Predictors of Bullying Behavior,Victimization,and Bully-Victim Risk Among High School Students With Disabilities

Abstract

The literature on bullying among students with disabilities is burgeoning. The purpose of this study was to examine risk factors for adolescents’ involvement in bullying across the bullying continuum. Drawing from the National Longitudinal Transition Study–2 (NLTS2), 2,870 adolescents with disabilities were sampled. Results from multinomial regression analyses indicated that internalizing symptoms and interpersonal skills were significant predictors of victimization and bully-victimization risk, respectively. Disability status emerged as a significant predictor only for bullying behavior. Ethnic differences were found for victimization roles, but not for bullying, suggesting that Caucasian students were most vulnerable to being the target of bullying or serving dual roles as bully-victims relative to students from other ethnic backgrounds. Implications for these findings are discussed.

Keywords

disability bullying special education victimization adolescents

The resultant effects of bullying on student’s risk for depression, suicide, and academic underachievement, has made bullying a national educational and public health concern (Borowsky, Taliaferro, & McMorris, 2013; Institute of Medicine & National Research Council, 2014; Klomek, Marrocco, Kleinman, Schonfeld, & Gould, 2008). For students with disabilities, the risk for direct bullying involvement may be far greater than for students without disabilities. Bully-victimization rates are 1 to 1½ times higher than the national average for children and adolescents with disabilities (Blake, Lund, Zhou, Kwok, & Benz, 2012). Students assume multiple and highly fluid roles in the bullying continuum, acting as a bully, victim, bully-victim, or bystander, with each role resulting in increasingly negative educational and psychological outcomes for involved youth (Jansen, Veenstra, Ormel, Verhulst, & Reijneveld, 2011; Rivers, Poteat, Noret, & Ashurst, 2009). Although the literature has clearly identified students with disabilities as being vulnerable to bully victimization, the factors that predict the propensity of students with disabilities to assume more aggressive roles during bullying is not well understood (Blake et al., 2014; T. W. Farmer et al., 2012; Rose, Espelage, & Monda-Amaya, 2009). The purpose of this study is to examine characteristics that predict risk of students with disabilities for bully involvement across the bullying continuum. Given that much of the research on the experience of students with disabilities has been isolated to elementary and middle school–age populations, a secondary goal of this study is to investigate the self-reported bullying experiences of high school students with disabilities (Rose & Espelage, 2012; Swearer, Wang, Maag, Siebecker, & Frerichs, 2012).

Beyond Victimization: Roles of Students With Disabilities as Aggressors in Bullying

Due to the power differential that characterizes bullying, whereby overly dominant and aggressive peers seek out students they perceive as weaker and unable to defend themselves, students with disabilities are presumed to be at greater risk of being the target of bullying than to act as perpetrators of bullying (Olweus, 1995; Sullivan, 2009; Thompson, Whitney, & Smith, 1994). Some support for this notion is based on two decades of research documenting that students with disabilities with significant social impairments and physical mobility limitations evidence elevated victimization rates in comparison with students without disabilities (for review, see Rose, Monda-Amaya, & Espelage, 2011). These findings are further corroborated by a multi-state investigation of bullying which found that, relative to students without disabilities, students with disabilities were more likely to be identified as bully-victims and victims as opposed to bullies (T. W. Farmer et al., 2012). However, at least one study has found contrasting results (Rose & Espelage, 2012).

Drawing from a sample of middle school students participating in anti-bullying intervention, Rose and Espelage (2012) reported that students receiving special education services under the category emotional disorders (ED) engaged in bullying and fighting behavior at higher rates and were less likely to endorse being subjected to victimization than students without disabilities. Although students were not classified into bullying roles, elevated bullying and aggression rates endorsed by students with ED implied that, similar to students in general education, students in special education assume multiple roles in the bullying continuum, but their involvement might vary by their disability type. What is unclear from the extant literature, however, is the reason for the tendency of students with disabilities to assume varying roles in bullying situations.

Predictors of Bullying Involvement

Adolescence marks a critical developmental period for the formation of youth identity and autonomy in which youth increasingly seek out and rely on peers rather than their parents to gauge their behavior and to evaluate their social competence. As such, the success in which adolescents are able to navigate the peer milieu to form and maintain positive peer and romantic relationships is central to their self-confidence and emotional well-being. Given that adolescent’s self-esteem hinges on peer acceptance, it is no coincidence that bullying involvement peaks during this period. Adolescence is the developmental period when youth are most vulnerable to peer opinion and more likely to use aggressive strategies to win the social approval of peers and to vie for the attention of potential romantic partners (Long & Pellegrini, 2003). National rates of bullying among youth without disabilities show that 21% of secondary students report experiencing bullying but that bullying victimization rates are highest during the transition to middle school (26%–27%) and high school (ninth grade; 23%), showing steady declines thereafter (Lessne & Cidade, 2015). At face value, these rates suggest that age serves as a protective factor against bullying involvement, with adolescents presenting less risk for being the target of bullying as they progress to higher grades in high school. However, this finding may not hold for students with disabilities (Blake et al., 2012).

The developmental demands of adolescence may make it particularly challenging for students with disabilities to escape bullying involvement given their lower social standing with peers and fewer social resources with which to draw from to combat bullying relative to students without disabilities (Estell et al., 2008; Nowicki & Sandieson, 2002). Friendships and social interactions with friendly peers as well as peer social support mitigate bullying behavior and protect against bully victimization (Demaray & Malecki, 2003). Yet, students with disabilities are less accepted by peers and have fewer friendships than same-age peers without disabilities, raising questions about their social capital for evading bullying (Boer, Pijl, Post, & Minnaert, 2013; Lindsay & McPherson, 2012). Although not all students with disabilities are socially excluded from peer networks, students with more visible disabilities (e.g., those who exhibit physical mobility limitations and autism spectrum disorders) are socially excluded (Estell et al., 2009; Lindsay & McPherson, 2012; Locke, Ishijima, Kasari, & London, 2010). These students tend to have limited contact with same-age peers in non-school activities and to not be fully integrated in peer activities within the classroom setting (Eriksson, Welander, & Granlund, 2007; T. Farmer, Wike, Alexander, Rodkin, & Mehtaji, 2015).

