Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Sep 1.
Published in final edited form as: Child Indic Res. 2015 Oct 12;9(3):743–756. doi: 10.1007/s12187-015-9343-

Ecological Factors of Being Bullied Among Adolescents: a Classification and Regression Tree Approach

Sung Seek Moon 1, Heeyoung Kim 2, Kristen Seay 1, Eusebius Small 3, Youn Kyoung Kim 3
PMCID: PMC5015696  NIHMSID: NIHMS741460  PMID: 27617043

Abstract

Being bullied is a well-recognized trauma for adolescents. Bullying can best be understood through an ecological framework since bullying or being bullied involves risk factors at multiple contextual levels. The purpose of the study was to identify the risk and protective factors that best differentiate groups along with the outcome variable of interest (being bullied) using Classification and Regression Tree (CART) analysis. The study used the Health Behavior in School-Aged Children (HBSC) data collected from a nationally representative sample of students in grades six through ten during the 2005–2006 school years. This study identified that for adolescents 12 and younger, lower parental support is a critical risk factor associated with bullying and among those 13 to 14 with lower parent support, adolescent with higher academic pressure reported experiencing more bullying. For the older group of adolescents (aged 15 and older), school related factors were identified to increase the risk level of being bullied. There was a critical age (15 years old) for implementing victimization interventions to reduce the damage from being bullied. Service providers working with adolescents aged 14 and less should focus more on family-oriented intervention and those working with adolescents aged 15 and more should offer peer- or school-related interventions.

Keywords: Being bullied, Bullying, Ecological risk factors, CART, Adolescents

1 Introduction

Being bullied is a well-recognized trauma for adolescents. According to the School Crime Supplement to the National Crime Victimization Survey (Robers et al. 2012), 28% of students aged 12 through 18 were victims of bullying at school one or more times in 2009. Sansone and Sansone (2008) reported one out of ten students report being bullied frequently by their peers. Experiencing bullying as an adolescent can influence a child’s perception of his or her safety as well as the overall quality of his or her school experience (Batsche and Knoff 1994; Borg 1998). Bullying has led to serious psychological and social impacts for its victims, even to the point of suicide (Bhutani et al. 2014; Ivarsson, Broberg, and Arvidsson 2005; McCabe et al. 2003; Owusu, Hart, Oliver, and Kang 2011; Schneider et al. 2012; Storch et al. 2003).

Studies on trends in risk and protective factors associated with bullying indicate there is a number of persistent risk factors that increase the probability of being bullied. Individual factors include child emotional, developmental, and behavioral problems (Shetgiri et al. 2012). Family factors include poor parental communication, lack of parental involvement, and parental anger with the child (Espelage, Bosworth, and Simon 2000; Flouri & Buchanan 2003; Shetgiri et al. 2012; Spriggs, Iannotti, Nansel, and Haynie 2001). Finally, poor relationships with classmates and negative peer groups are considered peer or school risk factors (Espelage et al. 2000; Spriggs et al. 2001). In the past few years, scholars have studied factors associated with bullying at multiple contextual levels: individual, family or peer, school, and community (Lee 2011; Lee and Song 2012). Although literature on the perpetration of bullying is increasing, less is known about the multilevel factors associated with who is a victim of bullying. The current study responds to this shortcoming in the previous research. Thus, this study aims to identify the risk and protective factors using Classification and Regression Tree (CART) analysis.

2 Literature Review

According to the ecological risk-factor approach, adolescents can be exposed to risks organized into four levels of their ecology: (1) the individual, (2) the family, (3) extrafamilial contexts, such as schools and peer groups, and (4) the macro system, such as media and public policies (Perkins 2002; Small and Luster 1994). The central idea of this approach suggests that as exposure to risk factors at multiple levels increases, the probability of being bullied increases.

2.1 Age and Being Bullied

At the individual level, age is a crucial variable to explore as a risk and protective factor associated with being bullied. There are distinct differences in how being bullied is experienced by age (Macklem 2003). Some research studies indicate that the percentage of students who are targeted by bullies decreases as age increases (Khoury-Kassabri et al. 2004; Smith, Madsen, and Moody 1999). Kaltiala-Heino et al. (2000) found that high school students (grades 11–12) experienced less victimization than students in junior high schools (grades 7–8). The 2009 School Crime Supplement showed that as the grade level of students increased from 6th through 12th grades, the total percentage of bullied students decreased from 39.4 to 20.4 %. Younger children may be at an increased risk for being bullied because they have not yet acquired the social and assertiveness skills to handle bullying incidents effectively in order to deter further bullying (Smith et al. 1999). The transition from elementary school to middle school or junior high school is a period in which students are more likely to be bullied than other time periods (Orpinas & Horne 2006).

