Bullying Victimization Among Chinese Middle School Students: The Role of Family Violence

Abstract

This study used the data from a representative sample to investigate the association between family violence (FV) and child bullying victimization (BV) in Xi’an city, China. Data on social demographic information and the prevalence of BV and FV were collected from a randomly selected sample with 3,175 middle school students aged 15 to 17 by self-administrated questionnaires. Results show that 55.9% and 30.3% of the participants have witnessed intimate partner violence (IPV), 37.7% and 30.8% have been victims of child abuse, and 54.9% and 44.6% have been bullied in a lifetime and in the preceding year, respectively. The lifetime and preceding-year co-occurrence rate of FV and BV are 45% and 30.4%, respectively. Multiple logistic regressions confirm FV as a unique risk factor in predicting both direct and relational BV after controlling for a number of confounding factors. This study suggests that FV experiences should be included in the screening and assessment of risk for child BV.

Keywords

bullying child abuse intimate partner violence family violence Chinese children

Introduction

Children experiencing bullying victimization (BV) is a serious but under-studied social issue in Chinese societies. Bullying may take both direct forms (including physical and verbal bullying) and relational forms (including deliberate social exclusion, isolation through rumor spreading, and relationship manipulation; Björkqvist, 1994; Crick, Casas, & Ku, 1999; Crick & Grotpeter, 1995; Olweus, 1999, 2003). Previous studies have documented that direct bullying and relational bullying are distinct in nature and boys are more likely to experience direct forms of BV, whereas girls tend to be bullied in relational forms (Björkqvist, Osterman, & Kaukiainen, 1991; Crick et al., 1999; Nansel et al., 2001; Seals & Young, 2003).

Among the limited number of studies on the issue of BV in China, various findings on the prevalence of bullying have been found depending on the varied samples and instruments adopted. Recent surveys within Chinese societies reported the varied prevalence of BV occurring at least once ranged from 13.3% to 87% (Zhu & Chan, 2015). These studies provide evidence that a sizable number of Chinese children had been bullied. Multiple factors, including demographic, socioeconomic, and behavioral factors, have been identified as being associated with BV among Chinese children. For example, Chinese victims of bullying tend to be boys (Chen & Yue, 2002; Zhang, Gu, Wang, Wang, & Jones, 2000); be of a young age (Chen & Yue, 2002); be lonely (Cheng et al., 2010); have no friends (Eslea et al., 2004); always have negative feelings, such as worry, sadness, unhappiness, or hopelessness (Hazemba, Siziya, Muula, & Rudatsikira, 2008); have low self-esteem (Huang, Zhou, & Guo, 2005); or even have suicide ideation (Cheng et al., 2010). Chinese children who report some externalizing problems, such as smoking cigarettes, drinking alcohol, and truancy, are also more likely to be bullied by their peers (Hazemba et al., 2008). Besides these commonly studied correlates, there is evidence showing that insecure attachment (Perry, Hodges, & Egan, 2001; Walden & Beran, 2010), antisocial personality disorder (APD; Sourander et al., 2007), borderline personality disorder (BPD; Sansone, Lam, & Wiederman, 2010), and posttraumatic stress disorder (PTSD; Cuevas, Finkelhor, Ormrod, & Turner, 2009), are also potential correlates for BV.

Children experiencing family violence (FV), which is characterized by any behavior within a family that causes direct or indirect physical, psychological, or sexual harm to the child, is also considered to be BV. Direct victimization within FV refers to violence against a child by caregivers who physically, psychologically, or sexually harm or neglect the child, similar to child abuse. Indirect victimization can be regarded as any form of exposure to violence at home, such as witnessing intimate partner violence (IPV; World Health Organization [WHO], 2004). Growing evidence suggests that children who suffer from any kind of victimization in one context may also suffer from victimization in other contexts (Boney-McCoy & Finkelhor, 1996; Finkelhor, Ormrod, & Turner, 2007; Saunders, 2003). Children who experience multiple victimizations have reported suffering from more psychological distress, lower academic grades, and more behavioral problems than children who experience a single form of or no victimization (Holt, Finkelhor, & Kantor, 2007b; Mrug, Loosier, & Windle, 2008).

Studies have shown a substantial overlap between FV and bullying. In a frequently cited study, Baldry (2003) found that bullying was strongly associated with exposure to parental violence. The perpetrators of bullying behavior were 1.8 times more likely to be exposed to IPV than those not exposed; overall, 60.8% of children with exposure to IPV were the perpetrators of bullying behavior and 71% of them were victims. In Dussich and Maekoya’s (2007) study, it was found that 71.6% of physically abused children reported experiencing bullying. Holt, Finkelhor, and Kantor (2007a) reported a relatively median overlap rate, with 34.6% of bullying victims and 24.1% of the perpetrators experiencing child maltreatment. Reviews of the rates of co-occurrence of FV and bullying ranged from 14.5% to 76.67% (Zhu, 2012). Different explanations have been provided for the associations between FV and BV. Some scholars pointed out that exposure to parental violence reduced the capacity of a child to be assertive, which could lead to the child being victimized at school (Baldry, 2003). Wilczenski et al. (1997) claimed that victimization at home caused the child to “learn” this behavior and signal to others that weakness had been internalized as a stable personal trait. In addition, Finkelhor et al. came up with a developmental victimology framework, which specifically emphasized the way in which experiencing some forms of victimization may create vulnerability to other forms of victimization (Finkelhor & Dziuba-Leatherman, 1994; Finkelhor & Kendall-Tackett, 1997).