The social isolation students with disabilities experience may have detrimental effects on youths’ social competence and feelings of school belongingness as it reduces opportunities for positive peer socialization and for students with disabilities to build peer relationships to buffer against attacks of aggression (Pavri & Monda-Amaya, 2001; Wilson, 2004). This social isolation coupled with peer torment might also lower perceptions of perceived safety and evoke retaliation on the part of students with disabilities to defend against or ward off intimidating and harassing behavior (Evans & Eder, 1993; Saylor & Leach, 2009; Thapa, Cohen, Guffey, & Higgins-D’Alessandro, 2013). These aggressive outbursts might reinforce further alienation and incite a cycle of peer ridicule and reactive aggression that leads to students with disabilities being classified as bully-victims (Swearer, Espelage, Valliancourt, & Hymel, 2010). Collectively, these finding speak to the importance of examining not only the interpersonal relationships and perceived social support of students with disabilities as predictors of bullying involvement but also the degree to which students feel included and safe within their school environment.

It is imperative to consider other interpersonal and demographic characteristics associated with bullying involvement identified in the literature. Whereas the display of internalizing problems is identified as the most stable predictor of victimization status of students with disabilities, followed by disability status, age, race/ethnicity, and contact with peers outside of school were all associated with elevated risk for being victimized (Blake et al., 2014; Elgar, Craig, Boyce, Morgan, & Vella-Zarb, 2009; Zeedyk, Rodriguez, Tipton, Baker, & Blacher, 2014). In light of this research, it is likely that interpersonal and demographic characteristics might be equally important risk factors for predicting the bullying involvement of adolescent students with disabilities. The purpose of this study was to examine the factors that predict bully, victim, and bully-victim risk status in adolescents with disabilities.

Method

Data Source

Data for this study were drawn from the National Longitudinal Transition Study–2 (NLTS2) funded by the U.S. Department of Education and conducted by SRI International (Wagner, Kutash, Duchnowski, & Epstein, 2005). Participants in NLTS2 were sampled through a two-stage stratified random sampling process to produce nationally representative samples of youth served in special education in the United States. The initial sample included 11,270 youth with disabilities who were 13 to 16 years old in 2000. NLTS2 spanned 10 years and encompassed five waves of data collection beginning during the 2000–2001 academic year and ending during the 2008–2009 academic year (Wagner et al., 2005). Variables for this study were selected from the Wave 2 parent survey, youth survey, and school program survey.

Sample

Participants were drawn from Wave 2 of the NLTS2 dataset (N = 6,860). To obtain a sample of youth-only respondents, participants were excluded if the status of bully and/or victimization were reported by youth’s parents and if there were missing values on the self-reported survey items related to bullying: (a) youth was bullied at or on the way to and from school (victim item) and (b) youth bullied or picked on other students (bully item). The final sample consisted of 2,870 students with disabilities; demographic information for the sample is presented in Table 1.

Table 1.

Demographic Characteristics of Participants.

Characteristics	n ^a	Weighted %	SE of weighted %
Age (years)
15	330	10.18	1.43
16	820	26.60	1.87
17	820	28.48	2.48
18	700	25.93	1.96
19	220	8.81	1.18
Gender
Male	1,810	64.21	2.17
Female	1,070	35.79	2.17
Ethnicity
Caucasian	1,950	61.98	4.51
African American	540	19.93	2.44
Hispanic	310	14.84	2.90
Other	70	3.25	0.43
Primary disability
LD	340	62.38	2.05
SLI	320	4.42	0.49
ID	230	12.41	1.27
ED	230	10.21	0.87
OI	370	1.19	0.14
OHI	380	4.68	0.70
AU	200	0.70	0.11
ADHD	230	1.15	0.14
Family household income
Low: US$25,000 or less	750	37.23	2.61
Medium: US$25,001–US$50,000	770	29.67	2.18
High: More than US$50,000	1,120	33.10	2.71

Note. LD = learning disability; SLI = speech or language impairment; ID = intellectual disability; ED = severe emotional disturbance; OI = orthopedic impairment; OHI = other health impaired; AU = autism; ADHD = attention deficit hyperactivity disorder.

Unweighted sample sizes are reported in accordance with U.S. Department of Education, Institute of Education Science, National Center for Education Statistics, IES Data Security Office (1999) standards.

Measures

Dependent variable

Two youth-reported items were extracted to measure bully-victimization risk: the victim item and bully item. A four-level categorical variable was created based on student reports on these items (a) the “bully-victim” group were youth who responded yes to both items, (b) the “victim-only” group were students who answered yes to the victim item and no to the bully item, (c) the “bully-only” group were students who responded no to the victim item but yes to the bully item, and (d) the “non-involved” group were students who indicated no to both items, serving as the reference group for all analyses.

Student demographic characteristics

Student demographic variables and a family characteristic variable were selected as predictors of student’s bully involvement from the parent and school program surveys shown in Table 1: student age, gender, ethnicity, and primary disability category. Student age was treated as a continuous variable. Ethnicity was coded as three dummy variables with Caucasian serving as the reference group for all comparisons. Due to low sample size, students reported as Asian, American Indian, or from bi- or multi-ethnic backgrounds were collapsed into the “Other” group. Gender was also coded dichotomously (female = 0, male = 1). Disability status was the students’ primary disability classification reported by the school district. The disability types investigated in this study are consistent with Individuals With Disabilities Education Act (IDEA; 1997) special education categories for severe emotional disturbance (ED), intellectual disability (ID), learning disability (LD), speech or language impairment (SLI), autism (AU), orthopedic impairment (OI), other health impaired (OHI) with one exception. The NLTS2 included attention deficit hyperactivity disorder (ADHD) as a separate disability category, which means students with ADHD were not included in the OHI category. Deviation coding was used to recode the categorical disability classification variable to reflect the comparison between students with a given disability and all others. Student’s household income was assessed with a single item from the NLTS2 parent interview in which three categories were created to classify student’s household income as low (US$25,000 or less), medium (US$25,001–US$50,000), and high (more than US$50,000).

Perceived safety

Students’ perception of safety at their school was assessed with a single self-reported item (i.e., how safe youth feels at school) based on a 4-point Likert-type scale (1 = not safe at all to 4 = very safe).

School belonging

School belonging was measured by the sum of two self-reported items using 4-point Likert-type scales: “how much youth enjoys school” and “how much youth feels like he/she is a part of the school” (1 = not at all to 4 = a lot)” (α = .62).

