.45, 1.02]	1.10	.26	[0.67, 1.75]	0.99	.24	[0.62, 1.58]
Priors	0.96	.04	[0.88, 1.03]	1.04	.05	[0.95, 1.13]	0.98	.05	[0.89, 1.07]
BSI	1.76	.28*	1.29, 2.41]	0.89	.19	[0.58, 1.39]	1.39	.27*	[0.94, 2.04]
PCL: YV	1.00	.01	[0.98, 1.02]	1.04	.02*	[1.01, 1.07]	1.04	.02*	[1.01, 1.07]

Note. BSI = Brief Symptom Inventory; CI = confidence interval. IQ = intelligence quotient; PCL: YV = Psychopathy Checklist: Youth Version; PTSD = posttraumatic stress disorder; RRR = relative risk ratio; SE = standard error; WASI = Wechsler Abbreviated Scale of Intelligence.

*p < .05.

In the second comparison of bully-only to no-status group, three variables are significant. Participants with greater temperance are less likely to be bullies and more likely to be in the no-status group, youth scoring high on the psychopathy checklist as well as females are more likely to be in the bully-only group compared to the no-status group.⁶

When comparing bully-victims to the no-status group, results show that four variables distinguish between the two groups. In particular, Black youth as well as youth reporting a higher degree of temperance were less likely to be in the bully-victim group compared to the no-status group, whereas youth scoring higher on the BSI and the psychopathy checklist were more likely to be bully-victims compared to the no-status group.

One additional analysis predicting only those youth who reported that they bullied others and were themselves victims of bullying was conducted. Those logistic regression results are presented in Table 3 in the form of odds ratios. As can be seen, three variables predict status as a bully-victim compared to the three other groups combined. As has been the case, Black youth as well as youth scoring higher on temperance were less likely to be bully-victims, whereas youth scoring high on the psychopathy checklist were more likely to be bully-victims.

Table 3.

Logistic Regression Predicting Bully-Victim Status.

Variable	Odds Ratio	SE	95% CI
Gender	0.90	.27	[0.50, 1.61]
Black	0.61	.16*	[0.37, 1.01]
Hispanic	0.93	.22	[0.59, 1.48]
Temperance	0.59	.08*	[0.45, 0.77]
Exposure to violence	1.04	.04	[0.96, 1.11]
WASI IQ	1.00	.01	[0.99, 1.02]
Family arrest	1.13	.29	[0.68, 1.87]
Depression	1.12	.42	[0.54, 2.33]
PTSD	1.13	.47	[0.50, 2.55]
Alcohol abuse	1.48	.42	[0.85, 2.58]
Drug abuse	1.09	.24	[0.69, 1.69]
Priors	0.98	.04	[0.90, 1.07]
BSI	1.18	.21	[0.83, 1.67]
PCL: YV	1.03	.01*	[1.001, 0.06]

Note. BSI = Brief Symptom Inventory; CI = confidence interval; IQ = intelligence quotient; PCL: YV = Psychopathy Checklist: Youth Version; PTSD = posttraumatic stress disorder; SE = standard error; WASI = Wechsler Abbreviated Scale of Intelligence.

*p < .05.

Predicting the Frequency of Rearrests

The final part of this investigation examines how membership in one of the three groups (bully-victim, victim-only, and bully-only) is related to the frequency of rearrests by the end of the 7-year study compared to the no-status group. As shown in the negative binomial regression results reported in Table 4, compared to the no-status group, only one of the other groups had a significantly higher number of rearrests, namely, the bully-victim group. Participants who reported bullying as well as being the victim of bullying accumulated a significantly higher number of rearrests than the no-status group (incidence rate ratio = 1.215). Moreover, these results held in another model that controlled for gender and race/ethnicity, the results of which showed, not surprisingly, that males were significantly more likely than females to incur a significantly higher number of rearrests in the follow-up period (see Table 5). Neither of the coefficients for Black and Hispanic was significant in this additional model. In short, bully-victims accumulate a higher number of rearrests in the follow-up period than those in the no-status group.⁷

Table 4.