2.2 Family Factors and Being Bullied

Influential factors for being bullied among adolescents are obviously related to their negative family experiences (Orpinas & Morne 2006). Some studies reported that poor parent-children relationships and communications increased the probability of victimization by bullying (Bowers et al. 1994; Moon et al. 2010; Stevens et al. 2002). Jaynes (2008) found that parental involvement was a predictor of being bullied by investigating the relationship between parental involvement and being bullied among 7th to 12th grade students. In the study of family factors associated with children’s resilience to being bullied, scholars found that maternal warmth, sibling warmth, and a positive atmosphere at home reduced the risk of being bullied (Bowes et al. 2010). They reported that warm and supportive family relationships helped to prevent children from the negative outcomes associated with bullying victimization (Bowes et al. 2010).

2.3 Peer Factors and Being Bullied

As students move into adolescence, they spend more time with friends and less time with their families (Larson & Richards 1991). In light of these changes among adolescents, peer factors and an increasing use of technology play significant roles in bullying today. Several studies indicate that the more adolescents use mobile phones, the more susceptible they are to cyberbullying (Cantone et al. 2015; D’Antona et al. 2010; National Children’s Home [NCH] 2005; Raskauskas and Stoltz 2007; Smith et al. 2008). Slonje and Smith (2008) asserted that cyberbullying is much dangerous than most traditional bullying because of the following three reasons: (1) the difficulty of getting away from it; (2) the breadth of potential audience; and (3) the invisibility of those doing the bullying.

2.4 School Factors and Being Bullied

In order to decrease school violence, developing a positive school climate is necessary (Colvin et al. 1998; Khoury-Kassabri et al. 2004; Moon, Karlson, and Kim 2015; Stephens 1994). To the extent that students do not attend school because they perceive themselves to be at high risk of being bullied, school climate, as measured by student and staff experiences with victimization (National School Climate Center 2015), could be a predictor of bullying. Astor, Benbenishty, Zeira, and Vinokur (2002) found that high school students’ decision not to attend school was related to their experiences of victimization by other students and school staff. Nansel et al. (2001) reported that poor relationships with classmates in classrooms pertained to increase the risk of bullying as well as being bullied. Moreover, low academic competence was associated with an increased risk of being bullied as well (Ma et al. 2009; Nansel et al. 2001).

Building upon the previous studies, we hypothesized that family, peer, and school factors would be significant predictors of being bullied among adolescents. Also, we tested another hypothesis that predictors of being bullied would vary by age.

3 Methods

3.1 Data and Participants

Data were collected from a nationally representative sample of students in grades six through ten participating in the Health Behavior in School-Aged Children (HBSC) study during the 2005–2006 school year (Wang et al. 2010). Health Behavior in School-aged Children (HBSC) is a cross-national survey, which is conducted every four year to examine health-related behavior among children and adolescents. The sample was selected through a three-stage stratified design, with census divisions and grades as strata and school districts as primary sampling units. African-American and Hispanic students were oversampled. Students completed an anonymous self-report questionnaire distributed in the classrooms. A total of 9,227 students completed the survey, but 8888 students were included for the study. Students were excluded from the final sample if they failed to respond to multiple items on the questionnaire, particularly if those items indicated whether the student experienced bullying or not.

3.2 Measures

From the larger HBSC study, 22 items were selected based on the ecological theoretical framework and the previous research. The questions for this study addressed: 1) individual factors (1 item; see Table 1); 2) family factors (16 items; see Table 1); 3) peer factors (1 item; see Table 1); and 4) school factors (4 items; see Table 1). Age was selected as the individual factor and calling and texting as the peer factor. Parental support and parental monitoring were used as the family factors. School performance, feelings about school, enjoy with classmate, and academic pressure were selected as the school factors.

Table 1.

