Abstract
This is the first study examining the association between victimization by bullying and attachment to both parents and teachers among students who report Learning Disorders (LD) and/or Attention Deficit Hyperactivity Disorder (ADHD). A total of 1,691 seventh- and eighth-grade students in six junior high schools completed questionnaires about LD/ADHD diagnosis, victimization, and attachment to mother, father, and teacher. A regression analysis was conducted to estimate the probability for victimization (infrequent and frequent) through the various variables. Only students who reported both LD and ADHD (but not one of them) were at greater risk of frequent victimization compared with students who did not report LD/ADHD. In addition, having a secure attachment pattern to mother was associated with a decrease in the likelihood of being frequently bullied. Attachment patterns to father and teacher were not significantly associated with victimization group membership. Results suggest that children’s perception of support and attachment to mother is important above and beyond their report on LD/ADHD diagnosis.
Victimization by bullying among youth is an international public health issue (Srabstein & Leventhal, 2010). The prevalence of victimization in Israel among youth ages 11 to 15 is 9.7% for boys and 6.7% for girls, which is comparable with the prevalence in other countries (Due et al., 2009).
Little available research exists on victimization among children and adolescents with Learning Disorders (LD; for example, Conti-Ramsden & Botting, 2004; Heiman & Olenik-Shemesh, 2013; Kaukiainen et al., 2002; Mishna, 2003) or Attention Deficit Hyperactivity Disorder (ADHD; for example, Fite, Evans, Cooley, & Rubens, 2014; Unnever & Cornell, 2003; Weiner & Mak, 2009). Most studies indicate that children with LD are at greater risk of being bullied (e.g., Estell et al., 2009; Martlew & Hodson, 1991; Mishna, 2003; Nabuzoka & Smith, 1993; Rose, Monda-Amaya, & Espelage, 2011). Mishna (2003) presents factors and characteristics that make youth with LD vulnerable to peer victimization. Various individual, social, and environmental risk factors make students with LD susceptible to victimization. Other studies, however, have found that students with LDs were just as likely to be victimized by bullies as were children who did not suffer from these disorders (e.g., Kaukiainen et al., 2002; Rose et al., 2011). Rose, Forber-Pratt, Epselage, and Aragon (2013) explain that the non-significant difference between the groups in their sample could be based on group overlap, where some individuals may be inaccurately diagnosed with specific learning disabilities and others may benefit from special education services but have not been formally identified with a disability. Another study, which has found that students with LDs were just as likely to be victimized by bullies as were children who did not have LD, included only 10 children who were victimized (Kaukiainen et al., 2002).
Studies on ADHD and bullying have found that children and adolescents with ADHD report higher levels of victimization by bullying than do unaffected children (e.g., Holmberg & Hjern, 2008; Sciberras, Ohan, & Anderson, 2012; Taylor, Saylor, Twyman, & Macias, 2010; Unnever & Cornell, 2003; Weiner & Mak, 2009). Sciberras et al.’s (2012) results indicate that adolescent girls with ADHD experienced more relational and overt victimization than adolescent girls without ADHD. Weiner and Mak (2009) and Taylor et al. (2010) found similar results for mixed samples of boys and girls.
There may be different explanations for why students with LD and/or ADHD are more prone to be victims of bullying. In this study, we focus on the quality of students’ attachment to parents and teachers (that is, the extent to which students feel secure in the availability, sensitivity, and responsiveness of these significant adults). There is accumulating evidence indicating that insecurities and worries in LD/ADHD children and adolescents related to their attachment to parents (Flouri & Buchanan, 2002; Georgiou, 2008; Storebo, Rasmussen, & Simonsen, 2013) and teachers (e.g., Al-Yagon, 2012; Al-Yagon & Margalit, 2006; Al-Yagon & Mikulincer, 2004b; Commodari, 2013; Pianta, 1999; Pianta, Nimetz, & Bennett, 1997; Pianta & Stuhlman, 2004) are associated with socio-emotional difficulties and psychological maladjustment. Not many studies have examined the association between attachment to parents and teachers and children’s roles as victims and bullies (Demanet & Van Houtte, 2012; Nickerson, Mele, & Osborne, 2010; Nikiforou, Georgiou, & Stavrinides, 2013; Runions & Shaw, 2013; Smith & Myron-Wilson, 1998; Troy & Sroufe, 1987; Walden & Beran, 2010; Williams & Kennedy, 2012). Nikiforou et al. (2013) found that poor quality of attachment with parents predicts bullying and victimization. Troy and Sroufe (1987) also found evidence for such an association with attachment to parents, but their study used only a small sample of preschool children. Similarly, Williams and Kennedy (2012), examined the association between parent–child attachment and bullying/victimization, but their sample was also relatively small and included undergraduate students who were asked to look back at their past victimization experiences. To the best of our knowledge, there are no studies that have examined the association between the frequency of victimization by bullying and attachment to both parents and teachers among students who report being diagnosed with LD and/or ADHD.