Negative Binomial Regression Predicting Rearrest Frequency.

Variable	Incidence Rate Ratio	SE	95% CI
Bully-victim	1.22	.12*	[1.01, 1.47]
Victim only	0.90	.07	[0.77, 1.04]
Bully only	.96	.10	[0.79, 1.17]

Note. CI = confidence interval; SE = standard error.

*p < .05.

Table 5.

Negative Binomial Regression Predicting Rearrest Frequency With Gender and Race/Ethnicity Controls.

Variable	Incidence Rate Ratio	SE	95% CI
Gender	0.44	.04*	[0.36, 0.53]
Black	0.89	.07	[0.77, 1.03]
Hispanic	0.94	.07	[0.80, 1.09]
Bully-victim	1.19	.11*	[0.99, 1.43]
Victim only	0.91	.07	[0.78, 1.06]
Bully only	0.96	.09	[0.79, 1.17]

Note. CI = confidence interval; SE = standard error.

*p < .05.

Discussion

Much of the extant research on bullying perpetration and victimization has been concentrated on samples of youth and adolescents in public schools. Importantly, it is less understood whether extent findings also replicate on samples of adolescent offenders, who represent a sample of particular policy relevance. As well, extant research on bullying perpetration and bullying victimization has, with some important exceptions, tended to treat bully offenders and bully victims as two distinct groups. Much less is known regarding the correlates and outcomes associated with the group of persons, commonly referred to as bully-victims, who are simultaneously bullies and victims of bullying (see Cook et al., 2010). This article extended prior research by examining the correlates of bully-victims and the extent to which bully-victims evince different patterns of antisocial behavior, in the context of official rearrests, in a large, longitudinal study of serious adolescent offenders who were followed for 7 consecutive years beginning in mid-adolescence. This is one of the first studies to examine bullying and victimization in a serious offender sample and to investigate correlates of membership in bully, victim, and bully-victim groups within such a sample. Perhaps most importantly, this is the first study to show that among a serious offender sample, bully-victims exhibit a higher frequency of rearrests when compared to all other groups combined or simply the no-status group. A few key observations are worth noting.

First, bullying was not ubiquitous among this serious offender sample. Over fifty percent of these youth who had already committed a serious offense reported that they had neither engaged in bullying nor been the victim of a bully. Problem behavior theory (Jessor & Jessor, 1977) might suggest that bullying would be correlated with other forms of antisocial behavior, but these results suggest that might not be the case, at least among this sample of serious adolescent offenders. The proportion of youth in this serious offender sample who reported being both bullies and victims, 11%, was in the general range that has been found with large samples of typically developing youth (7% in a large survey of urban youth in Grades 7–12; Carlyle & Steinman, 2007, and 15% in a large survey of students in Grades 7–12; Stein, Dukes, & Warren, 2007).

For this serious offender sample, there were some differences in correlates of being the victim of a bully, being a bully perpetrator, and being a bully-victim. For example, compared to a no-status group, bully victims were disproportionately White (compared to Blacks and Hispanics), and a higher number of symptoms as assessed by the BSI that was related to a higher likelihood of being the victim of a bully. On the other hand and in contrast to research for nonoffender samples, bully perpetrators when compared to the no-status group were disproportionately women and were youth who scored high on the psychopathy checklist, while youth evincing higher temperance were less likely to be bully perpetrators and more likely to be in the no-status group. Persons who were bully-victims were distinguishable from other groups in having a higher score on the psychopathy checklist, whereas Black youth and youth with greater temperance were less likely to be bully-victims. Thus, the patterns of results across the comparisons indicate some differences but also some similarities—namely, temperance appears to lessen the risk of being a bully and being a bully-victim, a higher degree of symptoms increases the risk of being the victim of bullying as well as being a bully-victim, whereas White youth are at increased risk of being a victim of bullying and being a bully-victim.