Summary of construct items

Variables Response Categories Reliability
(Cronbach’s Alpha)		 Excluded items
Individual Factor Age How old are you? 11=age 11 or younger,
12=age 12, 13=age 13,
14=age 14, 15=age 15,
16=age 16, 17=age 17
Family factors Parental support Q50A-Q50E & Q50H Parent/Guardian: ① Helps me as much as I need, ② Lets me do things I like doing, ③ Is loving, ④ Understands my problems, ⑤ Likes me to make own decisions, ⑧ Makes me feel better when upset 1=About every day,
2=About every week,
3=More than once a
week, 4=About every
month, 5=Rarely or never		 0.82 ⑥ Tries to control things I do, ⑦ Treats me like a baby
Parental monitoring Q48A-Q48E & Q49A-Q49E Mother knows: ① Who your friends are, ② How you spend money, ③ Where you are after school, ④ Where you go at night, ⑤ What you do with free time & Father knows: ① Who your friends are, ② How you spend money, ③ Where you are after school, ④ Where you go at night, ⑤ What you do with free time (1=She/he knows a lot,
2=She/he knows a little,
3=She doesn’t know
anything, 4=Don’t have/
see mother/guardian)		 0.88
Peer factor Calling & texting Q56 How often call/text friends 1=rarely or never, 2=1 or
2 days a week, 3=3 or
4 days a week, 4=5 or
6 days a week, 5=every
day
School factors Teacher’s opinion Q58 Teacher’s opinion of your school performance 1=very good, 2=good,
3=average, 4=below
average
Feelings about school Q59 Present feelings about school 1=I like it a lot, 2=I like
it a bit, 3=I don’t like it
very much, 4=I don’t
like it all
Enjoy classmates Q60A Student in my class: Enjoy being together 1=Strongly agree, 2=Agree,
3=Neither agree nor
disagree, 4=Disagree,
5=Strongly disagree
Academic pressure Q61 Amount of pressure from school work 1=Not at all, 2=A little,
3=Some, 4=A lot
Bullied Q 63A-Q63I How often got bullied: ① Called names, teased, ② Left out of things, ③ Hit, kicked, pushed, ④ Others lied about me ⑤ For my race/color, ⑥ For my religion, ⑦ Made sexual jokes to me ⑧ Using a computer/e-mail, ⑨ Using a cell phone 1=I haven’t been bullied,
2=only once or twice,
3=2 or 3 times a month
a month, 4=about once
a week, 5=Several times a
week→1=not bullied
(NO), 2–5=bullied (YES)
Open in a new tab

Family factors are parental support and parental monitoring. Six items were combined to create a measure of parental support: 1) parent/guardian helps me as much as I need; 2) parent/guardian lets me do things I like doing; 3) parent/guardian is loving; 4) parent/guardian understands my problems; 5) parent/guardian likes me to make my own decisions; and 6) parent/guardian makes me feel better when upset. Responses were measured on a 5-point Likert scale (about every day, about every week, more than once a week, about every month, and rarely or never). We used reverse scores. This means that higher cores would indicate higher support. The Cronbach Alpha for the measure of parental support was .84.

Ten items were combined to create a measure of parental monitoring. Five items asked about mothers and five asked about fathers. These five questions inquiring about parental monitoring were: 1) mother/father knows who your friends are; 2) mother/father knows how you spend money; 3) mother/father knows where you are after school; 4) mother/father knows where you go at night; 5) mother/father knows what you do with free time. Responses were measured on a 4-point Likert scale (she/he knows a lot, she/he knows a little, she/he doesn’t know anything, and I don’t have/see mother/father/guardian). We used reverse scores. This means that higher cores would indicate higher monitoring. The Cronbach Alpha for the measuring of parental monitoring was .88.

School factors are school performance, feelings about school, enjoy with classmate, and academic pressure. Each school factor was measured by a single item, so we did not report the Cronbach Alpha of them.

The variable, being bullied, was based on respondents’ self-report of how often they got bullied. Nine items inquired about the frequency of being bullied: 1) called names and teased; 2) left out of things; 3) hit, kicked, and pushed; 4) others lied about me; 5) for my race/color; 6) for my religion; 7) made sexual jokes to me; 8) using a computer/e-mail; and 9) using a cell phone. Responses for these items were measured on a 5-point Likert scale (I haven’t been bullied, only once or twice, two or three times a month, about once a week, and several times a week). The responses were recoded into two categories, not being bullied and bullied. Nine items were combined into a single score ranged from 0 to 9. The Cronbach Alpha of ‘being bullied’ was .90.

3.3 Analytic Method

We used descriptive statistics to examine the distribution of risk factors among the sample of adolescents and employed t-test and chi-square tests to evaluate bivariate relationships between bullying status and risk/protective factors. Also, Classification and Regression Trees (CART) were used to identify population (audience) segments of adolescents based on risk and protective factors for being bullied. CART analyses were conducted in CART 6.0 software (Salford Systems 2008). CART is one of the data mining algorithms for classification and regression (Breiman, et al. 1984). CART can efficiently handle both continuous and categorical variables in the data set. From an algorithmic point of view, CART has a forward stepwise procedure that adds model terms and a backward procedure for pruning, and it conducts variable selection by only including significant variables in the model. The output of CART models is a hierarchical structure that consists of a series of if-then rules to predict the dependent variable (Moon et al. 2011).