Students’ social interactions

Two types of social interactions were measured from youth interview items: structured and unstructured social interactions. Structured social interactions were operationalized through dichotomous student responses (i.e., yes = 1, no = 0) to two statements surrounding their participation in school activities outside of class and participation in community activities in the last 12 months. A structured social interaction score was created by summing these two items, with higher scores indicating more opportunities for students to engage in structured social interactions with same-age peers (α = .43).

The unstructured social interaction variable was created by summing four standardized items: “[youth] get together with friends” (based on a 5-point scale, 0 = never to 5 = 6 or 7 days a week); “invited to social activities” (0 = no; 1 = yes), “friends call youth on the phone” (based on a 6-point scale, 1 = never to 6 = every day), and “hangs out with friends” (on a 4-point scale, 1 = not at all to 4 = 5 or more times a week) during the last 12 months (α = .38). Higher summed scores indicated a higher frequency in which youth were involved in unstructured social interactions.

Emotional and behavioral functioning

Students completed a series of items regarding their interpersonal relationships and display of internalizing behaviors. Interpersonal skills were measured by summing two items assessing the degree to which student had trouble getting along with teachers and peers (α = .52). Students’ display of internalizing behavior was assessed by summing three items measuring the frequency in which students experienced depression, felt lonely, or were disliked by others (α = .68). Higher scores were indicative of greater difficulty with social skills and elevated levels of internalizing behavior problems.

Student’s externalizing behaviors were assessed by items extracted from the school program report in which school officials reported on the frequency in which students argued with others, acted impulsively, fought, and were easily distracted (based on 3-point scale, 1 = never to 3 = very often). Responses from these items were summed to create an externalizing behavior score (α = .80). Higher scores for these variables are interpreted as more difficulty with behavioral functioning.

Perceived social support

Student’s perception of social support was measured with four self-reported items on a 5-point Likert-type response scale (1 = not at all to 5 = very much). Students indicated how much they felt supported by important people in their lives, including adults, parents, friends, and family (α = .65). Elevated scores on this variable were indicative of student feeling highly supported.

Data Analysis

A multinomial logistic regression model was tested to examine which individual characteristics predicted involvement in bullying for adolescent students with disabilities: student demographic characteristics, perceptions of perceived safety and school belonging, student’s social interactions, emotional and behavioral functioning, and perceived social support. Descriptive statistics and correlations of study variables are presented in Table 2. Because the dependent variable, bully involvement, was categorical, results were reported in odds ratios (ORs) and d_HH was used to measure effect size for each OR (Hasselblad & Hedges, 1995). To accurately and clearly interpret the OR, relative risk ratios (RRs) were calculated for only significant ORs. Because NLTS2 participants were selected through a stratification and cluster sampling design, sampling weights were used in Mplus (Version 6.1, Muthén & Muthén, 1998–2010). It is recommended that when two or more sources of data are merged for analysis, the provided weights from the source with the smallest number of participants be used (Godard et al., 2007). Therefore, weights from the youth survey were applied to address sample selection bias.

Table 2.

Descriptive Statistics and Item Correlations.

Variables	1	2	3	4	5	6	7	8	M	SD	Minimum	Maximum
1. Perceived safety	1	.09**	.01	.26**	−.16**	.22**	−.92**	−.21**	3.29	0.03	1.00	4.00
2. Structured social interaction		1	.11**	.24**	−.05**	.09**	−.06	−.07**	.75	0.04	0.00	2.00
3. Unstructured social interaction			1	.01	−.03	.06**	−.02	−.14**	11.65	0.20	2.00	16.00
4. School belonging				1	−.17**	.30**	−.10**	−.19**	5.74	0.07	2.00	8.00
5. Interpersonal skills					1	−.16**	.21**	.30**	4.08	0.09	2.00	10.00
6. Social support						1	−.09**	−.32**	17.64	0.11	4.00	20.00
7. Externalizing behavior							1	.19**	6.78	0.21	4.00	12.00
8. Internalizing behavior								1	4.71	0.08	3.00	12.00

Note. Means and standard deviations are weighted.

Correlation is significant at the .01 level (two-tailed).

The average missing data rate was 5%. Chi-square tests for categorical variables and t tests for continuous variables were conducted to compare group differences in the demographic characteristics of students with complete information and students with missing data. The lack of significant group difference suggests that the data were missing at random or completely at random (Little & Rubin, 1987; Merkle, 2011). Full information maximum likelihood (FIML) estimation method in Mplus was adopted to handle the missing data (Version 6.1, Muthén & Muthén, 1998–2010). Based on the literature, interaction effects between disability type and (a) interpersonal skills, (b) internalizing behaviors, and (c) externalizing behaviors were tested. These effects were non-significant. Therefore, the interaction terms were not included to achieve a parsimonious final model.

Results

Overall, 3.7% (n = 110) of students reported they were involved in bullying as both a bully and victim (i.e., bully-victim group), 10.3% students indicated they experienced victimization only (n = 300; victim-only group), 6.5% reported engaging in bullying only (n = 190; bully-only group), and 79.5% (n = 2,280) of the sample indicated not being involved in bullying (i.e., not involved group). The results of the multinomial regression model shown in Table 3 display the odds of students with disabilities being involved in bullying as a bully-victim, victim, or bully relative to students who were not involved in bullying (i.e., not involved group) and effect sizes.

Table 3.

Multinomial Logistic Regression Model Results.

