Predictors of Dating Violence Among Chinese Adolescents

Abstract

In Chinese societies, violence among adolescent dating partners remains a largely ignored and invisible phenomenon. The goal of this study is to examine the relationships among gender-role beliefs, attitudes justifying dating violence, and the experiences of dating-violence perpetration and victimization among Chinese adolescents. This study has used self-reporting measures to collect data from a probability sample of 976 adolescents (mean age = 15.9) in three Chinese societies: Taiwan, Hong Kong, and Shanghai. Research results reveal a high prevalence of dating violence (including physical violence, sexual violence, and controlling behavior) among Chinese adolescents with dating experience: the perpetration rate is 27.3% and the victimization rate is 39%. Study results demonstrate that adolescents who endorse traditional gender-role beliefs tend to view dating violence as acceptable behavior. Boys’ endorsement of traditional gender roles, boys’ attitudes justifying boy-on-girl violence, and boys’ attitudes against girl-on-boy violence predict boys’ actual sexual-violence behavior. Moreover, boys’ attitudes justifying boy-on-girl dating violence is the strongest predictor of boys’ perpetration of physical and sexual dating violence. This study also shows that boys’ hostility is a significant predictor of boys’ controlling behavior. Programs for preventing dating violence should include components designed to challenge traditional gender-role beliefs and attitudes justifying dating violence.

Keywords

dating violence adolescents gender role justification of violence Chinese

Introduction

Dating violence is a pervasive and serious problem worldwide (Straus, 2008). Contemporary Western research has substantially advanced our understanding of the prevalence, severity, and detrimental consequences of dating violence (Lewis & Fremouw, 2001). For example, U.S. studies have shown that between a third and a half of all girls and boys will experience some form of physical partner violence, which makes physical violence a common event rather than an isolated problem in the lives of adolescents (Barter, 2009). Moreover, dating violence is approximately three times more prevalent than violence between married couples (Straus, 2008).

Researchers have documented that dating violence victimization is damaging to the victims’ physical health, mental health, sexual health, and social functioning (e.g., Banyard & Cross, 2008; Lewis & Fremouw, 2001; Silverman, Raj, Mucci, Lorelei, & Hathaway, 2001). In more serious instances, intimate partner violence (IPV) can lead to homicide and suicide (Harper & Voigt, 2007).

Adolescents are at a crucial developmental stage when a dating relationship is first initiated and when the risk of dating violence first emerges (Hickman, Jaycox, & Aronoff, 2004). Moreover, gender-role identities are being challenged and shaped during adolescence. Research suggests that adherence to traditional gender-role ideology is associated with the justification and the actual perpetration of relationship violence (Reitzel-Jaffe & Wolfe, 2001). Furthermore, behaviors and attitudes that concern intimate relationships and that are learned during these formative years frequently develop into lifelong patterns (Smith, White, & Holland, 2003). Therefore, adolescence represents a critical stage for fundamental prevention and intervention. In these regards, this study examines two critical factors associated with dating violence: gender-role beliefs and attitudes justifying dating violence.

Gender-role stereotypes are viewed as key elements contributing to dating violence (Mahlstedt & Welsh, 2005) although previous studies indicated inconsistent findings (Sugarman & Hotaling, 1998), especially regarding different types of violence (e.g., physical vs. sexual). Recent studies found that traditional gender-role beliefs are associated with sexual dating-violence perpetration and victimization. For example, one study surveyed 324 boys and 309 girls in Canada and revealed that boys’ use of sexual dating violence and girls’ use of psychological dating violence were linked to risk factors suggesting an enactment of social scripts associated with their respective gender roles (Sears, Byers, & Price, 2007). A longitudinal study (n = 1,291 American adolescents) revealed that girls’ adherence to traditional gender stereotypes predicted the girls’ chronic victimization from sexual dating violence, but not boys’ victimization (Foshee, Benefield, Ennett, Bauman, & Suchindran, 2004). Studies using samples of Chinese American college students indicate that gender-role beliefs are not associated with physical IPV (Yick & Agbayani-Siewert, 2000).