We selected CART to examine complex interactions among multiple risk factors that may not be apparent or may be difficult to interpret in a traditional regression analysis and for its ability to identify and segment homogeneous and possibly high risk subgroups of the population. Based on similar characteristics, these subgroups may benefit from different or tailored intervention strategies. CART generates a multi-dimensional examination of individuals who are members of a subgroup, whereas regression is based on the identification of variables as they relate to the outcome, averaged over all individuals (Swan et al. 2004). Specifically, CART was used to develop a classification and regression tree to stratify the study sample into meaningful homogenous subgroups in relation to a particular target variable. In our case, the dependent variable is bullying status (bullied or not bullied). Predictors include individual, family, peer, and school factors described in the measurement section.

4 Results

4.1 Sample Description

Table 2 shows the percentage of adolescents identified as being bullied by age and compares the students who were identified as ‘being bullied’ to those identified as ‘not being bullied with respect to individual, family, peer, and school risk factors.

Table 2.

Comparing age, individual, family, peer and school risk factors between ‘being bullied group’ and ‘not being bullied group’

N=8888 Not Bullied (n=6241) Bullied (n=2647) t Chi-square
   70.2 % 29.8 %
Individual factor
  Age (%)
    11 or younger (10.7) 62.3 % 37.7 % 133.02***
    12 (21.7) 64.1 % 35.9 %
    13 (20.6) 70.2 % 29.8 %
    14 (18.0) 70.7 % 29.3 %
    15 (16.7) 76.1 % 23.9 %
    16 (10.3) 80.3 % 19.7 %
    17 (0.6) 76.0 % 24.0 %
  Family factors
    Parental support 1.53 (0.44) 1.64 (0.47) −9.89***
    Parental monitoring 1.71 (0.60) 1.76 (0.59) −3.53***
  Peer factor
    Calling and texting 3.34 (1.53) 3.16 (1.53) 5.07***
  School factors
    Teacher’s opinion 2.11 (0.87) 2.22 (0.89) −5.18***
    Feeling about school 2.12 (0.90) 2.28 (0.92) −7.42***
    Enjoy classmates 2.43 (0.99) 2.66 (1.09) −9.63***
    Academic pressure 2.47 (1.02) 2.66 (1.00) −8.33
Open in a new tab
***

p<0.001

4.2 Differences Between Bullied and not Bullied Adolescents

The results of the analyses revealed a significant relationship between being bullied and age. The percentage of students who were not bullied increased gradually until age 16 (60 %–80 %) and decreased at age 17 (76.0 %). Students who were bullied peaked at age 11 or younger then gradually decreased until age 17. The results of the analyses revealed that there was a significant difference between students who were bullied and those who were not bullied by age. The analyses show that students who were bullied have significantly lower levels of parental support and parental monitoring compared to students who were not bullied (t=−9.89, t=−3.53, p<0.001). As for peer factors, bullied students reported making fewer calls and texts to friends (t=5.07, p<0.001). As with school factors, students who were bullied had more negative opinions about school performance (t=−5.18, p<0.001), negative feelings about school (t=−7.42, p<0.001), and enjoyed being with classmates less comparing to students who were not bullied (t=−9.63, p<0.001).

4.3 CART Analysis

The final model contained 15 nodes and had an overall classification accuracy of 59.8 %. The target class was set as ‘being bullied.’ After pruning the tree, 8 variables remained in the model, including age, enjoying being with classmates, parental support, academic pressure, frequency of calling and texting friends, present feelings about school, and teacher’s opinion of school performance. Terminal nodes ranged in size from N=52 to N=2,137. The most important variable for determining ‘being bullied’ was age, which split from the root node (see Table 2). Age split between a value of 14 and 15. Adolescents aged 14 and less were more likely to be bullied (32.6 %) than adolescents aged 15 and more (22.4 %). Figure 1 represents the overall tree structure and the size and position of each node.

Fig. 1.

Fig. 1

Open in a new tab

CART model of predicting ‘being bullied’

4.3.1 Adolescents Aged 15 and More

In adolescents age 15 and older, enjoying spending time with classmates less and poorer feelings about school contributed to higher level of being bullied. Specifically, adolescents age 15 and older who enjoyed being with classmates less were more likely to be bullied (34.3 %) than their counterparts (20.5 %). The risk level of being bullied among adolescents age 15 and older increased much more when they had negative feeling about school (37.1 %: terminal node). No other predictor variables explained ‘being bullied’ in adolescents age 15 and older.