Existing studies reveal that the association between FV and bullying experiences appears in at least three forms. First, past FV experience is a risk factor for later bullying perpetration (Strassberg, Dodge, & Pettit, 1994), victimization (Schwartz, Dodge, Pettit, & Bates, 2000), or both bullying and victimization (Schwartz, Dodge, Pettit, & Bates, 1997). Second, FV correlates with bullying perpetration (Dussich & Maekoya, 2007; Fitzpatrick, Dulin, & Piko, 2007), victimization (Mohr, 2006; Shields & Cicchetti, 2001), or both (Baldry, 2003). Third, bullying can be a risk factor for future FV perpetration (Connolly, Pepler, Craig, & Taradash, 2000). However, these studies did not control for the confounding factors or investigate the unique role of FV in children’s experiences of BV. In addition, these studies were mainly based on clinical samples, small community samples, or convenience sampling; this greatly limits their generalizability in regard to the wider population.

At present, little is known about the association between FV and children’s BV. Given the significantly heightened maladjustment of the co-occurring FV and BV, understanding this association is particularly important. In addition, relatively few studies in Western societies and no studies within Chinese societies have been conducted to systematically explore the relationship between FV and BV. Based on a representative population of Chinese middle school students aged 15 to 17 and residing in Xi’an city, the current study aims to estimate the prevalence of both BV and victimization of FV among middle school students in mainland China, and to investigate the associations between FV and BV after controlling for a number of confounding factors. This study represents an initial step in investigating the unique role of FV in BV by using a cohort of representative samples of school students in China.

Method

Design and Participants

Complete profiles of 3,175 children aged 15 to 17 were successfully collected in Xi’an city in Shanxi Province. In recent years, Xi’an city has emerged as one of the leading cities experiencing dramatic social and economic development in western China. Previous studies have been conducted in more developed coastal and middle parts of China, such as Beijing (Hazemba et al., 2008), Tianjin (Chen & Yue, 2002), Hangzhou, and Wuhan city (Cheng et al., 2010), but these studies may be biased because they only represent limited parts of China. Significant disparities in economic and social development exist in the three regions of mainland China (i.e., eastern, middle, and western China), and cities in western China still lag behind in terms of infrastructure, environment, investment, economic adjustment, human capital, and welfare of the population compared with other big cities in eastern China (Lai, 2002). Xi’an is the most developed city in all aspects when compared with other cities in western China, and so the data obtained in Xi’an can add to the representative data regarding child population in western China.

The data collection was conducted from September 2009 to June 2010 in Xi’an city. We adopted a three-stage stratified sampling design to identify eligible participants. In the first stage, we randomly chose two urban districts and one rural county from all the administrative units in Xi’an. In the second stage, six schools, including ordinary secondary schools, high schools, and technical high schools, were randomly selected from each of the three administrative regions. In the next step, a list of all children aged 15 to 17 years in the sampled 18 schools was obtained from school officials and those students were invited to complete a self-administrated questionnaire on regular school days. In total, 3,175 students aged 15 to 17 in the sampled schools were successfully recruited, representing a respondent rate of 94.4%.

Participation in this study was totally voluntary. All the participants signed a written consent form before taking part in the study and strict confidentiality of information was guaranteed. They were briefed by trained interviewers in the classrooms. The questionnaire was self-administrated, and required around 30 min to complete. Participants were informed that they could refuse to answer any question and terminate the interview at any time. Each completed questionnaire was sealed in an envelope by the participant to ensure privacy. All procedures were approved by the Institutional Review Board of the University of Hong Kong/Hospital Authority of Hong Kong West Cluster. We trained all interviewers in ethical issues. An information card containing the details of social services for violence prevention in Xi’an city was given to the participants, in case a need for help arose.

Measurements

A standardized demographic questionnaire was used to examine demographic and socioeconomic characteristics of the participants and to estimate their associations with BV. The questionnaire included items asking for information about age, gender, ethnicity, grades, parents’ education level, parents’ marital status, whether the participant had an unemployed father or mother, how many siblings the participant had at home, whether the participant’s family was receiving social security, and family income. In addition, school information, such as school location and school type, were provided by the local research team in Xi’an and were obtained at the sampling stage.

BV

We used the peer and sibling module of the Juvenile Victimization Questionnaire (JVQ; Finkelhor, Hamby, Ormrod, & Turner, 2005) to measure the prevalence of direct BV in terms of both lifetime and the preceding year. This module contained five items: gang or group assault, peer or sibling assault, nonsexual genital assault, threat of force, and verbal bullying. The peer and sibling module of JVQ in this study demonstrated a satisfactory internal consistency reliability (Cronbach’s α = .782). Relational BV was assessed by the modified Relational Aggression Scale (RAS; Crick & Grotpeter, 1995) and appeared in the questionnaire as the following five items: rumor spreading, social exclusion, threatened to obey others’ wills, ignorance, and threatened to hurt and take things. In this study, the Chinese version of RAS demonstrated a satisfactory internal consistency reliability (Cronbach’s α = .836).

Victimization of FV

This study included both a direct form (child abuse) and an indirect form (witnessing IPV between parents) of FV experiences. We used the child maltreatment module of JVQ to assess the prevalence of child abuse. This module contained assessments of four different types of child abuse (physical abuse, psychological abuse, neglect, and custodial interference). Participants used a 5-point Likert-type scale to rate the frequency of abusive experiences. The five response categories (from 0 to 4) are defined as follows: 0 = once, 1 = twice, 2 = three times, 3 = four times, 4 = five times or more. Participants experiencing at least one instance of abuse in either the preceding year or in their lifetime were categorized as victims of child abuse. The modified sexual victimization module had a Cronbach’s alpha of .667, indicating good internal consistency. Four items modified from the Abuse Assessment Screen (AAS; Soeken, McFarlane, Parker, & Lominack, 1998) were used to assess whether the participants had witnessed father-to-mother or mother-to-father physical assault (e.g., pushing, twisting, slapping, beating, and kicking) or psychological aggression (e.g., insults or swearing, shouting or yelling, said something to spite the other, destroyed something belonging to the other, threats, and being ignored). Respondents who reported any of these physical or psychological assaults between their parents were coded as having witnessed IPV. In this study, the self-constructed scale for witnessing IPV demonstrated adequate internal consistency reliability (Cronbach’s α = .674).