Researchers have demonstrated that perception of violence as justifiable under certain circumstances increases the risk of dating violence (Jackson, 1999; Lewis & Fremouw, 2001). O’Keefe (1997) found a significant relationship between perpetration of male-to-female violence and justification of male-to-female violence among boys. Reitzel-Jaffe and Wolfe’s study (2001) revealed that young undergraduates (n = 585 males) who endorsed both traditional gender-role ideology and attitudes condoning relationship violence were more likely to physically assault partners than were those endorsing either traditional gender-role ideology or attitudes condoning relationship violence alone. A few studies have revealed the cultural differences in attitudes justifying dating violence. For example, Chan and Straus (2008) compared 651 Hong Kong college students with 1,085 U.S. college students and found that Hong Kong students were more approving of dating violence than were the U.S. students.

In summary, literature on dating violence suggests connections among gender-role beliefs, attitudes justifying violence, and actual violent behaviors. Social role theory suggests that sexist ideologies offer justifications for male violence toward female intimate partners who violate gender prescriptions, such as when a wife commits an act of infidelity (Rudman & Glick, 2008).

In addition to gender-role beliefs and attitudes justifying violence, three correlates derived from previous studies on adolescent dating violence (Hickman et al., 2004) were used as control variables in the current study’s multivariate analyses: age, hostility, and dating violence victimization. Reports of the National Crime Victimization Survey (NCVS) show that physical and sexual victimization by intimate partners is relatively low for adolescents between 12 and 15 years old, but higher for those between 16 and 19 years old (as cited in Hickman et al., 2004). O’Keefe (1997) examined contexts of dating violence perpetration among high school students and found that boys and girls reported anger as their primary reason for inflicting violence and that the strongest predictor in multivariate analyses of an adolescent’s perpetration of violence was the adolescent’s experience as a recipient of dating violence.

Dating Violence in Chinese Societies

Although researchers have significantly advanced our knowledge of dating violence, they have surveyed primarily nonrandom college or high school White students, with little representation of ethnic minorities (Jackson, 1999; Lewis & Fremouw, 2001). Dating violence experiences in other populations or societies are understudied. In Chinese societies, violence among adolescent dating partners remains a largely ignored phenomenon, as compared with violence among married adults. A few studies on dating violence have targeted college students in Taiwan, Hong Kong, and China (e.g., Chan & Straus, 2008; Shen, 2008; Straus, 2008), but no previous study, so far, has surveyed middle school and high school students’ dating violence in Chinese societies.

Moreover, the majority of studies on adolescent dating violence have focused on physical violence, with limited attention to either sexual violence or controlling behavior. Given the lack of empirical data on dating violence among Chinese adolescents and given the patriarchal social structure in Chinese societies, this study uses a probability sample to examine relationships among the gender-role beliefs, the attitudes justifying violence, and the dating violence experiences of adolescents in Taiwan, Hong Kong, and Shanghai. Specifically, the present study has tested two hypotheses based on prior research and social role theory:

Hypothesis 1: Adolescents’ traditional gender-role beliefs are associated with the frequency of both dating violence perpetration and dating violence victimization.

Hypothesis 2: Adolescents’ attitudes justifying violence are associated with the frequency of both dating violence perpetration and dating violence victimization.

Method

Research Design and Procedure

The research design of this study is cross-sectional and correlational. This research project was approved and funded by the National Science Council of Taiwan and National Taiwan University.

Prior to data collection, informed consent was obtained from school teachers and students of a required class scheduled for data collection. Trained research assistants explained the research purpose and procedures to students prior to the questionnaire administration. The voluntary and anonymous nature of the research was emphasized. Research assistants informed students of their rights to refuse or discontinue participation at any time. Self-report questionnaires were then distributed to consenting students in group sessions scheduled during class hours. Questionnaire administration averaged 25 min in length.