4.3.2 Adolescents age 14 and Younger

For adolescents age 14 and younger, (1) parental support, (2) enjoying being with classmates, and (3) amount of academic pressure were found to be important contributors to being bullied. Adolescents with higher parental support were less likely to be bullied (27.3 %) than adolescents with lower parental support (37.1 %). Parental support had a stronger association with bullying among adolescents age 12 and younger than adolescents over age 12. For adolescents between ages 13 and 14, amount of academic pressure was a significant predictor of being bullied. Adolescents reporting higher academic pressure in this age group were more likely to report being bullied (35.3 %) than their counterparts (24.3 %).

5 Discussion

Over the past few years, experts have suggested that bullying can best be understood through an ecological framework because bullying or being bullied involves risk factors at multiple contextual levels (Lee 2011; Lee and Song 2012). This study confirms existing research findings that there are indeed group differences between those students being bullied and those who are not bullied. From the results of this study it is clear that there were significant correlates at multiple levels distinguishing adolescents who experienced bullying from those who did not. At the bivariate level, adolescents who were bullied had lower levels of parental support and monitoring. The results of the CART model indicate that adolescents who are bullied have less parental support than adolescents who are not bullied. This result supports the finding that family patterns are strongly related to children experiencing bullying and violence (Bowes et al. 2010; Yan et al. 2010). Research shows that poor parental relationships and lower involvement increase the probability of victimization by bullying (Bowers et al. 1994; Stevens et al. 2002; Jaynes 2008). With the close connections in the literature between parental warmth and parental supervision, the results of the CART model provide some indication that parental support is a stronger protective factor against bullying among Korean adolescents than parental monitoring. Although statistically significant at the bivariate level, it should be noted that the practical difference in levels of parental monitoring between students who experienced bullying and those that did not was small, M=1.71 (SD=0.60) not bullied compared to M=1.76 (SD=0.59) bullied. The similarity in levels of parental monitoring between groups may be one reason parental monitoring was not a significant correlate in the CART model.

Concerning peer factors, the study supports prior research suggesting that having poorer quality friendships or no friendships increases the risk of peer victimization. Prior research has found that victims report having fewer friends, having poorer peer relationship quality (Smith, Shu, and Madsen 2001), and experiencing greater peer rejection (Parkhurst and Hopmeyer 1998). Additionally, peer relationships and peer sociability have been found to be important protective factors against victimization for girls (Paul and Cillessen 2003). In our study, victimized adolescents reported calling and texting friends less frequently. Potentially, it raises the possibility that bullying leads adolescents to call and text less frequently. In our study, teacher perception of school performance, negative feelings about school, and disconnectedness with classmates appeared as risk factors for victimization.

The results of this study suggest that individual, family, peer, and school factors are important to consider when thinking about ways to prevent peer victimization. Therefore, multilevel prevention efforts that include strategies to increase parental support, to enhance friendship quality, and to decrease academic pressure while still improving school performance may have the greatest impact on reducing peer victimization.

Classification and regression trees were used to identify bullying-related risk profiles for subgroups of adolescents. Our results demonstrate that complex combinations of bullying-related risk factors differ among subgroups of adolescents. Similar to other studies, we found that age was a significant variable for bullying involvement (Khoury-Kassabri et al. 2004; Smith et al. 1999). Adolescents aged 14 and younger were more likely to be bullied than adolescent aged 15 and older. This result of our study substantiates previous research findings that bullying and victimization decrease with age and grade level (Khoury-Kassabri et al. 2004). In addition, our study found that different risk constellations emerged for adolescents aged 14 and younger compared to adolescents age 15 and older. For the younger group of adolescents (age 14 and younger), lack of parental support, lack of enjoyment being with classmates, and academic pressure were identified as important risk factors. In Orpinas and Horne’s (2006) study, the transition period from elementary school to middle or junior high school was a more vulnerable time for children to experience bullying than any other period. Interestingly, the results of our study identified two distinct risk factor groups for adolescents transitioning to early middle school (typically age 12 and younger) compared to adolescents transitioning to late middle school (typically between 13 and 14 years of age). For adolescents 12 and younger, less parental support is a critical risk factor associated with bullying. Among children 13 to 14 with less parent support, adolescent with higher academic pressure reported experiencing more bullying. For the older group of adolescents (age 15 and older), school related factors that were identified to increase the risk level of being bullied are enjoying being with classmates less and poorer feelings about school.