The aim of the current study is to examine the risk of victimization by bullying among students who report being diagnosed with LD and/or ADHD and to consider the possible role of attachment to parents and teachers. Specifically, we made the following hypotheses: first, students who report LD or ADHD would be victimized more frequently compared with students who do not report LD/ADHD diagnoses; second, students with less secure attachment styles to parents would experience more victimization compared with students with better attachment to parents; third, students with insecure attachment style to teachers would experience more frequent victimization compared with students with more secure attachment to teachers and; finally, we will explore to what extent variance in victimization frequencies could be explained by self-reports of LD/ADHD diagnosis, and attachment styles to parents and teacher while controlling for students’ gender and grade level.
Method
The study population included all 1,691 seventh- and eighth-grade students in six junior high schools in Herzliya, a city in central Israel. The study was carried out during May and June of 2013. After obtaining parental consent, the questionnaires were distributed to the children by research assistants. Students filled in the questionnaires anonymously during a school lesson, and all questionnaires were sealed in an envelope in front of the students and brought to the researchers. The schools selected did not have a formal bully prevention program in place during the administration of the survey.
In all, 21.5% of the students did not attend the classes chosen for our evaluation or were missing for the entire day on which sampling took place, and 18 (1.06%) students refused to participate in the study. Another 10 (0.6%) questionnaires were discarded by the researchers because they were incompletely or inappropriately filled in. The total sample included 1,298 students (76.75% response rate). More than three quarters of the students (79.20%; n = 1,009) reported living with both parents. The Ministry of Educations’ Scientific Review Board and the Interdisciplinary Center (IDC) Institutional Review Board approved the study.
Measures
Demographic questionnaire
The demographic questionnaire included information on age, grade level, self-reports of LD/ADHD diagnosis, gender, household composition, and family socioeconomic status. Students were considered to have only LD, only ADHD, both LD and ADHD, neither LD nor ADHD based on their self-report of being diagnosed with LD/ADHD.
Attachment to parents
Participants completed the 15-item Hebrew adaptation (Granot & Maysless, 2001) of the Attachment Security Style Scale (Kerns, Klepac, & Cole, 1996). This scale assesses adolescents’ attachment security in parent–child relationships, using Harter’s 4-point “some kids . . . other kids” format. Participants completed two versions of the scale. In one version, they were asked about their relationship with their mother. In the other version, they were asked about their relationship with their father. Previous findings demonstrated the validity and reliability of the Attachment Security Style Scale (Kerns et al., 1996) in early adolescents (Lieberman, Doyle, & Markiewicz, 1999). Ratings on these 4-point scales were summed across the 15 items comprising each of them, to form an attachment security score to mother and an attachment security score to father ranging from 15 to 60. Higher scores indicated more secure relationships. In line with previous research (Al-Yagon, 2007; Kerns et al., 1996), a score of 45 served as a specific categorical cut-off point to distinguish secure from insecure child–mother and child–father attachments. In the current study, the Cronbach’s alpha for the adolescent–mother scale was .83 and for the adolescent–father scale was .87.