Participants

The current study has analyzed a subsample of 976 adolescents with dating experience. The original participants consisted of 3,138 adolescents studying in middle and high schools. Two-step stratified random samples were drawn from the available school lists in three sites: Taiwan, Hong Kong, and Shanghai. The random sample was first stratified by school and then stratified by class. Target subjects were those in the 2nd through 6th years of secondary schools. A total of 42 schools (out of 53 schools contacted) agreed to participate in this study, and this total consisted of 18 schools in Taiwan, 11 schools in Hong Kong, and 13 schools in Shanghai. The average school-participation rate was 79% among the three sites, resulting in a valid sample size of 3,138 students. The pilot study was conducted in the winter of 2008 with 361 subjects among the three sites, and the formal data were collected in the spring of 2008. Out of the 3,138 participants, 31.7% (n = 994) had reported dating experience that lasted at least 1 month. Of the 994 participants with dating experience, 18 were excluded from the statistical analyses because of their unknown gender. Therefore, the sample of 976 adolescents with dating violence was analyzed in the current study.

Measures

Measures in the self-report, pen-and-paper questionnaire yielded information about demographics, dating experiences, gender-role beliefs, and attitudes justifying violence. Before being administered to the pilot-study participants, the questionnaire was first examined by scholars in Taiwan, Hong Kong, and Shanghai for content validity.

Gender-role beliefs

This study used the Attitudes Toward Women Scale (ATWS; Spence & Helmreich, 1978), modifying it to measure traditional or egalitarian gender-role beliefs among Chinese adolescents. An example of an item is “The intellectual leadership of a community should be largely in the hands of men.” The measure uses a 4-point Likert-type scale response format, ranging from 1 (strongly disagree) to 4 (strongly agree). The egalitarian items were reverse scored. The item scores were added up to obtain a total scale score for each participant. The modified ATWS has 12 items and scale scores ranging from 12 to 48, with the higher scores indicating more traditional attitudes. The ATWS has been widely used in other studies and has shown good internal consistency and test-retest reliability (Spence & Hahn, 1997). The Cronbach alpha for this study’s ATWS (i.e., the Chinese version used with the sample of Chinese adolescents) was .78, showing good internal consistency reliability.

The original ATWS has 15 items. The internal consistency reliability and content validity analyses of the pilot-study data for the Chinese version of ATWS led to the decision to exclude three original ATWS items: Items 1, 3, and 13 (Item 1: “Swearing and obscenity are more repulsive in the speech of a woman than in that of a man”; Item 3: “It is insulting to women to have the ‘obey’ clause still in the marriage service”; and Item 13: “Economic and social freedom is worth far more to women than acceptance of the ideal of femininity, which has been set up by men”).

Attitudes justifying violence

One of the current study’s authors (Shen, 2008) developed the Attitudes Justifying Dating Violence Scale on the basis of previous studies (O’Keefe, 1998; Pflieger & Vazsonyi, 2006; Yick & Agbayani-Siewert, 2000) to assess the extent to which participants would agree or disagree with assertions regarding whether or not certain circumstances justify hitting a dating partner. The scale contains 12 circumstances: He or she (a) threatens to break up, (b) is drunk, (c) hits first during an argument, (d) is cheating on her or him, (e) calls the other an unpleasant name, (f) flirts with someone else, (g) insults the other in front of friends, (h) forbids the other to go out with friends at night, (i) refuses to have sex, (j) antagonizes the other, (k) disobeys the other, and (l) screams hysterically. The scale employs a 4-point Likert-type scale, ranging from 1 (strongly disagree) to 4 (strongly agree). The item scores were added up, yielding two total scale scores: attitudes justifying boy-on-girl violence and attitudes justifying girl-on-boy violence for each participant. Scores ranged from 12 to 48: the higher the score, the greater the agreement with contextual justifications of partner-directed violence. The current sample’s Cronbach alpha is .89 for attitudes justifying boy-on-girl violence and .84 for attitudes justifying girl-on-boy violence, demonstrating strong internal consistency reliability.