5.1 Implications for Practice and Research

To reduce the harm from being bullied, focusing and implementing age-modified interventions is necessary. Service providers working with adolescents age 15 and under should focus more on family-oriented interventions and those working with adolescents age 15 and older should offer peer-or school-related interventions to prevent them from being bullied. Furthermore, it is important to note that CART can be a compelling data analysis method for social work researchers to examine complex interactions among multiple risk factors. This data analysis method can be useful to identify and segment homogeneous and possibly high-risk subgroups of the population, based on similar characteristics that may benefit from different or tailored intervention strategies.

5.2 Limitations and Suggestions for Future Studies

As a secondary data analysis, this study is limited to the variables available in the HBSC study. Future research should examine the influence of adolescents’ exposure to media on bullying behavior. Furthermore, this analysis utilizes a cross-sectional data sample. The analysis can effectively identify subgroups of adolescents who experience higher rates of bullying. However, longitudinal data would allow for a stronger understanding of how earlier risk and protective factors influence experiencing bullying for these subgroups. Despite these limitations, this study used a large national probability sample of adolescents to obtain these subgroups. Moreover, as the first CART analysis examining risk factors for being bullied, this study provides important information that can be used to tailor bullying intervention strategies for adolescents.

Acknowledgments

This study was financially supported by Namseoul University. Kristen Seay is the recipient of training fellowships from the National Institute on Drug Abuse (F31DA034442, K. Seay, PI; 5T32DA015035).