Adolescent Attachment Questionnaire (AAQ; West, Rose, Spreng, Sheldon-Keller, & Adam, 1998) consists of three subscales of a child’s attachment to his or her parents: angry distress, availability, and goal-corrected partnership, with 5-point Likert-type responses ranging from strongly disagree to strongly agree. In the present study, the alpha coefficient of AAQ is .66, showing acceptable internal consistency reliability.

The University of California, Los Angeles, PTSD Reaction Index for DSM-IV (UCLA PTSD-RI, Revision 1, Adolescent version; Pynoos, Rodriguez, & Steinberg, 1998) is a 22-item self-report instrument used to report the presence of any type of traumatic event and the frequency of Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) PTSD symptoms by using a 5-point Likert-type scale from 0 (none of the time) to 4 (most of all the time) to rate PTSD symptoms. Sample questions for each criterion include “Upsetting thoughts, pictures, or sounds of what happened come into my mind when I do not want them to” (intrusion); “I try not to talk about, think about, or have feelings about what happened” (avoidance); and “I have trouble going to sleep or I wake up often during the night” (arousal). A total score was calculated as the mean response across all items. This scale demonstrates strong internal consistency reliability in the present study (Cronbach’s α = .95).

The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) was adopted to assess the extent of participants’ self-esteem. The RSES consists of five positively worded items and five negatively worded items. The scores for the 10 items were summarized, and the higher the score, the higher the respondent’s self-esteem. The internal consistency Cronbach’s alpha of the self-esteem scale in this study was .73.

Beck Depression Inventory–II (BDI-II) was applied to assess the intensity of depression among children (Beck, Steer, & Brown, 1996). It contained 21 statements that are structured on a 4-point scale ranging from 0 (symptoms not present) to 3 (symptoms strongly present), with resulting summary scores ranging from 0 to 63. Higher total scores indicate more severe depressive symptoms. In this study, this scale demonstrates strong internal consistency reliability (Cronbach’s α = .815).

Personal and Relationships Profile (PRP) is a multi-scale and validated instrument measuring individual and relationship factors along 21 subscale items and is widely used in clinical screening and research on FV (Chan, Straus, Brownridge, Tiwari, & Leung, 2008; Straus, Hamby, Boney-McCoy, & Sugarman, 1999). Antisocial Personality and Borderline Personality subscales were adopted to assess the extent of participants’ APD and BPD. In this study, the alpha coefficient of the Antisocial Personality Subscale and Borderline Personality Subscale of PRP is .777 and .815, respectively.

Data Analysis

The first stage consisted of descriptive analyses, where we assessed the gender differences in sociodemographic and the prevalence of violence by the chi-square test and t test. The co-occurrence rate was computed and tested. Logistic regression is an appropriate technique for predicting a dichotomous dependent variable from a set of independent variables. A structured three-phase logistic regression was performed to examine the associations between child BV, sociodemographic, individual characteristics, and victimization in FV. In Phase 1, logistic regressions regarding direct and relational BV in both the lifetime and preceding year was performed with all sociodemographic variables (i.e., gender, age, parents’ education level) while adjusting the remaining demographic factors. In Phase 2, BV was regressed in regard to individual characteristic factors (i.e., gambling, depression, PTSD) in separate logistic regression models by controlling for all the demographic factors and the remaining individual characteristics. In phase 3, BV was regressed regarding FV victimization by controlling for all the variables listed in Phase 1 and in Phase 2.

Multicollinearity was checked among all independent variables to make sure they were not highly correlated in a multiple regression model. In this study, all the variance inflation factors (VIFs) were less than 2.5, which is lower than the guideline values of VIF. Therefore, no multicollinearity problem was identified. In all regression models, an odds ratio (OR) greater than one indicates that an increase in the independent variable is associated with an increase in the odds of the dependent variable; the reverse is true if the OR is below one. Statistical significance was tested using p values (5% level) and 95% confidence intervals (CIs). The Hosmer and Lemeshow (2000) (H–L) test was used to calculate the model fit. Missing data were handled with listwise deletion in all analyses. SPSS Version 17 was used for the statistical analysis.

Results

Prevalence

Table 1 showed that the total prevalence rates of witnessing IPV were 55.9% and 30.3%, 37.7% and 30.8% of participants reported being abused in the lifetime and the preceding year, respectively. Female participants were significantly more likely to be exposed to IPV, whereas males were more likely to be victims of child abuse in both time frames; 54.9% and 44.6% of the respondents reported having been bullied in their lifetime and in the year preceding the study, respectively. Relational victimization (44.1%) occurred slightly more than direct BV (43.9%) for lifetime prevalence, whereas the preceding-year prevalence of direct BV (35.1%) is a little bit higher than that of relational victimization (34.5%).

Table 1.

Lifetime and Preceding-Year Prevalence of FV and BV by Gender.

	Lifetime Prevalence—In Total				Annual Prevalence—In Total
	Prevalence (%)	Boys (%)	Girls (%)	t Test	Prevalence (%)	Boys (%)	Girls (%)	t Test
Witnessing IPV	55.9	52.6	58.8	.000 ***	30.3	29.4	31.0	.333
CAN	37.7	40.9	34.9	.001 ***	30.8	33.4	28.5	.004 **
Direct BV	43.9	50.1	38.5	.000 ***	35.1	41.9	29.1	.000 ***
Relational BV	44.1	43.8	44.4	.772	34.5	36.5	32.8	.030 *
BV (in total)	54.9	57.3	52.8	.013 *	44.6	49.6	40.2	.000 ***

Note. IPV = intimate partner violence; CAN = child abuse and neglect; BV = bullying victimization.

p < .05. **p < .01. ***p < .001.