Dating Violence Scale

One of the authors (Shen, 2008) developed the Dating Violence Scale to assess youths’ experiences with dating-violence perpetration and victimization on the basis of previous studies (Huang & Wang, 2005; O’Keefe, 1998; Straus, 1979). As summarized in Table 2, the instrument contains 15 pairs of items measuring both perpetration and victimization relative to physical violence (9 items), sexual violence (2 items), and controlling behaviors and stalking (4 items). Participants were categorized as perpetrators of dating violence if they reported that they had committed one or more violent acts on one or more occasions against their current or last partner during the course of their dating. Similarly, participants were categorized as having experienced dating violence victimization if they reported that their current or last partner had committed one or more violent acts on one or more occasions against them during the course of their dating.

In addition to identifying the occurrence of dating violence, this study focuses on frequency of violence. Participants responded to items on a 7-point Likert-type scale (from 0 = never to 6 = 20 times or more) to rate the frequency with which they had experienced each violent act either as a perpetrator or as a victim with their current or last dating partner. The current study has calculated the Perpetration and Victimization Scale scores by adding the midpoints of each participant’s responses in the scale’s response categories. The Perpetration Scale scores represent the number of incidents of violent acts experienced by a participant as a perpetrator; and the Victimization Scale scores represent the number of incidents of violence acts experienced by a participant as a victim. The midpoints are the same as the response-category numbers for categories 0, 1, and 2. For category 3 (3-5 times), the midpoint is 4; for category 4 (6-10 times), it is 8; for category 5 (11-20 times), it is 15; and for category 6 (more than 20 times), it is 25 (as suggested by Straus, Hamby, Boney-McCoy, & Sugarman, 1996). The internal consistency Cronbach alpha of the total scale for the present sample is .82 for perpetration items and .81 for victimization items, showing good internal consistency reliability.

Control measures

Three additional measures were used as control variables in the current study’s hierarchical regression analyses: age (measured as a single item), hostility (via the Chinese Hostility Inventory), and dating violence victimization (via the Victimization Scale).

The Chinese Hostility Inventory (CHI)

The current study used the CHI (Weng et al., 2008) to measure participants’ hostility and anger. The measure uses a 4-point Likert-type scale response format, ranging from 1 (never) to 4 (always). Sample items include “It is hard for me to control my temper” and “If someone hits me first, I will hit back.” The item scores were added up so that a total scale score for each participant could be obtained. Scores ranged from 20 to 80, with the higher scores indicating more hostility.

The CHI has exhibited good psychometric properties: the internal consistency Cronbach alpha is .89 and the 4-week test-retest reliability is .80 for the total inventory with the original sample. The internal consistency Cronbach alpha of the total scale for the present sample is .87, showing good internal consistency reliability.

Data Analyses

This study employed the SPSS 16.0.1 for Windows software (SPSS, 2007) to analyze data, and conducted correlation and multiple regression analyses to examine the relationships among gender-role beliefs, attitudes justifying violence, and dating violence experience. In addition, this study conducted chi-square and t-test analyses to test the gender differences in participants’ demographics, relationship characteristics, and independent and dependent variables. Using the pairwise method, this study excluded some cases from the data analyses when missing values existed in certain variables, resulting in different total case numbers in different statistical analyses.

Results

Participant Characteristics

Of the 976 participants with dating experience, 49% were boys (n = 478), 46.5% were studying in middle school (n = 454), 53.5% were studying in high school (n = 522), 42.9% were living in Hong Kong (n = 419), 39.5% were living in Taiwan (n = 386), and 17.5% were living in Shanghai (n = 171). The mean age of the participant was 15.9 years with a range between 13 and 20 years. The mean age for beginning dating was 13.7 years and the average length of the dating relationship lasted for 7.7 months (see Table 1). The age ranges for the participants in the 2nd through 6th year of secondary education were as follows respectively: 13 to 16 years old (M = 14.2), 14 to 18 years old (M = 14.9), 15 to 18 years old (M = 16.4), 16 to 18 years old (M = 16.7), and 16 to 20 years old (M = 17.9).