These results are only somewhat consistent with those from previous studies with typically developing samples. Past research has shown that bully-victims have difficulties with emotion regulation, a construct at least somewhat akin to temperance (Toblin, Schwartz, Gorman, & Abou-ezzzedine, 2005). However, other correlates differed for this sample of serious adolescent offenders. Research with community samples shows that males are higher than females on bullying perpetration (Carlyle & Steinman, 2007), whereas among this serious offender group, being female predicted engaging in bullying—which may be a function of the selection of certain kinds of females into the juvenile justice system. For a community sample, African American youth were more likely to be bully-victims, whereas for this serious offender sample, bully-victims were more likely to be White (Carlyle & Steinman, 2007). Additional research will be needed to both replicate and explore the reasons for these differences.

Perhaps the most important finding was that adolescents identified as bully-victims at the baseline interview accumulated a higher number of official rearrests at the 7-year follow-up when compared to all other groups combined or simply the no-status group. In short, bully-victims tend to experience higher symptomology, score high on psychopathy, exhibit less temperance, and offend at higher rates than their nonbully-victim counterparts. Interestingly enough, these same correlates have also been identified as key risk factors for antisocial behavior more generally, especially among serious adolescent offenders (see Loeber & Farrington, 1998; Moffitt, 1993; Monahan, Steinberg, & Piquero, 2015).

Being both a perpetrator and victim of bullying highlights the need for theory as well as risk/needs-based assessments when bully-victims come to the attention of the mental health and juvenile/criminal justice systems. Studies of both typically developing and offender samples could also be helpful in guiding the development of theory. Research with typical samples suggests that involvement in bullying and victimization may be related to poor emotional regulation (Shields & Cicchetti, 2001; Wilton, Craig, & Pepler, 2000), especially for youth who are both aggressive and victims of bullying (Toblin et al., 2005). The results of this study of serious adolescent offenders fit this general theoretical picture. Bully-victims were lower in temperance and higher on overall psychopathology, which may indicate emotional dysregulation. Further analysis is needed to understand the individual and mental health correlates associated with membership across these groups across a range of samples as well as demographic distinctions within samples in order to build better theories and to help better inform prevention and intervention efforts—especially within the mental and juvenile justice systems which more often than not operate independently of one another. This is an especially important issue because as this research shows, bully-victims do evince mental health problems and do differentially offend. Better coordination across these social service agencies is needed. On this policy-related issue, there are some evidence-based strategies that appear to be useful with respect to bullying more generally, most of which are school-based interventions (Bradshaw, Mitchell, & Leaf, 2010; Merrell, Gueldner, Ross, & Isava, 2008; Vreeman & Carroll, 2007), which is not a surprise because most of the attention regarding bullying has been centered among school-based populations. Yet, practitioners in the mental and especially the juvenile justice systems also need to not only consider screening for bully-victims but also explore the extent to which school-based programs translate to adolescent offenders in juvenile treatment-based settings as well.

Several limitations regarding this study should be noted. First, as is the case with all of the research on bullying perpetration and bullying victimization, definitional issues should be kept in mind. The current work used very simple but straightforward questions aimed at measuring offending and victimization, which permitted the respondent to interpret what bullying, meant to him or her. Replication of this work with other more specific bullying assessments should be undertaken. Second, although an extensive range of correlates for which to differentiate between the different bully-victim groups was used, other correlates should be considered in subsequent research. Third, this work did not examine the context in which bullying perpetration and bullying victimization occurred. Perhaps the bullying that exists on a playground, a shopping mall, a school, or a correctional facility varies according to both individual and situational characteristics. Further understanding of how persons and environments meet in time and place to lead to a heightened risk for perpetration and victimization would be an especially interesting direction for future analysis. Given the extent to which offending and victimization influence the development of adolescents more generally—and permeate throughout the life course, continued research on these topics is important.