Table 2 showed the lifetime and preceding-year prevalence of the co-occurrence of FV and BV. In the current sample, a vast majority of children reported either being victims of bullying or of FV. A small proportion of child participants reported that they had been victims of bullying only (9.6%) and FV only (21.2%) in their lifetime, whereas 14.2% reported being bullied only and 17.3% reported FV only in the past 12 months. Among the participants who came from families characterized by IPV or child abuse, 68.1% and 63.6% had been bullied by their peers in their lifetime and the preceding year, respectively. Conversely, among the child victims of bullying, 82.4% (lifetime) and 68.1% (preceding year) also reported being victims of FV in the forms of physical and psychological aggression. The huge overlap between bullying and FV victimization led to a relatively high lifetime and preceding year co-occurrence rate of around 45% and 30.4%, respectively.

Table 2.

Co-Occurrence of FV and BV.

	Total		No Victimization		Being Bullied Only^a		FV Only		Co-Occurrence of FV and BV		% of BV Among FV (%)	% of FV Among BV (%)
	n	%	n	%	n	%	n	%	n	%	% of BV Among FV (%)	% of FV Among BV (%)
Lifetime prevalence	3,151	99.2	753	23.9	304	9.6	668	21.2	1,426	45.3	68.1	82.4
Preceding-year prevalence	3,147	99.1	1,198	38.1	447	14.2	546	17.3	956	30.4	63.6	68.1

Note. FV = family violence; BV = bullying victimization.

BV included victimization of direct bullying and relational bullying.

Association Between FV and BV

Table 3 presents the association between FV victimization and BV after controlling for all demographic variables listed in Phase 1 and individual characteristics in Phase 2. The results show that lifetime FV victimization contributed the highest risk of lifetime direct BV (adjusted odds ratio [aOR] = 4.15; 95% CI = [3.18, 5.412]) and relational victimization (aOR = 3.89; 95% CI = [2.982, 5.072]), and preceding-year FV victimization also significantly increased the odds of preceding-year direct BV (aOR = 3.5; 95% CI= [2.617, 4.68]) and relational victimization (aOR = 3.36; 95% CI = [2.503, 4.512]). The model fit for the prediction of both direct and relational BV was significantly increased in Phase 3 by including the sociodemographic and individual variables, and FV victimization. Specifically, the Nagelkerke R² (e.g., 26.6% and 24.2% for direct BV; 29.5% and 25.6% for relational BV) implied a moderately strong relationship between the predictors and the lifetime and preceding-year prevalence for both direct and relational BV. The results of the H–L goodness-of-fit test in Phase 3 were greater than the results reported for each model in Phases 1 and 2.

Table 3.

Structured Multiphase Logistic Regression Analysis of the Association Between FV and BV.