Table 1.

Participant Characteristics (N = 976)

	Overall (n = 976)		Boys (n = 478)		Girls (n = 498)
Variable	%	n	%	n	%	n	χ² test
Site							.85
Taiwan	39.5	386	38.1	182	41.0	204
Hong Kong	42.9	419	43.9	210	42.0	209
Shanghai	17.5	171	18.0	86	17.1	85
Schooling							6.8**
Middle school	46.5	454	42.3	202	50.6	252
High school	53.5	522	57.7	276	49.4	246
Parental status							.001
Married	83.3	813	83.1	397	83.5	416
Divorced	10.1	99	10.9	52	9.4	47
Other	5.7	56	5.4	26	6.0	30
Dating experience							3.5
Currently in relationship	41.7	407	38.7	185	44.6	222
Yes, but not currently	58.3	569	61.3	293	55.4	276
Dating Partner is a							30.7***
Working adult	3.1	30	0.2	1	5.8	29
Schoolmate	33.4	326	34.9	167	31.9	159
Classmate	22.4	219	24.5	117	20.5	102
Student in other school	30.2	295	26.6	127	33.7	168
Other	1.3	13	1.3	6	1.4	7
Perpetration rate²	27.3	239	25.7	106	28.7	133	.95
Controlling behavior	18.8	171	19.4	84	18.2	87	.25
Physical violence	19.2	175	17.6	76	20.7	99	1.4
Sexual violence	7.1	65	10.9	47	3.8	18	17.4***
Victimization rate²	39.0	346	46.2	194	32.5	152	17.5***
Controlling behavior	24.3	221	24.3	105	24.2	116	.001
Physical violence	27.5	250	39.4	170	16.7	80	58.9***
Sexual violence	9.7	88	10.9	47	8.6	41	1.4
	Mean	SD	Mean	SD	Mean	SD	t test
Age	15.9	1.6	16.1	1.6	15.8	1.6	3.4***
Age when first dating	13.7	1.9	13.8	2.1	13.7	1.7	0.7
No. of dates altogether	2.1	2.1	2.1	2.5	2.1	1.6	−.2
No. of months the latest relationship lasted	7.7	9.4	7.9	9.0	7.6	9.7	0.5
Attitudes Toward Women Scale scores	21.8	4.8	23.8	4.6	20.0	4.3	13.2***
Attitudes Justifying Boy-on-girl Violence (Scale scores)	23.3	6.9	23.5	7.4	23.0	6.4	0.9
Attitudes Justifying Girl-on-boy Violence (Scale scores)	27.8	6.8	28.1	6.9	27.5	6.7	1.3

Note: (1) Totals or percentages may not equal 976 or 100% owing to missing data or rounding error; (2) valid percentage.

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

Experiences With Dating Violence

Of the participants with dating experience (n = 976), 27.3% (n = 239) reported that they had perpetrated violence (including controlling behavior, physical violence, and sexual violence) toward their dating partners, while 39% (n = 346) reported that they had experienced dating violence victimization (valid percentages). Of participants who had dating violence experience (n = 392), 46.7% (n = 183) had experienced both being a perpetrator and being a victim, while 14.3% (n = 56) had experienced perpetration only and 39% (n = 153) had experienced victimization only. Boys reported higher prevalence rates of sexual violence perpetration than girls did (10.9% vs. 3.8%; χ² = 17.4, p = .000). Boys also reported higher prevalence rates of physical violence victimization than girls did (39.4% vs. 16.7%; χ² = 58.9, p = .000). There were no gender differences in the prevalence rates of controlling behavior (both perpetration and victimization), physical violence perpetration, and sexual violence victimization (see Table 1).