The myriad of issues surrounding bullying, both perpetration and victimization, are unfortunately part of the experience that many children and adolescents must face on a daily basis. One can turn to any media platform to learn of these bullying hardships. Yet, characterizations of offenders and victims of bullying, and especially of those who are both bullies and victims (bully-victims), are more often-than-not depicted in an armchair quarterback style; that is, based more on fiction than fact. Our study, using a sample of serious adolescent offenders who are making important life-course transitions from mid-adolescence to early adulthood, highlights some of the important risk factors that distinguish bully-victims from their counterparts. Our research also uncovered potentially adverse negative consequences manifested by increased offending among such persons. Studies like ours and future ones are necessary for theoreticians, policy makers, and the general public to have a better understanding of the facts associated with bully-victims in order to develop frameworks and prevention/intervention efforts that are grounded in science, a science that opens one’s eyes rather than makes one blind.

Footnotes

Authors’ Note

The content of this article, however, is solely the responsibility of the authors and does not necessarily represent the official views of these agencies.

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: The project described was supported by funds from the following: Office of Juvenile Justice and Delinquency Prevention (2007-MU-FX-0002), National Institute of Justice (2008-IJ-CX-0023), John D. and Catherine T. MacArthur Foundation, William T. Grant Foundation, Robert Wood Johnson Foundation, William Penn Foundation, Center for Disease Control, National Institute on Drug Abuse (R01DA019697), Pennsylvania Commission on Crime and Delinquency, and the Arizona Governor’s Justice Commission. We are grateful for their support.

Notes

References

Bradshaw

C. P.

Mitchell

M. M.

Leaf

P. J.

(2010). Examining the effects of school-wide positive behavioral interventions and supports on student outcomes: Results from a randomized controlled effectiveness trial in elementary schools. Journal of Positive Behavior Interventions, 12, 133–148.

Carlyle

K. E.

Steinman

K. J.

(2007). Demographic differences in the prevalence, co-occurrence, and correlates of adolescent bullying at school. Journal of School Health, 77, 623–629.

Connell

N. M.

Morris

R. G.

Piquero

A. R.

(2015). Exploring the link between being bullied and adolescent substance use. Victims & Offenders. doi:10.1080/15564886.2015.1055416

Connell

N. M.

Morris

R. G.

Piquero

A. R.

(2016). Predicting bullying: Exploring the contributions of childhood negative life experiences in predicting adolescent bullying behavior. International Journal of Offender Therapy and Comparative Criminology, 60, 1082–1096.

Cook

C. R.

Williams

K. R.

Guerra

Kim

(2009). Variability in the prevalence of bullying and victimization: A cross-national methodological analysis. In Jimerson

S. R.

Swearer

S. M.

Espelage

D. L.

(Eds.), The handbook of bullying in schools: An international perspective (pp. 347–362). London, England: Routledge.

Cook

C. R.

Williams

K. R.

Guerra

N. G.

Kim

T. E.

Sadek

(2010). Predictors of bullying and victimization in childhood and adolescence: A meta-analytic investigation. School Psychology Quarterly, 25, 65–83.

Derogatis

Melisaratos

(1983). The brief symptom inventory: An introductory report. Psychological Medicine, 13, 595–605.

Espelage

D. L.

Horne

(2008). School violence and bullying prevention: From research-based explanations to empirically based solutions. In Brown

Lent

(Eds.), Handbook of counseling psychology (4th ed., pp. 588–606). Hoboken, NJ: Wiley.

Farrington

D. P.

Coid

J. W.

Murray

(2009). Family factors in the intergenerational transmission of offending. Criminal Behaviour & Mental Health, 19, 109–124.

10.

Forth

Kosson

Hare

(2003). The hare psychopathy checklist: Youth version, technical manual. New York, NY: Multi-Health Systems.

11.