Variable	Adjusted OR[95% CI]Lifetime (Direct BV)	Adjusted OR[95% CI]Lifetime (Relational BV)	Adjusted OR[95% CI]Preceding Year (Direct BV)	Adjusted OR[95% CI]Preceding Year (Relational BV)
Phase 1	(Nagelkerke R² = 8.8%, p^HL = .421)	(Nagelkerke R² = 5.7%, p^HL = .023)	(Nagelkerke R² = 12.2%, p^HL = .113)	(Nagelkerke R² = 7.4%, p^HL = .06)
Age (M, SD)	0.9 [0.801, 1.01]	1.05 [0.935, 1.177]	0.89 [0.783, 1.003]	1.00 [0.882, 1.127]
Gender
Male	1.68*** [1.408, 2.015]	1.03 [0.861, 1.23]	2.01*** [1.665, 2.432]	1.33** [1.099, 1.602]
Female	1.00	1.00	1.00	1.00
Education (father)
Form 3 or below	1.72** [1.142, 2.591]	1.26 [0.852, 1.868]	1.93** [1.221, 3.044]	1.54 [0.997, 2.373]
Form 4-7	1.48* [1.009, 2.169]	1.13 [0.788, 1.633]	1.7* [1.106, 2.616]	1.31 [0.875, 1.975]
Tertiary or above	1.00	1.00	1.00	1.00
Education (mother)
Form 3 or below	0.98 [0.64, 1.506]	1.07 [0.71, 1.626]	0.94 [0.583, 1.501]	0.99 [0.626, 1.559]
Form 4-7	0.83 [0.553, 1.241]	0.99 [0.672, 1.461]	0.84 [0.537, 1.322]	0.9 [0.585, 1.385]
Tertiary or above	1.00	1.00	1.00	1.00
No. of brothers or sisters
0	0.77* [0.617, 0.957]	0.87 [0.702, 1.083]	0.71** [0.563, 0.902]	0.79* [0.624, 0.995]
1 or more	1.00	1.00	1.00	1.00
Parents’ marital status
Divorced/separated	0.99 [0.279, 3.494]	2.45 [0.592, 10.105]	0.56 [0.141, 2.249]	2.12 [0.557, 8.052]
Widow	0.88 [0.526, 1.48]	0.79 [0.472, 1.322]	0.74 [0.418, 1.311]	0.64 [0.358, 1.15]
Married/cohabited	1.00	1.00	1.00	1.00
Unemployed (father)	1.09 [0.823, 1.436]	1.42* [1.075, 1.882]	1.11 [0.832, 1.476]	1.65*** [1.244, 2.192]
Unemployed (mother)	1.11 [0.805, 1.539]	1.1 [0.792, 1.526]	1.09 [0.78, 1.521]	1.11 [0.789, 1.55]
Receiving social security (family)	1.23 [0.915, 1.643]	0.96 [0.713, 1.288]	1.32 [0.977, 1.791]	1.2 [0.886, 1.636]
Family income (per month)
No income	1.03 [0.76, 1.394]	1.01 [0.744, 1.362]	0.91 [0.662, 1.251]	0.83 [0.606, 1.143]
Low income	0.92 [0.692, 1.225]	0.96 [0.725, 1.276]	0.81 [0.599, 1.099]	0.81 [0.598, 1.088]
Low-mid-level	0.86 [0.652, 1.135]	0.95 [0.728, 1.249]	0.82 [0.61, 1.108]	0.86 [0.643, 1.151]
Mid-high	1.00	1.00	1.00	1.00
School district
Urban	0.65** [0.478, 0.883]	0.73* [0.532, 0.989]	0.58*** [0.423, 0.797]	0.64** [0.461, 0.879]
Rural	1.00	1.00	1.00	1.00
School type
Public key school	1.39* [1.033, 1.87]	1.83*** [1.355, 2.463]	1.13 [0.828, 1.55]	1.57** [1.139, 2.158]
Public ordinary school	1.4* [1.023, 1.906]	2.14*** [1.564, 2.94]	1.29 [0.938, 1.772]	1.65** [1.189, 2.278]
Private school	2.08*** [1.446, 3.005]	2.98*** [2.062, 4.304]	2.08*** [1.433, 3.021]	2.47*** [1.691, 3.596]
Vocational school	1.00	1.00	1.00	1.00
Phase 2	(Nagelkerke R² = 17.2%, p^HL = .547)	(Nagelkerke R² = 23.6%, p^HL = .349)	(Nagelkerke R² = 12.2%, p^HL = .907)	(Nagelkerke R² = 20.2%, p^HL = .173)
Gambling	0.87 [0.576, 1.309]	0.96 [0.63, 1.477]	1.13 [0.745, 1.711]	1.00 [0.649, 1.552]
Smoking	1.55* [1.024, 2.337]	1.19 [0.778, 1.835]	1.37 [0.902, 2.077]	1.28 [0.826, 1.971]
Alcohol abuse	1.15 [0.776, 1.698]	1.04 [0.689, 1.562]	1.01 [0.673, 1.506]	0.93 [0.612, 1.411]
Drug abuse	0.4 [0.049, 3.238]	0.05* [0.004, 0.684]	0.74 [0.092, 5.954]	0.29 [0.02, 4.043]
Self-esteem (M)	0.93 [0.669, 1.283]	1.55* [1.101, 2.174]	0.94 (0.661, 1.328)	1.38 (0.969, 1.979)
Depression (M)	1.04*** [1.019, 1.052]	1.06*** [1.043, 1.079]	1.04*** [1.023, 1.057]	1.05*** [1.035, 1.07]
APD (M)	0.98 [0.702, 1.373]	0.81 [0.57, 1.141]	1.11 [0.777, 1.578]	1.13 [0.791, 1.622]
BPD (M)	1.36* [1.018, 1.824]	1.6** [1.178, 2.164]	1.15 [0.843, 1.56]	1.47* [1.073, 2.025]
PTSD (M)	1.55*** [1.294, 1.856]	1.99*** [1.642, 2.402]	1.39*** [1.152, 1.683]	1.63*** [1.345, 1.978]
Attachment (M)	0.78 [0.584, 1.036]	0.94 [0.697, 1.262]	0.79 [0.585, 1.074]	0.85 [0.626, 1.167]
Phase 3	(Nagelkerke R² = 26.6%, p^HL = .79)	(Nagelkerke R² = 29.5%, p^HL = .151)	(Nagelkerke R² = 24.2%, p^HL = .5151)	(Nagelkerke R² = 25.6%, p^HL = .709)
FV (ever)	4.15*** [3.18, 5.412]	3.89*** [2.982, 5.072]	3.5*** [2.617, 4.68]	3.36*** [2.503, 4.512]
FV (preceding year)	3.17*** [2.532, 3.968]	2.66*** [2.122, 3.334]	3.61*** [2.826, 4.605]	3.09*** [2.429, 3.941]

Note. OR = odds ratio; CI = confidence interval; BV = bullying victimization; APD = antisocial personality disorder; BPD = borderline personality disorder; FV = family violence; PTSD = posttraumatic stress disorder.

p < .05. **p < .01. ***p < .001, statistically significant ORs.

Discussion

Using a large and representative sample of school students in Xi’an city of China, this study has generated reliable estimates of the prevalence of BV and FV experiences, and provided evidence for the unique role of FV in the increase in the risk of child involving in BV. Consistent with past research in Chinese communities (Chan, 2011; Chan, Brownridge, Yan, Fong, & Tiwari, 2011), our findings showed both BV and FV are prevalent in Xi’an, with 55.9% and 30.3% of Chinese children aged 15 to 17 witnessing IPV, whereas 37.7% and 30.8% of children from Xi’an schools have been victims of parental child abuse in both their lifetime and the preceding year, respectively. The findings also demonstrated that child experiencing BV was not a rare phenomenon in China; almost one in every two school students aged 15 to 17 in this study had experienced some form of BV in their lifetime, and more than 40% of the children in the sample had been victims of bullying in the preceding year. These values commensurate with previous cross-national comparative studies, which reported a low of 6% to 20% to a high of 67% prevalence of BV (Delfabbro, Winefield, & Metzer, 2006; Due & Holstein, 2008; Kim, Koh, & Leventhal, 2004; Nansel et al., 2001; Undheim & Sund, 2010).