Variables	Bully-victim			Victim-only			Bully-only
Variables	OR	95% CI	d _HH	OR	95% CI	d _HH	OR	95% CI	d _HH
Age	0.81	[0.39, 1.23]	−0.12	0.95	[0.69, 1.21]	−0.03	1.24	[0.96, 1.52]	0.12
Gender (vs. Male)	1.09	[0.11, 2.07]	0.05	1.08	[−0.17, 2.33]	0.04	1.78	[1.06, 2.50]	0.32
Primary disability (vs. all others)
LD	1.87	[0.94, 2.80]	0.35	0.81	[0.20, 1.42]	−0.12	1.07	[0.47, 1.67]	0.04
SLI	1.22	[0.19, 2.25]	0.11	0.53	[−0.18, 1.24]	−0.35	1.55	[0.97, 2.13]	0.24
ID	1.30	[0.28, 2.32]	0.14	0.72	[−0.02, 1.46]	−0.18	1.21	[0.22, 2.20]	0.11
ED	0.94	[−0.19, 2.07]	−0.03	0.98	[0.34, 1.62]	−0.01	1.00	[0.18, 1.82]	0.00
OI	0.45	[−0.7, 1.60]	−0.44	1.40	[0.81, 1.99]	0.19	0.62	[−0.04, 1.28]	0.26
OHI	0.20	[−1.65, 2.05]	−0.89	0.41	[−0.81, 1.63]	−0.49	0.20**	[−0.90, 1.30]	0.89
AU	1.06	[−0.20, 2.32]	0.03	1.18	[0.37, 1.99]	0.09	0.45	[−0.57, 1.47]	0.44
ADHD	5.87	[3.58, 8.16]	0.98	3.67	[2.31, 5.03]	0.72	5.49**	[4.27, 6.71]	0.94
Ethnicity (vs. Caucasians)
African American	0.14***	[−1.03, 1.31]	−1.08	0.13***	[−0.74, 1.00]	−1.13	0.51	[−0.46, 1.48]	0.37
Hispanic	0.06***	[−1.20, 1.32]	−1.55	0.34	[−0.94, 1.94]	−0.60	0.39	[−1.00, 1.78]	0.52
Other	0.12*	[−1.95, 2.19]	−1.17	0.50	[0.57, 1.37]	−0.38	0.29	[−1.22, 1.80]	0.68
Household income	0.68	[0.11, 1.25]	−0.21	0.97	[0.15, 0.89]	−0.02	0.73	[0.36, 1.10]	0.17
Perceived safety	1.23	[0.49, 1.97]	0.11	0.52***	[0.72, 1.14]	−0.36	1.67	[1.11, 2.23]	0.28
School belonging	0.81	[0.47, 1.15]	−0.12	0.93	[0.81, 1.73]	−0.04	0.94	[0.71, 1.17]	0.03
Social interactions
Structured	1.51	[0.72, 2.30]	0.23	1.27	[0.82, 1.22]	0.13	0.71	[0.14, 1.28]	0.19
Unstructured	0.87	[0.55, 1.19]	−0.08	1.02	[0.62, 1.08]	0.01	1.49***	[1.26, 1.72]	0.22
Emotional and behavioral functioning
Externalizing behavior	1.30	[0.96, 1.64]	0.14	0.85	[0.62, 1.08]	0.09	1.03	[0.75, 1.31]	0.02
Internalizing behavior	1.18	[1.00, 1.36]	0.09	1.31***	[1.16, 1.46]	0.15	1.07	[0.88, 1.26]	0.04
Interpersonal skills	1.32**	[1.12, 1.52]	0.15	1.26**	[1.10, 1.42]	0.13	1.10	[0.95, 1.25]	0.05
Social support	0.84*	[0.69, 0.99]	−0.10	0.92	[0.81, 1.03]	0.05	0.95	[0.82, 1.08]	0.03

Note. For all analyses, reference group is “not involved” group. OR = odds ratio; CI = confidence interval; LD = learning disability; SLI = speech or language impairment; ID = intellectual disability; ED = severe emotional disturbance; OI = orthopedic impairment; OHI = other health impaired; AU = autism; ADHD = attention deficit hyperactivity disorder.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Bully-Victim Versus Not Involved

With the exception of ethnicity, students’ demographic characteristics were not predictive of bully-victim risk. Compared with Caucasian students, African American (OR = 0.14; RR = 0.144; p ≤ .001) and Hispanic students (OR = 0.06; RR = 0.061; p ≤ .001), and students from other ethnic backgrounds (OR = 0.12; RR = 0.125; p ≤ .05) showed decreased odds of being a bully-victim relative to non-involved students. That is, Caucasian students with disabilities were about 7 times more likely to be a bully-victim than African American students with disabilities and more than 16 times more likely to be a bully-victim than Hispanic students. This finding also held when Caucasian students were compared with other racial ethnic minority students, with Caucasian students 8 times more likely to be bully-victims. Respondents who endorsed elevated levels of difficulties in their interpersonal relationships (OR = 1.32, p ≤ .01), and who had less social support (OR = 0.84, p ≤ .05) were more likely to be bully-victims than non-involved students.

Victim-Only Versus Not Involved

Ethnicity also emerged as a significant predictor of victimization risk in that Caucasian students (OR = 0.13; RR = 0.138; p ≤ .001) were more than 7 times more likely to be victimized than African American students. Whereas students’ perception of their school as safe lowered risk of students with disabilities for being victimized (OR = 0.52, p ≤ .001), respondents who had elevated internalizing behavior problems (OR = 1.31, p ≤ .001) and interpersonal difficulties (OR = 1.26, p ≤ .01) were more likely to be bullied. No significant differences were detected on gender, age, disability type, household income, school belonging, social interactions, and social support between students not involved in bully-victim behavior versus students who were bullied by others.

Bully-Only Versus Not Involved

Students with ADHD (OR = 5.49; RR = 3.82; p ≤ .01) were more than 3 times more likely to bully others than students with other disabilities. In contrast, students with OHI (OR = 0.20; RR = 0.21; p ≤ .01) were more than 4 times less likely to bully others compared with all other students with disabilities. More importantly, students who engaged in unstructured activities were more likely to bully others than students who did not (OR = 1.49, p ≤ .001).

Discussion

Research on the bullying experiences of adolescents with disabilities is in its infancy. The purpose of this study was to advance the bullying disability literature by examining predictors of bullying involvement for adolescents with disabilities. The results of the present study suggested that bullying involvement of adolescents with disabilities, in terms of students engaging in bullying behavior, being victimized, and acting as a bully-victim, was associated with unique interpersonal and demographic predictors.