Furlong

M. J.

Sharkey

J. D.

Felix

E. D.

Tanigawa

D. I. A. N. E.

Greif-Green

(2010). Bullying assessment: A call for increased precision of self-reporting procedures. In Jimerson

S. R.

Swearer

S. B.

Espelage

D. L.

(Eds.), Handbook of bullying in schools: An international perspective (pp. 329–346).New York, NY: Routledge.

12.

Geerken

M. R.

(1994). Rap sheets in criminological research: Considerations and caveats. Journal of Quantitative Criminology, 10, 3–21.

13.

Georgiou

S. N.

Stavrinides

B. P

. (2008). Bullies, victims, and bully-victims: Psychosocial profiles and attribution styles. School Psychology International, 29, 574–589.

14.

Hansen

T. B.

Steenberg

L. M.

Palic

Elklit

(2012). A review of psychological factors related to bullying victimization in schools. Aggression and Violent Behavior, 17, 283–387.

15.

Hawker

D. S. J.

Boulton

M. J.

(2000). Twenty years research on peer victimization and psychosocial maladjustment: A meta-analytic review of cross-sectional studies. Journal of Child Psychology and Psychiatry, 41, 441–455.

16.

Haynie

D. L.

Nansel

Eitel

Crimp

A. D.

Saylor

Simons-Morton

(2001). Bullies, victims, and bully/victims: Distinct groups of at-risk youth. Journal of Early Adolescence, 21, 29–49.

17.

Hong

J. S.

Espelage

D. L.

(2012). A review of research on bullying and peer victimization in school: An ecological system analysis. Aggression and Violent Behavior, 17, 311–322.

18.

Huizinga

Elliott

D. S.

(1986). Reassessing the reliability and validity of self-report delinquent measures. Journal of Quantitative Criminology, 2, 293–327.

19.

Jensen

P. S.

Rubio-Stipec

Canino

Bird

H. R.

Dulcan

M. K.

Schwab-Stone

M. E.

Lahey

B. B.

(1999). Parent and child contributions to diagnosis of mental disorders: Are both informants always necessary? Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1569–1579.

20.

Jessor

(1977). Problem behavior and psychological development: A longitudinal study of youth. New York, NY: Academic Press.

21.

Kirk

D. S.

(2006). Examining the divergence across self-report and official data sources on inferences about the adolescent life-course of crime. Journal of Quantitative Criminology, 22, 107–129.

22.

Klomek

A. B.

Sourander

Elonheimo

(2015). Bullying by peers in childhood and effects of psychopathology, suicidality, and criminality in adulthood. The Lancet Psychiatry, 2, 930–941.

23.

Loeber

Farrington

D.P.

(Eds.) (1998). Serious and violent juvenile offenders: Risk factors and successful interventions. Newbury Park, CA: Sage.

24.

Merrell

K. W.

Gueldner

B. A.

Ross

S. W.

Isava

D. M.

(2008). How effective are school bullying intervention programs? Meta-analysis of intervention research. School Psychology Quarterly, 23, 26–42.

25.

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.

26.

Moffitt

T. E.

(1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological review, 100, 674–701.

27.

Monahan

Steinberg

L.D.

Piquero

A. R.

(2015). Juvenile justice policy and practice: A developmental perspective. In Tonry

(Ed.), Crime and justice: A review of research (Vol. 44., pp. 577–619). Chicago, IL: University of Chicago Press.

28.

Mulvey

E. P.

(2011). Highlights from pathways to desistance: A longitudinal study of serious adolescent offenders. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

29.

Mulvey

E. P.

Steinberg

Fagan

Cauffman

Piquero

A. R.

Chassin

… Losoya

(2004). Theory and research on desistance from antisocial activity among serious adolescent offenders. Youth Violence and Juvenile Justice, 2, 213–236.

30.

Olweus

(1978). Aggression in the schools: Bullies and whipping boys. Washington, DC: Hemisphere/Wiley.

31.

Piquero

A. R.

Connell

N.M.

Piquero

N.L.

Farrington

D.P.