Table 2 shows the percentages attributable to the different violent acts experienced by the participants in their dating relationships. For all participants, the most prevalent types of violent acts experienced were “My partner pushed, grabbed, or shoved me” (15%) and “My partner restricted my actions” (12.5%). The most prevalent violent acts inflicted by participants on their partners were “I threw something at him/her” (8.6%) and “I restricted his/her actions” (8.2%). For boys, the most prevalent type of violent act experienced was “My partner pushed, grabbed, or shoved me” (21.8%); and the most prevalent violent act inflicted by boys on their partners was “I prohibited him/her from contacting relatives or friends” (8.4%). For girls, the most prevalent type of violent act experienced was “My partner restricted my actions” (12%); and the most prevalent violent act inflicted by girls on their partners was “I threw something at him/her” (9.8%).

Table 2.

Type and Percentage of Dating Violence Experienced and Perpetrated by Participants (All Sample N = 976; boys N = 478; girls N = 498)

	Perpetration Percentage			Victimization Percentage
Perpetration Items	All	Boy	Girl	All	Boy	Girl	Victimization Items
Controlling behavior
I restricted his/her actions.	8.2	7.3	9.0	12.5	13.0	12.0	My partner restricted my actions.
I harassed him/her with phone calls.	7.3	6.9	7.6	9.9	9.8	10.0	My partner harassed me with phone calls.
I prohibited him/her from contacting relatives or friends.	7.1	8.4	5.8	11.0	12.1	9.8	My partner prohibited me from contacting relatives or friends.
I stalked him/her.	2.8	4.0	1.6	3.5	3.3	3.6	My partner stalked me.
Physical violence
I threw something at him/her.	8.6	7.3	9.8	11.4	16.9	6.0	My partner threw something at me.
I pushed, grabbed, or shoved him/her.	7.7	7.1	8.2	15.0	21.8	8.4	My partner pushed, grabbed, or shoved me.
I hit him/her with something.	4.4	4.0	4.8	8.0	13.2	3.0	My partner hit me with something.
I kicked, bit, or hit him/her with a fist.	4.2	3.1	5.2	8.3	14.2	2.6	My partner kicked, bit, or hit me with a fist.
I slapped him/her.	2.8	2.3	3.2	4.7	7.7	1.8	My partner slapped me.
I choked him/her.	1.4	2.1	0.8	2.7	3.6	1.8	My partner choked me.
I beat him/her up.	1.2	2.1	0.4	2.4	4.6	0.2	My partner beat me up.
I used a knife or a weapon to hurt him/her.	0.8	1.3	0.4	1.0	1.5	0.6	My partner used a knife or a weapon to hurt me.
I threatened him/her with a knife or weapon.	0.4	0.8	0.0	0.7	1.3	0.2	My partner threatened me with a knife or weapon
Sexual violence
I touched his/her body in an unwanted way.	2.6	4.4	0.8	6.1	6.7	5.6	My partner touched my body in an unwanted way.
I used force to make him/her have sex.	1.1	2.3	0.0	0.7	1.0	0.4	My partner used force to make me have sex.

Note: (1) Perpetration items are listed by the rank order of percentages in each violence type. (2) The percentage was calculated by totaling the percentages of responses 1 (one time) through 6 (more than 20 times).

Attitudes Justifying Dating Violence

Study results show that 12.1% to 64.9% of the adolescents agreed that hitting one’s dating partner is acceptable under certain circumstances. Adolescents were mostly in agreement that it is okay for a girl to hit her boyfriend if (a) he hits her first during an argument (64.9%; this and the following percentages were calculated by summing the percentage of agree and strongly agree responses), (b) he is cheating on her (61.8%), and (c) he is flirting with someone else (56.4%). For boy-on-girl violence, adolescents were mostly in agreement that it is okay for a boy to hit his girlfriend if (a) she is flirting with someone else (37.6%), (b) she is cheating on him (36.8%), and (c) she hits him first during an argument (23.1%). The mean score across 12 items assessing attitudes justifying girl-on-boy violence was 2.3 (4 = strongly agree; 1 = strongly disagree), and the mean score for attitudes justifying boy-on-girl violence was 1.9. These results suggest that participants were more accepting of girls’ perpetration of physical violence against dating partners than boys’ perpetration of physical violence against dating partners. In terms of the total scale scores, independent t-test analyses show that there were no gender differences for either attitudes justifying girl-on-boy violence (n = 957, df = 955, t = 1.3, p = .20) or boy-on-girl violence (n = 950, df = 948, t = .93, p = .35).