Consistent with prior research, victimization risk was associated with perceptions of students with disabilities of feeling unsafe at school and display of internalizing symptoms (Zeedyk et al., 2014). These findings are not surprising given research which suggests that sadness and emotional distress serve as markers of vulnerability for bullies, making students who exhibit these symptoms more susceptible to aggression than their uninvolved peers (D’Esposito, Blake, & Riccio, 2011; T. W. Farmer, Irvin, et al., 2015). The association between perceptions of school safety and victimization risk of student with disabilities also aligns with prior research (Thapa et al., 2013). The lack of safety that students with disabilities perceive in their school likely reflects broader indicators of school disorganization and how unsafe environments contribute to a school culture that reinforces violence (G. D. Gottfredson, Gottfredson, Payne, & Gottfredson, 2005). Although prior research has found disability status to be an important predictor of victimization risk, this finding was not maintained for adolescents with disabilities in the present study. In the case of victimization, it seems that emotional functioning tends to be more predictive of victimization risk than disability status alone (Blake et al., 2014; T. W. Farmer, Irvin, et al., 2015). Ethnicity also emerged as a significant predictor of victimization risk, with African American students reporting less risk for being bullied than Caucasian students. Additional research is needed to understand why cultural differences in bullying reporting exist.

Similar to our findings related to predictors of victimization risk, we found that disability status was not a significant predictor of student’s risk for being a bully-victim. Instead, our results show that adolescents with disabilities who had poor interpersonal skills, marked by impairments in their ability to relate to others in prosocial ways, were more likely to be bully-victims than non-involved students. Further highlighting the interpersonal difficulties that bully-victims face is our finding that low social support was associated with greater bully-victim risk. These findings underscore the importance of examining social skills in bully-victimization risk for adolescents with disabilities rather than just disability status (T. Farmer, Wike, et al., 2015). However, given the low reliability for the interpersonal skills variable, a measure used to assess social skills, caution is encouraged in the interpretation of this finding. Ethnicity also emerged as a significant predictor of bully-victim risk. As noted for victimization, Caucasian students demonstrated the greatest vulnerability for being a bully-victim relative to Hispanic, African American, and students from other backgrounds. This finding is consistent with the growing research base surrounding the differential rates of victimization endorsement by ethnically diverse youth (Sawyer, Bradshaw, & O’Brennan, 2008).

Whereas disability category was not predictive of victimization or bully-victimization risk, having a diagnosis of ADHD or being eligible in the OHI special education category was predictive of bullying behavior. These findings lend support to the hypothesis that impulsivity and dysregulated emotional control, characteristics of ADHD, are associated with aggression (Taylor, Saylor, Twyman, & Macias, 2010). An unexpected but not completely surprising finding was that bullying risk increased for students with disabilities when they socialized in unstructured settings. Considerable research suggests that bullying can be reduced by increasing adult supervision and structure in the classroom setting as well as in hallways and bathrooms (Swearer et al., 2010). Therefore, it is plausible that risk for bullying involvement increases for some students with disabilities in settings where there is limited adult monitoring and supervision. However, given that the reliability for our unstructured interaction variables did not reach the threshold for psychometric adequacy, we suggest that this finding be reevaluated in future research.

Interestingly, across bullying roles, we did not find gender, household income, or age to be significant predictors of bullying involvement, which deviates from prior research (Blake et al., 2014). Our inability to detect gender differences in risk for bullying behavior, victimization, and bully-victim risk might be due to the measure we used to assess student bullying. Gender differences in bullying rates are sensitive to the forms of bullying assessed—physical and relational aggression (Blake, Kim, McCormick, & Hayes, 2011). Given that our bullying and victimization items were based on a general term, it is likely that our measure of bullying and victimization was not sensitive enough to detect the forms of bullying typically experienced by females (Wang, Iannotti, & Nansel, 2009). Our inability to detect differences in bullying involvement based on family wealth is unexpected and warrants further investigation (Elgar et al., 2009). In our past work on victimization with students with disabilities, we found family wealth to be a strong predictor of initial victimization risk (Blake et al., 2014); however, that pattern of findings was not observed in the present study. In contrast to findings on declining bullying rates with age for typically developing high school students, our results suggest that age is not a protective factor for bullying involvement for students with disabilities (Finkelhor, Ormrod, & Turner, 2009). Although these findings depart from the extant literature about developmental shifts in bullying, it does suggest that the unique social challenges that adolescents with disabilities face may make them more susceptible to bullying involvement regardless of their age.

Limitations

The results of this study should be viewed in the context of study limitations. Much of the large-scale studies on bullying experiences of students with disabilities have been based on parent report (Nordhagen, Nielsen, Stigum, & Köhler, 2005; Zablotsky, Bradshaw, Anderson, & Law, 2014). Our findings provide an advancement in the field in that it offers a portrait of the bullying involvement of adolescents with disabilities from their own perspective. To do this, this study relied solely on adolescents’ self-reported ratings of their bullying and school experiences, which may have introduced bias by increasing common method variance across study variables (Lindell & Whitney, 2001). Although this study draws from a nationally representative sample of students with disabilities, our use of a subsample of adolescents limits the generalizability of our findings to high functioning adolescents.

It is well supported that school contextual factors significantly affect student aggression and bullying (Bradshaw, Sawyer, & O’Brennan, 2009; G. D. Gottfredson et al., 2005). Unfortunately, we were unable to control for school-level variables that predict bullying involvement such as administrator disciplinary practices, student–teacher ratios, school size, and school community location due to data limitations (D. C. Gottfredson & DiPietro, 2011; Gregory et al., 2010). Failure to account for school-level characteristics associated with bullying and to investigate these predictors in a multi-level design significantly limit the generalizability of our findings. We recommend future research to examine predictors of bullying for students with disabilities while considering key school variables associated with bullying. Because this study was based on a cross-sectional sample, we are also unable to make conclusion about causality.

Given the secondary nature of the study, some of our key constructs were measured by a single-item variable or dichotomous items. This data structure limited our ability to construct variables to address core concepts of interest related to bullying with depth as indicated by our low reliabilities for some predictor variables. Whereas single-item constructs and dichotomous variables are common in large-scale national datasets, our inability to construct bullying, perceived safety, school belongingness, and student interactions variables based on multiple continuous items reduced our ability to fully assess the effects of these constructs in our study (Harris, 2013).