Significant gender differences appeared in children reporting FV and BV. In line with some Western research (Briere & Elliott, 2003; Gershoff, 2002), this study found that girls are more likely to witness IPV than boys, whereas boys were more likely to be abused by parents. Consistent with previous studies (Björkqvist et al., 1991; Crick et al., 1999; Nansel et al., 2001; Seals & Young, 2003), boys were found to be at a higher risk of being bullied. These significant gender differences may be the result of generally stricter parental supervision over girls than boys in Chinese societies (Chan, 2014; Sun, Li, Ji, Lin, & Semaana, 2008). Girls are always required to stay at home rather than going out or playing outside. Because girls spend more time at home, they are less likely to be victims of violence outside the family and more likely to witness other types of FV happening at home.

The co-occurrence rates of FV victimization and BV in the present study were 45.3% and 30.4% in the lifetime and preceding year, respectively. These rates are comparable with those reported in past research on the co-existence of victimization in FV and bullying for children (Baldry, 2003; Dussich & Maekoya, 2007; Holt et al., 2007a). The alarming co-occurrence rate and the results of structured logistic regressions provide supportive evidence for the notion that victimization of FV is the strongest correlate with BV, after controlling for other covariates. This finding is in keeping with earlier findings that children exposed to IPV and those who were victims of child abuse were at a higher risk of being bullied by peers (Bauer et al., 2006; Shields & Cicchetti, 2001). This multi-finality of the linkage between dual victimization offers implications for future research to examine more carefully, especially in regard to the processes and mechanisms of the co-occurring victimizations. Understanding why some victims of FV become bullies, whereas others become victims of bullying, would be especially valuable. Furthermore, the co-occurrence of FV and BV was evident within the same time frame, indicating that pervasive interpersonal deficits are quickly represented and manifested in children in their new social groups.

There are several limitations that should be noted in this study. Despite the present study showing significant associations between FV and BV, it was limited in its ability to establish causal relationship between the variables, due to its cross-sectional design. Second, the reliance on self-reported data may be subject to social desirability and response biases, which might lead to an underestimate of the prevalence of FV and BV. Third, the present study only targeted children aged 15 to 17 as participants; replication studies in the future should include other age groups for comparison, and delineate a clearer picture for tracing BV occurrence along with ages. Fourth, the present study examined the victims of bullying only, and did not make the comparison between victims and bully victims. In light of the more severe consequences for bully victims consistently documented in previous research (e.g., Georgiou & Stavrinides, 2008; Solberg, Olweus, & Endresen, 2007), further differentiation of bully victims from general victims, identification of their specific correlates, and the implementation of relevant social services for each group are important. Finally, the correlates in the present research are mainly related to individual and family factors. Future research should include more school- and family-level correlates, and extend the examinations to more broad-level risk factors.

Despite these limitations, this study provides preliminary evidence that FV is a salient risk factor for child BV. The results have important implications for future research and practice. Given that bullying victims with FV histories suffer from accumulated dramatization and are more likely to suffer from more severe emotional and behavioral problems (Duncan, 1999a; Finkelhor et al., 2007), this research highlights the importance of screening and assessing FV experiences when treating bullying victims. Moreover, as recommended by some researchers (Lynch & Ciccheti, 1998), contexts characterized by adversity should be taken into account in universal screening for BV. This principle should not be limited to child abuse or witnessing IPV only, but should be extended to cover all forms of violence against children in multiple contexts. Interdisciplinary collaborations between health professionals, child protection agencies, and services for FV working on different types of violence are needed to provide integrated programs to better address the co-existence of multiple forms of victimization accounted for by a single child.

Conclusion

Using a large and representative sample of schoolchildren, this study provides reliable estimates of the prevalence of child experiencing BV and FV in China. The present study also confirms the strong associations between FV and the likelihood of an increasing risk of child BV.

Children’s involvement in BV has surprisingly incomplete epidemiological information, yet it attracts inadequate public attention in mainland China. This research points out that more evidence-based research is needed on the magnitude and impacts of the problem, as well as its costs to wider society. It also calls for the promotion of interdisciplinary collaborations among local policy makers, researchers, and social work practitioners, and the development of an integrative screening tool and comprehensive interventions and prevention programs to end BV and FV. Effective, sustainable, and family-based interventions and services may help curb dual victimizations.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Beijing Social Science Fund (Grant number: 15SHB025).

Author Biographies

Yuhong Zhu is an assistant professor of social work at the School of Sociology and Population Studies, Renmin University of China, China.

Ko Ling Chan is an associate professor at Department of Social Work & Social Administration, the University of Hong Kong, Hong Kong; School of Social Welfare, Yonsei University, Republic of Korea.

Jinsong Chen is a professor of social work at the School of Sociology and Population Studies, Renmin University of China, China.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Baldry

A. C.

(2003). Bullying in schools and exposure to domestic violence. Child Abuse & Neglect, 27, 713-732.

Bauer

N. S.

Herrenkohl

T. I.

Lozano

Rivara

F. P.

Hill

K. G.

Hawkins

J. D.

(2006). Childhood bullying involvement and exposure to intimate partner violence. Pediatrics, 118, 235-242.

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Manual for Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation.

Björkqvist

(1994). Sex differences in physical, verbal, and indirect aggression: A review of recent research. Sex Roles, 30, 177-188.

Björkqvist

Osterman

Kaukiainen

(1991). The development of direct and indirect aggressive strategies in males and females. In Bjorkqvist

Niemela

(Eds.), Of mice and women: Aspects of female aggression (pp. 51-64). New York, NY: Academic Press.