References

  1. Astor RA, Benbenishty R, Zeira A, Vinokur A. School climate, observed risky behaviors, and victimization as predictors of high school students’ fear and judgments of school violence as a problem. Health Education and Behavior. 2002;29:716–736. doi: 10.1177/109019802237940. [DOI] [PubMed] [Google Scholar]
  2. Batsche G, Knoff H. Bullies and their victims: understanding a pervasive problem in the schools. School Psychology Review. 1994;23:165–75. [Google Scholar]
  3. Bhutani R, Sudhir PM, Philip M. Teasing experiences, interpersonal sensitivity, self-schema and psychological distress in youth: an exploratory study. Psychological Studies. 2014;59(3):241–251. [Google Scholar]
  4. Borg M. The emotional reactions of school bullies and their victims. Educational Psychology: An International Journal of Experimental Educational Psychology. 1998;18:433–444. [Google Scholar]
  5. Bowers L, Smith PK, Binney V. Perceived family relationships of bullies, victims and bully/victims in middle childhood. Journal of Social and Personal Relationships. 1994;11:215–232. [Google Scholar]
  6. Bowes L, Maughan B, Caspi A, Moffitt TE, Arseneault L. Families promote emotional and behavioural resilience to bullying: evidence of an environmental effect. Journal Of Child Psychology And Psychiatry. 2010;51(7):809–817. doi: 10.1111/j.1469-7610.2010.02216.x. [DOI] [PubMed] [Google Scholar]
  7. Breiman L, Fiedman J, Stone CJ, Olshen RA. Classification and regression trees. Oxford: Taylor and Francis; 1984. [Google Scholar]
  8. Cantone E, Piras AP, Vellante M, Preti A, Daníelsdóttir S, D’Aloja E, Bhugra D. Interventions on bullying and cyberbullying in schools: a systematic review. Clinical Practice and Epidemiology in Mental Health: CP & EMH. 2015;11(M4) Suppl 1:58–76. doi: 10.2174/1745017901511010058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Colvin G, Tobin T, Beard K, Hagan S, Sprague J. The school bully: assessing the problem, developing interventions and future research directions. Journal of Behavioral Education. 1998;8:293–319. [Google Scholar]
  10. D’Antona R, Kevorkian M, Russom A. Sexting, texting, cyberbullying and keeping youth safe online. Journal of Social Science. 2010;6:521–526. [Google Scholar]
  11. Espelage DL, Bosworth K, Simon TR. Examining the social context of bullying behaviors in early adolescence. Journal of Counseling & Development. 2000;78:326–333. [Google Scholar]
  12. Flouri E, Buchanan A. The role of father involvement and mother involvement in adolescents' psychological well-being. British Journal Of Social Work. 2003;33(3):399–406. [Google Scholar]
  13. Ivarsson T, Broberg AG, Arvidsson T. Bullying in adolescence: psychiatric problems in victims and bullies as measured by the youth self-report (YSR) and the depression self-rating scale (DSRS) Nordic Journal of Psychiatry. 2005;59:365–373. doi: 10.1080/08039480500227816. [DOI] [PubMed] [Google Scholar]
  14. Kaltiala-Heino R, Rimpela M, Rantanen P, Rimpela A. Bullying at school: an indicator of adolescents at risk for mental disorders. Journal of Adolescence. 2000;23:661–674. doi: 10.1006/jado.2000.0351. [DOI] [PubMed] [Google Scholar]
  15. Khoury-Kassabri M, Benbenishty R, Astor RA, Zeira A. The contributions of community, family, and school variables to student victimization. American Journal of Community Psychology. 2004;34:187–204. doi: 10.1007/s10464-004-7414-4. [DOI] [PubMed] [Google Scholar]
  16. Larson R, Richards MH. Daily companionship in late childhood and early adolescence-changing developmental contexts. Child Development. 1991;62:284–300. doi: 10.1111/j.1467-8624.1991.tb01531.x. [DOI] [PubMed] [Google Scholar]
  17. Lee C. An ecological systems approach to bullying behaviors among middle school students in the United States. Journal of Interpersonal Violence. 2011;26:1664–1693. doi: 10.1177/0886260510370591. [DOI] [PubMed] [Google Scholar]
  18. Lee C, Song J. Functions of parental involvement and effects of school climate on bullying behaviors among South Korea middle school students. Journal of Interpersonal Violence. 2012;27:2437–2464. doi: 10.1177/0886260511433508. [DOI] [PubMed] [Google Scholar]
  19. Ma L, Phelps E, Lerner JV, Lerner RM. Academic competence for adolescents who bully and who are bullied. Journal of Early Adolescence. 2009;29:862–897. [Google Scholar]
  20. Macklem GL. Bullying and teasing social power in children’s groups. New York: Kluwer Academic/Plenum Publishers; 2003. [Google Scholar]
  21. McCabe R, Antony M, Summerfeldt L, Liss A, Swinson R. Preliminary examination of the relationship between anxiety disorders in adults and self-reported history of teasing or bullying experiences. Cognitive Behaviour Therapy. 2003;32:187–193. doi: 10.1080/16506070310005051. [DOI] [PubMed] [Google Scholar]
  22. Moon SS, Patton J, Rao U. An ecological approach to understanding youth violence: the mediating role of substance Use. Journal of Human Behavior In The Social Environment. 2010;20(7):839–856. doi: 10.1080/10911351003751918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Moon SS, Kang S-Y, Jitpitaklert W, Kim SB. Religious involvement, physical health, psychological distress, and smoking behavior among older adults. Expert Systems with Applications: An International Journal. 2011;39:445–451. [Google Scholar]
  24. Moon SS, Karlson A, Kim Y. Peer victimization and adolescent suicide: the mediating effect of risk behaviors. Child and Adolescent Social Work Journal. 2015;32:257–268. [Google Scholar]