Attachment representations of teachers
Participants completed the 25-item Children’s Appraisal of Teacher as a Secure Base (CATSB) scale (Al-Yagon & Mikulincer, 2006). This scale assesses adolescents’ perceptions of their homeroom teacher as an attachment figure along a 7-point scale. Previous findings demonstrated the validity and reliability of this scale (e.g., Al-Yagon & Mikulincer, 2006). The Availability and Acceptance subscale is comprised of 17 items, assessing the teacher as caring and as available in times of need (e.g., “My teacher is always there to help me when I need her”). The Rejection subscale is composed of eight items tapping the extent to which the adolescent perceives the teacher as rejecting (e.g., “My teacher makes me feel unwanted”). In the current study, Cronbach’s alphas for the Availability–Acceptance subscale were .96 and for the Rejecting subscale .92. On this basis, we computed two total scores for each participant by summing up the scores of the 17 availability–acceptance items (scale range 17–119) and the eight rejecting items (scale range 8–56). Higher scores reflect appraisals of homeroom teachers as more available–accepting and as more rejecting.
Bullying questionnaire
The questionnaire was based on Sourander et al. (2010). Traditional victimization by bullying was elicited with the question “How often have you been bullied at school during the past 6 months?”. Available responses varied between “1 = not at all, 2 = less than once a week, 3 = more than once a week, and 4 = most days.” To estimate victimization frequencies, the sample was divided into three groups on the basis of the victimization questionnaire. Students were assigned to the group not a victim (n = 952; 73.3%), if they indicated not at all on the bullying questionnaire. The second group, infrequent victim (n = 219; 16.9%), included students who indicated either less than once a week or more than once a week. The third group, frequent victim (n = 127; 9.8%), comprised students who indicated being victimized on most days. Table 1 presents the characteristics of the sample by victimization groups (i.e., not victimized, infrequently victimized, and frequently victimized).
Descriptive Characteristics for the Entire Sample by Victimization Frequency Groups.
Note. n.s. = not significant.
Not a victim. bInfrequent victim. cFrequent victim.
p < .05. **p < .01. ***p < .001.
Data Analyses
Chi-square tests were conducted to describe the prevalence of demographic characteristics (e.g., gender, grade, living with both parents) among the victimization frequency groups and to examine associations of victimization frequency with students’ perception of LD/ADHD diagnosis, and attachment to mother and father. Univariate analyses of variance (ANOVAs) and Scheffe’s post hoc comparisons were computed to examine differences in victimization groups’ perceptions of attachment to teacher. In addition, a Multinomial Logistic Regression (MLR) estimated the probability of being placed in either of the two groups of victimization frequencies (infrequent victim and frequent victim) versus the reference group (not a victim). The analysis allowed us to estimate the probability for victimization (infrequent and frequent) through the demographics and independent variables which were found to be significantly related to victimization: gender, grade (seventh or eighth grade), students’ reports on both LD/ADHD diagnosis, attachment to mother, attachment to father, and attachment to teacher. In essence, the results revealed which factors are important in differentiating the three groups of victimization (never, infrequent victim, and frequent victim). Analyses of variance in students’ victimization revealed that the between-school variance was not large enough to warrant multilevel analyses. All analyses were conducted using SPSS (22).
Results
Significant correlations were found between the study variables (Table 1). Chi-square test yielded significant associations between victimization and students’ gender, χ2(2) = 19.74; p < .000. Male students were significantly more likely to be in the infrequent victims (52.30%; n = 114) followed by the frequent victims groups (59.80%; n = 76), whereas female students were significantly more likely to be in the non-victimized group (58.20%; n = 552), followed by the infrequent victims groups (47.70%; n = 104). Analysis further reveal significant associations between victimization and grade level, χ2(2) = 6.991; p < .05; seventh graders were significantly more likely to be in the infrequent victims (59.70%; n = 129) followed by the frequent victims groups (53.30%; n = 65), whereas eighth graders were significantly more likely to be in the non-victimized group (50.20%; n = 475) followed by the frequent victims group (46.70%; n = 57).
Association Between Victimization Frequency and Self-Reports of LD/ADHD Diagnosis
The majority of the students (68.6%; n = 860) did not report any disorder. The smallest proportion of the sample reported they were diagnosed with only LD (5.60%; n = 73); more than one tenth (11.30%; n = 147) reported only on ADHD diagnosis, and additional 14.50% (n = 188) reported on both LD and ADHD diagnosis; see Table 1). Results indicate that only the self-reporting of both LD and ADHD was significantly associated to victimization, χ2(2) = 7.929; p < .05. Students who reported on both LD and ADHD diagnosis were significantly more likely to be in the frequent victims group (22.80%; n = 29), whereas students who did not report on both LD and ADHD diagnosis were significantly more likely to be in the not victimized (86.40%; n = 823) or in the infrequent victims groups (86.30%; n = 189).