Boney-McCoy

Finkelhor

(1996). Is youth victimization related to trauma symptoms and depression after controlling for prior symptoms and family relationships? A longitudinal, prospective study. Journal of Consulting and Clinical Psychology, 64, 1406-1416.

Briere

Elliott

(2003). Prevalence and psychological sequelae of self-reported childhood physical and sexual abuse in a general population of sample of men and women. Child Abuse & Neglect, 27, 1205-1222.

Chan

K. L.

(2011). Children exposed to child maltreatment and intimate partner violence: A study of co-occurrence among Hong Kong Chinese families. Child Abuse & Neglect, 35, 532-542.

10.

Chan

K. L.

(2014). Child victims and poly-victims in China: Are they more at-risk of family violence? Child Abuse & Neglect, 38, 1832-1839.

11.

Chan

K. L.

Brownridge

D. A.

Yan

Fong

D. Y.

Tiwari

(2011). Child maltreatment polyvictimization: Rates and short-term effects on adjustment in a representative Hong Kong sample. Psychology of Violence, 1, 4-15.

12.

Chan

K. L.

Straus

M. A.

Brownridge

D. A.

Tiwari

Leung

W. C.

(2008). Prevalence of dating partner violence and suicidal ideation among male and female university students worldwide. Journal of Midwifery & Women’s Health, 53, 529-537.

13.

Chen

S. P.

Yue

G. A.

(2002). A survey of bully behavior in the school campus. Journal of Psychological Science, 25, 355-356. (In Chinese)

14.

Cheng

Y. L.

Newman

I. M.

Mbulo

Chai

Chen

Shell

D. F.

(2010). Being bullied and psychosocial adjustment among middle school students in China. Journal of School Health, 80, 193-199.

15.

Connolly

Pepler

Craig

Taradash

(2000). Dating experiences of bullies in early adolescence. Child Maltreatment, 5, 299-310.

16.

Crick

N. R.

Casas

J. F.

H. C.

(1999). Relational and physical forms of peer victimization in preschool. Developmental Psychology, 35, 376-385.

17.

Crick

N. R.

Grotpeter

J. K.

(1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66, 710-722.

18.

Cuevas

C. A.

Finkelhor

Ormrod

Turner

(2009). Psychiatric diagnosis as a risk marker for victimization in a national sample of children. Journal of Interpersonal Violence, 24, 636-652.

19.

Delfabbro

Winefield

Metzer

(2006). Peer and teacher bullying/victimization of South Australian secondary school students: Prevalence and psychosocial profiles. British Journal of Educational Psychology, 76, 71-90.

20.

Due

Holstein

B. E.

(2008). Bullying victimization among 13 to 15 year old school children: Results from two comparative studies in 66 countries and regions. International Journal of Adolescent Medicine and Health, 20, 209-221.

21.

Duncan

R. D.

(1999). Maltreatment by parents and peers: The relationship between child abuse, bully victimization, and psychological distress. Child Maltreatment, 4(1), 45-55.

22.

Dussich

J. P. J.

Maekoya

(2007). Physical child harm and bullying-related behaviors: A comparative study in Japan, South Africa, and the United States. International Journal of Offender Therapy and Comparative Criminology, 51, 495-509.

23.

Eslea

Menesini

Morita

O’Moore

Mora-Merchan

J. A.

Pereira

Smith

P. K.

(2004). Friendship and loneliness among bullies and victims: Data from seven countries. Aggressive Behavior, 30, 71-83.

24.

Finkelhor

Dziuba-Leatherman

(1994). Children as victims of violence: A national survey. Pediatrics, 94, 413-420.

25.

Finkelhor

Hamby

S. L.

Ormrod

R. K.

Turner

H. A.

(2005). The Juvenile Victimization Questionnaire: Reliability, validity, and national norms. Child Abuse & Neglect, 29, 383-412.

26.

Finkelhor

Kendall-Tackett

(1997). A developmental perspective on the childhood impact of crime, abuse, and violent victimization. In Ciccheti

Toth

(Eds.), Developmental perspectives on trauma: Theory, research, and intervention (pp. 1-32). New York, NY: University of Rochester Press.

27.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2007). Re-victimization patterns in a national longitudinal sample of children and youth. Child Abuse & Neglect, 31, 479-502.

28.

Fitzpatrick

K. M.

Dulin

A. J.

Piko

B. F.

(2007). Not just pushing and shoving: Bullying among African-American adolescents. Journal of School Health, 77, 16-22.

29.

Georgiou

S. N.

Stavrinides

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

30.

Gershoff

E. T.

(2002). Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin, 128, 539-579.

31.

Hazemba

Siziya

Muula

A. S.

Rudatsikira

(2008). Prevalence and correlates of being bullied among in-school adolescents in Beijing: Results from the 2003 Beijing Global School-Based Health Survey. Annals of General Psychiatry, 7, Article 6. doi:10.1186/1744-859X-7-6

32.

Holt

M. K.

Finkelhor

Kantor

G. K.

(2007a). Hidden forms of victimization in elementary students involved in bullying. School Psychology Review, 36, 345-360.

33.

Holt

M. K.

Finkelhor

Kantor

G. K.

(2007b). Multiple victimization experiences of urban elementary school students: Associations with psychosocial functioning and academic performance. Child Abuse & Neglect, 31, 503-515.

34.

Hosmer

D. W.

Lemeshow

(2000). Applied logistic regression (2nd ed.). New York, NY: John Wiley.

35.

Huang

X. Q.

Zhou

J. X.

Guo

L. T.

(2005). Bully in primary school and its impact on psychosocial health. Chinese Mental Health Journal, 19, 676-678.

36.

Kim

Y. S.

Koh

Y. J.

Leventhal

B. L.

(2004). Prevalence of school bullying in Korean middle school students. Archives of Pediatrics & Adolescent Medicine, 158, 737-741.