  25. Nansel T, Overpeck M, Pilla RS, Ruan WJ, Simons-Morton B, Scheidt P. Bullying behaviors among U.S. Youth: prevalence and association with psychosocial adjustment. Journal of the American Medical Association. 2001;285:2094–2100. doi: 10.1001/jama.285.16.2094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. National School Climate Center. Measuring school climate. [Retrived October 8, 2015];2015 from http://www.schoolclimate.org/climate/practice.php. [Google Scholar]
  27. NCH. Putting u in the picture: Mobile phone bullying survey 2005. 2005 Retrieved from http://www.filemaker.co.uk/educationcentre/downloads/articles/Mobile_bullying_report.pdf. [Google Scholar]
  28. Orpinas P, Morne AM. Bullying prevention: Creating a positive school climate and developing social competence. Washington: American Psychological Association; 2006. [Google Scholar]
  29. Owusu A, Hart P, Oliver B, Kang M. The association between bullying and psychological health among senior high school students in Ghana, West African. Journal of School Health. 2011;81:231–238. doi: 10.1111/j.1746-1561.2011.00590.x. [DOI] [PubMed] [Google Scholar]
  30. Parkhurst JT, Hopmeyer A. Sociometric popularity and peer-perceived popularity: Two distinct dimensions of peer status. Journal of Early Adolescence. 1998;18:125–144. [Google Scholar]
  31. Paul J, Cillessen T. Dynamics of peer victimization in early adolescence: results from a four-year longitudinal study. Journal of Applied School Psychology. 2003;19:25–44. [Google Scholar]
  32. Perkins DF. An ecological risk-factor examination of suicide ideation and behavior of adolescents. Journal of Adolescent Research. 2002;17:3–26. [Google Scholar]
  33. Raskauskas J, Stoltz AD. Involvement in traditional and electronic bullying among adolescents. Developmental Psychology. 2007;43:564–575. doi: 10.1037/0012-1649.43.3.564. [DOI] [PubMed] [Google Scholar]
  34. Robers S, Zhang J, Truman J, Snyder TD. Indicators of School Crime and Safety: 2011 (NCES 2012-002/NCJ 236021) Washington, D.C.: National Center for Education Statistics, U.S. Department of Education, and Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice; 2012. [Google Scholar]
  35. Salford Systems. CART 6.0 User’s Guide. 2008 Available at http://dist.salford-systems.com/CART6/CART6.pdf. [Google Scholar]
  36. Sansone RA, Sansone LA. Bully victims: psychological and somatic aftermaths. Psychiatry. 2008;5(6):62–64. [PMC free article] [PubMed] [Google Scholar]
  37. Schneider SK, O’Donnell L, Stueve A, Coulter RWS. Cyberbullying, school bullying, and psychological distress: a regional census of high school students. American Journal of Public Health. 2012;102(1):171–177. doi: 10.2105/AJPH.2011.300308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shetgiri R, Lin H, Avila RM, Flores G. Parental characteristics associated with bullying perpetration in US children aged 10 to 17 years. American Journal of Public Health. 2012;102(12):2280–2286. doi: 10.2105/AJPH.2012.300725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Slonje R, Smith PK. Cyberbullying: another main type of bullying? Scandinavian Journal of Psychology. 2008;49(2):147–154. doi: 10.1111/j.1467-9450.2007.00611.x. [DOI] [PubMed] [Google Scholar]
  40. Small SA, Luster T. Adolescent sexual activity: an ecological risk-factor approach. Journal of Marriage and the Family. 1994;56:181–192. [Google Scholar]
  41. Smith PK, Madsen KC, Moody JC. What causes the age decline in reports of being bullied at school? towards a developmental analysis of risks of being bullied. Educational Research. 1999;41:267–285. [Google Scholar]
  42. Smith PK, Shu S, Madsen K. Characteristics of victims of schoolbullying: developmental changes in copingstrategies ands skills. In: Juoven J, Graham S, editors. Peer harassment at school: the plight of the vulnerable and victimised. New York, NY: Guilford Press; 2001. pp. 332–352. [Google Scholar]
  43. Smith PK, Mahdavi J, Carvalho M, Fisher S, Russell S, Tippett N. Cyberbullying: its nature and impact in secondary school pupils. Journal of Child Psychology and Psychiatry. 2008;49:376–385. doi: 10.1111/j.1469-7610.2007.01846.x. [DOI] [PubMed] [Google Scholar]
  44. Spriggs AL, Iannotti RJ, Nansel TR, Haynie DL. Adolescent bullying involvement and perceived family, peer, and school relations: commonalities and differences across race/ethnicity. Journal of Adolescents Health. 2001;41:283–293. doi: 10.1016/j.jadohealth.2007.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Stephens RD. Planning for safer and better schools: School violence prevention and intervention strategies. School Psychology Review. 1994;23:204–215. [Google Scholar]
  46. Stevens V, De Bourdeaudhuij I, Oost PV. Relationship of the family environment to children’s involvement in bully/victim problems at school. Journal of Youth and Adolescence. 2002;31:419–428. [Google Scholar]
  47. Storch E, Brassard M, Masia-Warner C. The relationship of peer victimization to social anxiety and loneliness in adolescence. Child Study Journal. 2003;33:1–18. doi: 10.1016/j.adolescence.2004.03.003. [DOI] [PubMed] [Google Scholar]
  48. Swan GE, Javitz HS, Jack LM, et al. Heterogeneity in 12-month outcome among female and male smokers. Addiction. 2004;99:237–250. doi: 10.1111/j.1360-0443.2003.00629.x. [DOI] [PubMed] [Google Scholar]
  49. Wang J, Iannotti RJ, Luk JW, Nansel TR. Co-occurrence of victimization from five subtypes of bullying: physical, verbal, social exclusion, spreading rumors, and cyber. Journal Of Pediatric Psychology. 2010;35(10):1103–1112. doi: 10.1093/jpepsy/jsq048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Yan FA, Howard DE, Beck KH, Shattuck T, Hallmark-Kerr M. Psychosocial correlates of physical dating violence victimization among Latino early adolescents. Journal of Interpersonal Violence. 2010;25:808–831. doi: 10.1177/0886260509336958. [DOI] [PubMed] [Google Scholar]

RESOURCES