Association Between Victimization Frequency and Attachment to Parents
Chi tests also examined victimization frequencies group differences in attachment patterns to parents (Table 2). Results revealed significant associations between victimization frequency and students’ attachment patterns to mother, χ2(2) = 22.045; p < .000, and father, χ2(2) = 9.234; p < .01. The non-victimized group was significantly more likely to have positive secure attachment to mother and father (66.20%; n = 630 and 49.50%; n = 471, respectively) followed by the infrequent victims (58.9%; n = 129 and 42.50%; n = 93, respectively), followed by the frequent victims (45.70%; n = 58 and 37.00%; n = 47, respectively).
Correlation Among Study Variables.
Note. All coefficients significant at p < .01.
Association Between Victimization Frequency and Attachment to Teacher
Univariate ANOVAs examined victimization frequencies group differences in attachment patterns to teacher (Table 3). Results reveal significant main effects for the victimization frequency groups’ perceptions of teachers’ acceptance, F(2, 1152) = 5.919; p < .01, and teachers’ rejection, F(2, 1152) = 5.891; p < .01. Scheffe’s post hoc comparison determined significant differences between the non-victims who were highest in their appraisals of their teachers as accepting and available in times of need and the frequent victims group who were the lowest in their appraisals of their teachers as accepting. Furthermore, the non-victims had significantly lower evaluation of rejection from their teachers compared with the frequent victims group. Infrequent victims did not differ significantly than the non-victims or the frequent victims in their evaluation of both attachment patterns (i.e., acceptance and rejection) to teacher.
Univariate ANOVAs for Attachment to Teacher by Victimization Group (N = 1,298).
Note. ANOVA = analyses of variance.
Higher scores indicate more positive/acceptance attachment. bHigher scores indicate more negative/rejection attachment.
p < .01.
Contribution of Attachment to Parents and Teacher, and Self-Reported LD/ADHD Diagnosis to Victimization
We estimated the probability for the frequency of victimization through self-reported LD + ADHD diagnosis, attachment to parents, and attachment to teacher, controlling for demographics, using MLR analysis. Coefficients and standard errors of the independent variables for the infrequent victim and frequent victim compared with the non-victim group (the reference group in the regressions) are presented in Table 2. Positive coefficients indicate an increase in odds of being in the infrequent or frequent groups as opposed to being in the non-victimized group; negative coefficients represent a decrease in such odds.
The results reveal that male students were 1.5 more likely to belong to the infrequent victims (odds ratio [OR] = 1.47; 95% confidence interval [CI] [1.07, 2.03]), and more than twice likely to belong to the frequent victims (OR = 2.20; 95% CI [1.47, 3.32]) compared with female students. Being in seventh grade was associated with decreased odds of being infrequent victim (OR = .67; 95% CI [0.49, 0.93], compared with non-victim. Students reporting on both LD and ADHD diagnosis were almost twice as likely to belong to the frequent victim group (OR = 1.70; 95% CI [1.01, 2.86]), compared with non-victims. Reports on both LD and ADHD, however, were not related to students’ chances of being an infrequent victim. Finally, students reporting on secure attachment pattern to mother were 0.5 less likely to be frequent victims (OR = 0.53; 95% CI [0.33, 0.84]), compared with non-victims. Attachment patterns to father and teacher were not significantly associated with victimization group membership (Tables 3 and 4).
Model 1: Multinomial Logistic Regression Parameter Estimates of Victimization Frequencies (N = 1,298).
Note. The reference group of this analysis was the not a victim group. CI = confidence interval; LD = Learning Disorders; ADHD = Attention Deficit Hyperactivity Disorder.
Higher scores indicate more positive/acceptance attachment. bHigher scores indicate more negative/rejection attachment.
p < .05. **p < .01. ***p < .001.
Discussion
Our main results indicate that students who report both LD and ADHD were at greater risk of frequent victimization by bullying compared with students who do not report LD/ADHD, whereas reporting only LD or only ADHD was not significantly associated with victimization frequency. In addition, having a secure attachment pattern to mother was associated with a decrease in the likelihood of being frequently bullied.