37.

Lai

H. H.

(2002). China’s Western development program. Modern China, 28, 432-466.

38.

Lynch

Ciccheti

(1998). An ecological-transactional analysis of children and contexts: The longitudinal interplay among child maltreatment, community violence, and children’s symptomatology. Development and Psychopathology, 10, 235-257.

39.

Mohr

(2006). Family variables associated with peer victimization: Does family violence enhance the probability of being victimized by peers? Swiss Journal of Psychology, 65, 107-116.

40.

Mrug

Loosier

P. S.

Windle

(2008). Violence exposure across multiple contexts: Individual and joint effects on adjustment. American Journal of Orthopsychiatry, 78, 70-84.

41.

Nansel

T. R.

Overpeck

Pilla

R. S.

Ruan

W. J.

Simons-Morton

Scheidt

(2001). Bullying behaviors among US youth: Prevalence and association with psychosocial adjustment. Journal of the American Medical Association, 285, 2094-2100.

42.

Olweus

(1999). Sweden. In Smith

P. K.

Morita

Junger-Tas

Olweus

Catalano

Slee

(Eds.), The nature of school bullying: A cross-national perspective (pp. 7-27). New York, NY: Routledge.

43.

Olweus

(2003). A profile of bullying. Educational Leadership, 60(6), 12-17.

44.

Perry

D. G.

Hodges

E. V. E.

Egan

S. K.

(2001). Determinants of chronic victimization by peers: A review and new model of family influence. In Graham

J. J. S.

(Ed.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 73-104). New York, NY: Guilford Press.

45.

Pynoos

R. S.

Rodriguez

Steinberg

A. S.

(1998). The UCLA PTSD Reaction Index for DSM IV (Revision 1). Los Angeles, CA: UCLA Trauma Psychiatry Program.

46.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

47.

Sansone

R. A.

Lam

Wiederman

M. W.

(2010). Being bullied in childhood: Correlations with borderline personality in adulthood. Comprehensive Psychiatry, 51, 458-461.

48.

Saunders

B. E.

(2003). Understanding children exposed to violence: Toward an integration of overlapping fields. Journal of Interpersonal Violence, 18, 356-376.

49.

Schwartz

Dodge

K. A.

Pettit

G. S.

Bates

J. E.

(1997). The early socialization of aggressive victims of bullying. Child Development, 68, 665-675.

50.

Schwartz

Dodge

K. A.

Pettit

G. S.

Bates

J. E.

(2000). Friendship as a moderating factor in the pathway between early harsh home environment and later victimization in the peer group. Developmental Psychology, 36, 646-662.

51.

Seals

Young

(2003). Bullying and victimization: Prevalence and relationship to gender, grade level, ethnicity, self-esteem, and depression. Adolescence, 38, 735-747.

52.

Shields

Cicchetti

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

53.

Soeken

K. L.

McFarlane

Parker

Lominack

M. C.

(1998). The abuse assessment screen: A clinical instrument to measure frequency, severity, perpetrator of abuse against women. In Campbell

J. C.

(Ed.), Empowering survivors of abuse: Health care for battered women and their children (pp. 195-204). Thousand Oaks, CA: Sage.

54.

Solberg

M. E.

Olweus

Endresen

I. M.

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

55.

Sourander

Jensen

Rönning

J. A.

Niemelä

Helenius

Sillanmäki

. . . Almqvist

(2007). What is the early adulthood outcome of boys who bully or are bullied in childhood? The Finnish “From a Boy to a Man” study. Pediatrics, 120, 397-404.

56.

Strassberg

Dodge

K. A.

Pettit

G. S.

(1994). Spanking in the home and children’s subsequent aggression toward kindergarten peers. Development and Psychopathology, 6, 445-461.

57.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1999). The Personal and Relationships Profile (PRP). Retrieved from http://pubpages.unh.edu/~mas2/PR10N.pdf

58.

Sun

Lin

Semaana

(2008). Child behaviour and parenting in HIV/AIDS-affected families in China. Vulnerable Children and Youth Studies, 3, 192-202.

59.

Undheim

A. M.

Sund

A. M.

(2010). Prevalence of bullying and aggressive behavior and their relationship to mental health problems among 12-to 15-year-old Norwegian adolescents. European Child & Adolescent Psychiatry, 19, 803-811.

60.

Walden

L. M.

Beran

(2010). Attachment quality and bullying behavior in school-aged youth. Canadian Journal of School Psychology, 25(1), 5-18.

61.

West

M. L.

Rose

M. S.

Spreng

Sheldon-Keller

Adam

(1998). Adolescent attachment questionnaire: A brief assessment of attachment in adolescence. Journal of Youth and Adolescence, 27, 661-672.

62.

Wilczenski

F. L.

Steegmann

Braun

Feeley

Griffin

Horowitz

Olson

(1997). Children as victims and victimizers: Intervention to promote “fair play.” School Psychology International, 18, 81-89.

63.

World Health Organization. (2004). Handbook for the documentation of interpersonal violence prevention programme. Geneva, Switzerland: Author.

64.

Zhang

W. X.

C. H.

Wang

M. P.

Wang

Y. W.

Jones

(2000). Gender differences in the bully/victim problem among primary and junior middle school students. Psychological Science, 23, 435-439.

65.

Zhu

Y. H.

(2012). A study of c Björkqvist hild bullying victimization in Xi’an, China: Prevalence, correlates and co-occurrence with family violence (Unpublished doctoral dissertation). The University of Hong Kong, Hong Kong.

66.

Zhu

Y. H.

Chan

K. L.

(2015). Prevalence and correlates of school bullying victimization in Xi’an, China. Violence and Victims, 30, 714-730.