Implications for Practice and Future Research

Despite these limitations, our study yields important insight into the bullying experiences of adolescents with disabilities. We found that students with disabilities engaged in range of bullying roles, challenging the predominant notion that students with disabilities are incapable of initiating acts of bullying or only do so in response to retaliation (Swearer et al., 2010). Instead, our findings indicate that high school students with disabilities are as likely as students without disabilities to engage in bullying behavior, but their risk of being a bullying may vary by disability type. The current emphasis on bullying prevention in schools has focused on elementary and middle school–age populations (Merrell, Gueldner, Ross, & Isava, 2008). Findings from this study argue for the inclusion of bullying prevention programs in high schools given continued risk of adolescents with disabilities for involvement in bullying throughout their secondary careers (ages 15–19 years). Bullying is a complex social phenomenon that requires systematic changes to school climate to thwart its effects and occurrence. As such, researchers and practitioners need to create developmentally appropriate school-wide bullying prevention programs for high school students to reduce the bullying involvement of adolescents with disabilities as well as students in general education who are transitioning to high school. Given research which suggests that social emotional learning programs and Positive Behavioral Interventions and Supports (PBIS) show preliminary effectiveness in reducing bullying behavior and victimization among student students with disabilities (Espelage, Rose, & Polanin, 2015; Sullivan, Sutherland, Farrell, & Taylor, 2015; Waasdorp, Bradshaw, & Leaf, 2012), schools may want to consider integrating adapted forms of social emotional learning programs within school-wide PBIS to address bullying involvement among adolescents in high schools (Bradshaw, Bottiani, Osher, & Sugai, 2014; Domitrovich et al., 2010).

Our findings underscore the importance of considering the unique interpersonal and demographic characteristics that predict victimization and bully-victimization risk for students with disabilities. We recommend that schools implement behavioral and emotional screening in tandem with school-wide bullying assessments to assist teachers and school psychologists with identifying adolescents with disabilities who exhibit internalizing and externalizing behaviors that place them at risk for being victimized (Blake, 2015; Kamphaus, DiStefano, Dowdy, Eklund, & Dunn, 2010; Stiffler & Dever, 2015). School-wide behavioral and emotional screening can aid in the identification of students with elevated internalizing and externalizing symptoms who require more intensive individualized support services to reduce bullying (Blake, Banks, Patience, & Lund, 2015). For students identified as displaying elevated internalizing and externalizing behaviors, special education teachers are encouraged to monitor these students for bullying involvement and to provide social skills training on emotional regulation in their classrooms. We recommend that school psychologists provide more intensive counseling for these students on emotion management and conflict resolution skills to reinforce the social skills instruction offered in classrooms. Collectively, these services, folded within a large school-wide bullying prevention program rooted in a multi-tiered system of support (e.g., integration of PBIS and social emotional learning programs), have the potential to be highly effective for reducing bullying involvement for students with disabilities.

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) received no financial support for the research, authorship, and/or publication of this article.

References

Blake

J. J.

(2015). The utility of behavioral-emotional screeners in detecting student victimization risk. Manuscript submitted for publication.

Blake

J. J.

Banks

C. S.

Patience

B. A.

Lund

E. M.

(2015). School-based mental health professionals’ bullying assessment practices: A call for evidenced-based bullying assessment guidelines. Professional School Counselor, 18, 136–147. doi:10.5330/2156-759X-18.1.136

Blake

J. J.

Kim

E. S.

Lund

E. M.

Zhou

Kwok

Benz

M. R.

(2014). Predictors of bully victimization in children with disabilities: A longitudinal examination with a national dataset. Journal of Disability Policy Studies, 26, 199–208. doi:10.1177/1044207314539012

Blake

J. J.

Kim

E. S.

McCormick

A. S.

Hayes

(2011). The dimensionality of social victimization: A preliminary investigation. School Psychology Quarterly, 26, 56–69. doi:10.1037/a0022712

Blake

J. J.

Lund

E. M.

Zhou

Kwok

Benz

M. R.

(2012). National prevalence rates of bully victimization among students with disabilities in the United States. School Psychology Quarterly, 27, 210–222. doi:10.1037/spq0000008

Boer

Pijl

S. J.

Post

Minnaert

(2013). Peer acceptance and friendships of students with disabilities in general education: The role of child, peer, and classroom variables. Social Development, 22, 831–844. doi:10.1111/j.1467-9507.2012.00670.x

Borowsky

I. W.

Taliaferro

L. A.

McMorris

B. J.

(2013). Suicidal thinking and behavior among youth involved in verbal and social bullying: Risk and protective factors. Journal of Adolescent Health, 53(1, Suppl.), S4–S12. doi:10.1016/j.jadohealth.2012.10.280

Bradshaw

C. P.

Bottiani

J. H.

Osher

Sugai

(2014). The integration of Positive Behavioral Interventions and Supports and Social Emotional Learning. In Weist

M. D.

Lever

N. A.

Bradshaw

C. P.

Owens

J. S.

(Eds.), Handbook of school mental health (2nd ed., pp. 101–118).

Bradshaw

C. P.

Sawyer

A. L.

O’Brennan

(2009). A social disorganization perspective on bullying-related attitudes and behaviors: The influence of school context. American Journal of Community Psychology, 43, 204–220. doi:10.1007/s10464-009-9240-1

10.

Demaray

M. K.

Malecki

C. K.

(2003). Perceptions of the frequency and importance of social support by students classified as victims, bullies, and bully/victims in an urban middle school. School Psychology Review, 32, 471–489.

11.

D’Esposito

S. E.

Blake

J. J.

Riccio

C. A.

(2011). Adolescents’ vulnerability to victimization: Interpersonal and intrapersonal predictors. Professional School Counseling, 14, 299–309. doi:10.5330/PSC.n.2011-14.299

12.

Domitrovich

C. E.

Bradshaw

C. P.

Greenberg

M. T.

Embry

Poduska

J. M.

Ialongo

N. S.

(2010). Integrated models of school-based prevention: Logic and theory. Psychology in the Schools, 47, 71–88. doi:10.1002/pits.20452

13.

Elgar

F. J.

Craig

Boyce

Morgan

Vella-Zarb

(2009). Income inequality and school bullying: Multilevel study of adolescents in 37 countries. Journal of Adolescent Health, 45, 351–359. doi:10.1016/j.jadohealth.2009.04.004

14.

Eriksson

Welander

Granlund

(2007). Participation in everyday school activities for children with and without disabilities. Journal of Developmental Physical Disabilities, 19, 485–502. doi:10.1007/s10882-007-9065-5

15.