The results indicating that the students at risk for bullying were those who reported both LD and ADHD are in line with other studies indicating that children with both LD and ADHD perform worse than children who only have one of the disorders (Pisecco, Baker, Silva, & Brooke, 2001; Rohrer-Baumgartner et al., 2013; Sexton, Gelhorn, Bell, & Class, 2011). Interestingly, our results indicated that even after controlling for self-reported LD/ADHD, having a secure attachment pattern to mother was associated with a decrease in the odds of being frequently victimized. This may imply that children’s perception of support and attachment to mother is important above and beyond their perception of their LD/ADHD diagnosis in terms of bullying prevention. These findings are in line with previous studies highlighting the importance of reliable interpersonal relationships with parents as a protective factor among adolescents with LD (e.g., Al-Yagon, 2007, 2010, 2011a, 2011b; Al-Yagon & Mikulincer, 2004a, 2004b; Murray & Greenberg, 2001). Our results may be another confirmation of the ecological model of bullying which points to the importance of the child–environment interaction rather than one level examination (Espelage & De La Rue, 2011). In other words, bullying is a multi-faceted issue, which is best understood in the larger social context in which it occurs.
The finding that there is a significant contribution of attachment to mother but not father to victimization are contradictory to those of Nikiforou et al. (2013), who found that the role of the father in peer victimization is stronger than that of the mother. Our findings, however, are in line with a recent study by Al-Yagon (2012) among LD students, which suggested a greater number of significant paths between adolescent–mother attachment relationships and adolescents’ socio-emotional measures than for adolescent–father attachment. These results are also in line with a recent study of ours (Klomek et al., 2013), which found that the effectiveness of an intervention with LD students was mediated predominantly by the relationship with mother but not with the father. Our finding are also in agreement with prior studies indicating that, from early to late adolescence, both female and male adolescents tend to use their father less often than their mothers for support and proximity (see Lieberman et al., 1999, for review). Moreover, early studies on the families of children with disabilities have described a difference in the maternal and paternal tendency toward their children (e.g., Konstantareas & Homatidis, 1989; Levy, 1970). It may be that attachment to father is based on different experiences (Grossmann et al., 2002; Lamb, 2002; Lieberman et al., 1999).
In the current study, family-level variables (i.e., parents in our study) were more significant than were the community-level variables (i.e., attachment to teachers). Specifically, our results indicated that attachment patterns to teacher were not significantly associated with victimization group membership, when controlling for attachment to father and mother. This finding may be explained by recent studies indicating a complex association between bullying involvement and teachers’ support. Bradshaw, Sawyer, and O’Brennan (2007), for example, found that teachers were not very proficient in identifying victims. Al-Yagon and Margalit (2006) found that students with LD were not different than those without LD on measures of acceptance by teachers. Indeed, Rose et al. (2013) have found that students who reported higher levels of teachers’ support also reported higher levels of victimization. It may be that as a group, victimized students do not view their teachers as a secure base, perhaps because they feel that their teachers were ineffective in preventing their victimization.
Limitations
The current study has a few limitations, which should be taken into consideration. First, LD/ADHD diagnoses, victimization, and attachment were all based on the students’ self-reports. In addition, this is a cross sectional study and causality cannot be inferred. Last, the total explained variance of victimization was only 7% as victimization is often explained by a wide range of risk and protective factors (e.g., the students’ personality factors, parenting behaviors, family factors, and the broader school environment), which were not included in the current study. However, this is the first study that examines the association between bullying, self-reported LD/ADHD, and attachment among adolescents, and therefore future studies may make use of this initial data as they refine our techniques. In addition, we examined victimization using three categories of victimization frequencies, allowing for a more nuanced analyses that reflect different frequencies of the phenomenon and not only existence or non-existence. Future studies should examine the association between a clinical LD/ADHD diagnosis, attachment (reported by both students and parents/teachers), and victimization in a longitudinal study. In these future studies, additional risk and protective factors should be included.
Implications for Practice
Given these limitations, the clinical implications of the study suggest the use of attachment theory as a foundation upon which to develop effective anti-bullying interventions among students who report LD/ADHD diagnosis. Interventions should include work on attachment to significant others and especially to the mother.
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.