Jennings

W.G.

(2013). Does adolescent bullying distinguish between male offending trajectories in late middle age? Journal of Youth and Adolescence, 42, 444–453.

32.

Piquero

A. R.

Farrington

Blumstein

(2003). The criminal career paradigm: Background and recent developments. In Tonry

(Ed.), Crime and justice: A review of research (Vol. 33, pp. 359–506). Chicago, IL: University of Chicago Press.

33.

Piquero

A. R.

Schubert

C. A.

Brame

(2014). Comparing official and self-report records of offending across gender and race/ethnicity in a longitudinal study of serious youthful offenders. Journal of Research in Crime and Delinquency, 51, 525–555.

34.

Schubert

C. A.

Mulvey

E. P .

Steinberg

Cauffman

Losoya

Hecker

… Knight

G. P.

(2004). Operational lessons from the pathways to desistance project. Youth Violence and Juvenile Justice, 2, 237–255.

35.

Schubert

C. A.

Mulvey

E. P.

Glasheen

(2011). Influence of mental health and substance use problems and criminogenic risk on outcomes in serious juvenile offenders. Journal of the American Academy of Child & Adolescent Psychiatry, 50, 925–937.

36.

Schwartz

(2000). Subtypes of victims and aggressors in children’s peer groups. Journal of Abnormal Child Psychology, 28, 181–192.

37.

Schwartz

Proctor

L. J.

Chien

D. H.

(2001). The aggressive victim of bullying: Emotional and behavioral dysregulation as a pathway to victimization by peers. In Juvonen

Graham

(Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 147–174). New York, NY: The Guilford Press.

38.

Selner-Ohagan

Kindlon

Buka

Raudenbush

Earls

(1998). Assessing exposure to violence in urban youth. Journal of Child Psychology and Psychiatry and Allied Disciplines, 39, 215–224.

39.

Shields

Cicchetti

(2001). Parental maltreatment and emotion dysregulation as risk factors for bullying and victimization in middle childhood. Journal of Clinical Child Psychology, 30, 349–363.

40.

Solberg

M. E.

Olweus

Endresen

I. M.

(2007). Bullies and victims at school: Are they the same pupils? British Journal of Educational Psychology, 77, 441–464.

41.

Stein

J. A.

Dukes

R. L.

Warren

J. I.

(2007). Adolescent male bullies, victims, and bully-victims: A comparison of psychosocial and behavioral characteristics. Journal of Pediatric Psychology, 32, 273–282.

42.

Toblin

R. L.

Schwartz

Gorman

A. H.

Abou-ezzeddine

(2005). Social–cognitive and behavioral attributes of aggressive victims of bullying. Journal of Applied Developmental Psychology, 26, 329–346.

43.

Ttofi

M. M.

Farrington

D. P.

(2011). Effectiveness of school-based programs to reduce bullying: A systematic and meta-analytic review. Journal of Experimental Criminology, 7, 27–56.

44.

Vaillancourt

McDougall

Hymel

Krygsman

Miller

Stiver

Davis

(2008). Bullying: Are researchers and children/youth talking about the same thing? International Journal of Behavioral Development, 32, 486–495.

45.

Vreeman

R. C.

Carroll

A. E.

(2007). A systematic review of school-based interventions to prevent bullying. Archives of Pediatrics and Adolescent Medicine, 161, 78–88.

46.

Wechsler

(1999). Wechsler abbreviated scale of intelligence. New York, NY: The Psychological Corporation: Harcourt Brace.

47.

Weinberger

D. A.

Schwartz

G. E.

(1990). Distress and restraint as superordinate dimensions of self-reported adjustment: A typological perspective. Journal of Personality, 58, 381–417.

48.

Wilton

M. M. M.

Craig

W. M.

Pepler

D. J.

(2000). Emotion regulation and display in classroom victims of bullying: Characteristic expressions of affect, coping styles, and relevant contextual factors. Social Development, 9, 226–245.