Espelage

D. L.

Rose

C. A.

Polanin

J. R.

(2015). Social-emotional learning program to reduce bullying, fighting, and victimization among middle school students with disabilities. Remedial and Special Education, 36, 299–311. doi:10.1177/0741932514564563

16.

Estell

D. B.

Farmer

T. W.

Irvin

M. J.

Crowther

Akos

Boudah

D. J.

(2009). Students with exceptionalities and the peer group context of bullying and victimization in late elementary school. Journal of Child and Family Studies, 18, 136–150. doi:10.1007/s10826-008-9214-1

17.

Estell

D. B.

Jones

M. H.

Pearl

Van Acker

Farmer

T. W.

Rodkin

P. C.

(2008). Peer groups, popularity, and social preference: Trajectories of social functioning among students with and without learning disabilities. Journal of Learning Disabilities, 41, 5–14. doi:10.1177/0022219407310993

18.

Evans

Eder

(1993). “NO EXIT”: Processes of social isolation in middle school. Journal of Contemporary Ethnography, 22, 139–170. doi:10.1177/089124193022002001

19.

Farmer

Wike

T. L.

Alexander

Q. R.

Rodkin

P. C.

Mehtaji

(2015). Students with disabilities and involvement in peer victimization: Theory, research, and considerations for the future. Remedial and Special Education, 36, 263–274. doi:10.1177/0741932514564563

20.

Farmer

T. W.

Irvin

M. J.

Motoca

L. M.

Leung

Hutchins

B. C.

Brooks

D. S.

Hall

C. M.

(2015). Externalizing and internalizing behavior problems, peer affiliations, and bullying involvement across the transition to middle school. Journal of Emotional and Behavioral Disorders, 23, 3–16. doi:10.1177/1063426613491286

21.

Farmer

T. W.

Petrin

Brooks

D. S.

Hamm

J. V.

Lambert

Gravelle

(2012). Bullying involvement and school adjustment of rural students with and without disabilities. Journal of Emotional and Behavioral Disorders, 20, 19–37. doi:10.1177/1063426610392039

22.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2009). The developmental epidemiology of childhood victimization. Journal of Interpersonal Violence, 24, 711–731. doi:10.1177/0886260508317185

23.

Godard

McCracken

Rollin

Van Campen

Valdes

Williamson

(2007). Special Elementary Education Longitudinal Study Waves 1, 2, & 3 data documentation and dictionary (Vol. 1). Menlo Park, CA: SRI International.

24.

Gottfredson

D. C.

DiPietro

S. M.

(2011). School size, social capital, and student victimization. Sociology of Education, 84, 69–89. doi:10.1177/0038040710392718

25.

Gottfredson

G. D.

Gottfredson

D. C.

Payne

Gottfredson

N. C.

(2005). School climate predictors of school disorder: Results from a national study of delinquency prevention in schools. Journal of Research in Crime & Delinquency, 42, 412–444. doi:10.1177/0022427804271931

26.

Gregory

Cornell

Fan

Sheras

Shih

T. H.

Huang

(2010). Authoritative school discipline: High school practices associated with lower bullying and victimization. Journal of Educational Psychology, 102, 483–496. doi:10.1037/a0018562

27.

Harris

K. M.

(2013). The Add Health Study: Design and accomplishments. Carolina Population Center, University of North Carolina at Chapel Hill. Retrieved from http://www.cpc.unc.edu/projects/addhealth/data/guides/DesignPaperWIIV.pdf

28.

Hasselblad

Hedges

L. V.

(1995). Meta-analysis of screening and diagnostic tests. Psychological Bulletin, 117, 167–178.

29.

Individuals With Disabilities Education Act of 1990, 20 U.S.C. § 1400 et seq. (1990) (amended 1997).

30.

Institute of Medicine & National Research Council. (2014). Building capacity to reduce bullying: Workshop summary. Washington, DC: The National Academies Press.

31.

Jansen

Veenstra

Ormel

Verhulst

F. C.

Reijneveld

S. A.

(2011). Early risk factors for being a bully, victim, or bully/victim in late elementary and early secondary education: The longitudinal TRAILS study. BioMed Central Public Health, 11, 1–7. doi:10.1186/1471-2458-11-440

32.

Kamphaus

R. W.

DiStefano

Dowdy

Eklund

Dunn

A. R.

(2010). Determining the presence of a problem: Comparing two approaches for detecting youth behavioral risk. School Psychology Review, 39, 395–407.

33.

Klomek

A. B.

Marrocco

Kleinman

Schonfeld

I. S.

Gould

M. S.

(2008). Peer victimization, depression, and suicidiality in adolescents. Suicide and Life-Threatening Behavior, 38, 166–180. doi:10.1521/suli.2008.38.2.166

34.

Lessne

Cidade

(2015). Student reports of bullying and cyber-bullying: Results from the 2013 school crime supplement to the national crime victimization survey. U.S. Department of Justice, Bureau of Justice Statistics, School Crime Supplement to the National Crime Victimization Survey. Retrieved from http://nces.ed.gov/pubs2015/2015056.pdf

35.

Lindell

M. K.

Whitney

D. J.

(2001). Accounting for common method variance in cross-sectional research designs. Journal of Applied Psychology, 86, 114–121. doi:10.1037//0021-9010.86.1.114

36.

Lindsay

McPherson

A. C.

(2012). Experiences of social exclusion and bullying at school among children and youth with cerebral palsy. Disability and Rehabilitation, 34, 101–109. doi:10.3109/09638288.2011.587086

37.

Little

R. J. A.

Rubin

D. B.

(1987). Statistical analysis with missing data. New York, NY: John Wiley.

38.

Locke

Ishijima

E. H.

Kasari

London

(2010). Loneliness, friendship quality and the social networks of adolescents with high-functioning autism in an inclusive school setting. Journal of Research in Special Education Needs, 10, 74–81. doi:10.1111/j.1471-3802.2010.01148.x

39.

Long

J. D.

Pellegrini

A. D.

(2003). Studying change in dominance and bullying with linear mixed modeling. School Psychology Review, 22, 401–417.

40.

Merkle

E. C.

(2011). A comparison of imputation methods for Bayesian factor analysis models. Journal of Educational and Behavioral Statistics, 36, 257–276. doi:10.3102/1076998610375833

41.

Merrell

K. W.

Gueldner

B. A.

Ross

S. W.

Isava

D. M.

(2008). How effective are school bullying intervention programs? A meta-analysis of intervention research. School Psychology Quarterly, 23, 26–42. doi:10.1037/1045-3830.23.1.26

42.

Muthén

L. K.

Muthén

B. O.

(1998–2010). Mplus User’s Guide (6th ed.). Los Angeles, CA: Author.

43.

Nordhagen

Nielsen

Stigum

Köhler

(2005). Parental reported bullying among Nordic children: A population-based study. Child: Care, Health and Development, 31, 693–701.

44.

Nowicki

E. A.

Sandieson

(2002). A meta-analysis of school-age children’s attitudes towards persons with physical or intellectual disabilities. International Journal of Disability, Development and Education, 49, 243–265. doi:10.1080/1034912022000007270

45.

Olweus

(1995). Bullying or peer abuse at school: Facts and intervention. Current Directions in Psychological Science, 4, 196–200. doi:10.1111/1467-8721.ep10772640

46.

Pavri

Monda-Amaya

(2001). Social support in inclusive schools: Student and teacher perspectives. Exceptional Children, 67, 391–411. doi:10.1177/001440290106700307

47.

Rivers

Poteat

V. P.

Noret

Ashurst

(2009). Observing bullying at school: The mental health implications of witness status. School Psychology Quarterly, 24, 211–223. doi:10.1037/a0018164

48.

Rose

C. A.

Espelage

D. L.

(2012). Risk and protective factors associated with bullying involvement of students with emotional and behavioral disorders. Behavioral Disorders, 37, 133–148.

49.

Rose

C. A.

Espelage

D. L.

Monda-Amaya

L. E.

(2009). Bullying and victimization rates among students in general and special education: A comparative analysis. Educational Psychology, 29, 761–776. doi:10.1080/01443410903254864

50.

Rose

C. A.

Monda-Amaya

L. E.

Espelage

D. L.

(2011). Bullying perpetration and victimization in Special Education: A review of the literature. Remedial and Special Education, 32, 114–130. doi:10.1177/0741932510361247

51.

Sawyer

A. L.

Bradshaw

C. P.

O’Brennan

L. M.

(2008). Examining ethnic, gender, and developmental differences in the way children being a victim of bullying self-report measures. Journal of Adolescent Health, 43, 106–114. doi:10.1016/j.jadohealth.2007.12.011

52.

Saylor

C. F.

Leach

J. B.

(2009). Perceived bullying and social support in students accessing special inclusion programming. Journal of Developmental Physical Disabilities, 21, 69–80.

53.

Stiffler

M. C.

Dever

B. V.

(2015). An example using the BASC-2 Behavioral and Emotional Screening System (BESS). In Mental Health Screening at School (pp. 107–120). Switzerland: Springer.

54.

Sullivan

P. M.

(2009). Violence exposure among children with disabilities. Clinical Child Family and Psychology Review, 12, 196–216. doi:10.1007/s10567-009-0056-1

55.

Sullivan

T. N.

Sutherland

K. S.

Farrell

A. D.

Taylor

K. A.

(2015). An evaluation of second step: What are the benefits for youth with and without disabilities? Remedial and Special Education, 36, 286–298. doi:10.1177/0741932515575616

56.

Swearer

S. M.

Espelage

Valliancourt

Hymel

(2010). What can be done about school bullying? Linking research to educational practice. Educational Researcher, 39, 38–47. doi:10.3102/0013189X09357622

57.

Swearer

S. M.

Wang

Maag

J. W.

Siebecker

A. B.

Frerichs

L. J.

(2012). Understanding the bully dynamic among students in special and general education. Journal of School Psychology, 50, 503–520. doi:10.1016/j.jsp.2012.04.001

58.

Taylor

L. A.

Saylor

Twyman

Macias

(2010). Adding insult to injury: Bullying experiences of youth with attention deficit hyperactivity disorder. Children’s Health Care, 39, 59–72. doi:10.1080/02739610903455152

59.

Thapa

Cohen

Guffey

Higgins-D’Alessandro

(2013). A review of school climate research. Review of Educational Research, 83, 357–385. doi:10.3102/0034654313483907

60.

Thompson

Whitney

Smith

P. K.

(1994). Bullying of children with special needs in mainstream schools. Support for Learning, 9, 103–106. doi:10.1111/j.1467-9604.1994.tb00168

61.

U.S. Department of Education, Institute of Education Science, National Center for Education Statistics, IES Data Security Office. (1999). Restricted-use data procedures manual (NCES Publication No. 96860rev). Retrieved from http://nces.Ed.gov/statprog/rudman/pubinfo.asp

62.

Waasdorp

T. E.

Bradshaw

C. P.

Leaf

P. J.

(2012). The impact of Schoolwide Positive Behavioral Interventions and Supports on bullying and peer rejection: A randomized controlled effectiveness trial. Archives of Pediatrics & Adolescent Medicine, 166, 149–156. doi:10.3102/0034654313483907

63.

Wagner

Kutash

Duchnowski

A. J.

Epstein

M. H.

(2005). The Special Education Elementary Longitudinal Study and the National Longitudinal Transition Study: Study designs and implications for youth with emotional disturbance. Journal of Emotional and Behavioral Disorders, 13, 25–41. doi:10.1177/10634266050130010301

64.

Wang

Iannotti

R. J.

Nansel

T. R.

(2009). School bullying among adolescents in the United States: Physical, verbal, social, and cyber. Journal of Adolescent Health, 45, 368–375. doi:10.1016/j.jadohealth.2009.03.021

65.

Wilson

(2004). The interface of school climate and school connectedness and relationships with aggression and victimization. Journal of School Health, 74, 293–299. doi:10.1111/j.1746-1561.2004.tb08286.x

66.

Zablotsky

Bradshaw

C. P.

Anderson

C. M.

Law

(2014). Risk factors among children with autism spectrum disorders. Autism, 18, 419–427. doi:10.1177/1362361313477920

67.

Zeedyk

S. M.

Rodriguez

Tipton

L. A.

Baker

B. L.

Blacher

(2014). Bullying of youth with autism spectrum disorder, intellectual disability, or typical development: Victim and parent perspectives. Research in Autism Spectrum Disorders, 8, 1173–1183. doi:10.1016/j.rasd.2014.06.001