Predictors of Dating Violence Frequency

Before testing the regression model, the current study conducted initial correlation analyses to examine the associated interplay of gender-role beliefs, attitudes justifying boy-on-girl and girl-on-boy violence, and the frequency of dating violence perpetration and victimization (Table 3). For all participants, both gender-role beliefs (r = .157, p = .000) and attitudes justifying boy-on-girl violence (r = .159, p = .000) were associated with dating violence perpetration (total scale frequency scores) and its three subscales: controlling behavior, physical violence, and sexual violence. Gender-role beliefs and attitudes justifying boy-on-girl violence were positively correlated (r = .20, p = .000). However, gender-role beliefs and attitudes justifying boy-on-girl violence were not correlated with the total scale frequency scores for dating violence victimization. Dating violence victimization was correlated only with dating violence perpetration (r = .282, p = .000). Attitudes justifying girl-on-boy violence had a significant positive association with attitudes justifying boy-on-girl violence (r = .34, p = .000), and had a significant negative association with sexual violence perpetration (r = −.064, p = .031).

Table 3.

Correlations Among Gender-Role Beliefs, Attitudes Justifying Boy-on-Girl and Girl-on-Boy Violence, and Dating Violence (All Sample N = 976; Boys N = 478; Girls N = 498)

	Gender-Role Beliefs	Attitudes Justifying Boy-on-Girl Violence	Attitudes Justifying Girl-on-Boy Violence	Hostility	Perpetration Scale	Victimization Scale	Controlling Behavior Perpetration	Physical Violence Perpetration	Sexual Violence Perpetration
Gender-role beliefs	1.000
Attitudes justifying boy-on-girl violence
All sample	.200***	1.000
Boys	.261***
Girls	.138**
Attitudes justifying girl-on-boy violence
All sample	.025	.340***	1.000
Boys	−.043	.300***
Girls	.070	.386***
Hostility
All sample	−.051	.140***	.136***	1.000
Boys	.085*	.174***	.180***
Girls	−.099*	.105*	.105*
Perpetration Scale
All sample	.157***	.159***	.005	.114***	1.000
Boys	.214***	.219***	−.047	.138**
Girls	.067	.065	.074	.090*
Victimization Scale
All sample	.050	.039	.025	.054	.282***	1.000
Boys	−.010	.027	.042	.049	.312***
Girls	.003	.053	−.009	.107*	.221***
Controlling behavior perpetration
All sample	.124***	.125***	.042	.107**	.742***	.171***	1.000
Boys	.181***	.156**	−.005	.184***	.752***	.150**
Girls	.059	.087*	.094*	.041	.737***	.199***
Physical violence perpetration
All sample	.094**	.119***	.005	.066*	.804***	.281***	.246***	1.000
Boys	.136**	.171***	−.034	.060	.790***	.360***	.228***
Girls	.043	.040	.054	.078*	.835***	.149**	.275***
Sexual violence perpetration
All sample	.174***	.132***	−.064*	.095**	.618***	.099**	.506***	.267***	1.000
Boys	.203***	.183***	−.102*	.137**	.731***	.104*	.635***	.333***
Girls	.080*	−.031	.047	.088*	.298***	.126**	.418***	.035

p < .05. **p < .01. ***p < .001, one-tailed.

Separate correlation analyses were conducted with the same variables for boys and girls. The results concerning associations among variables remained the same as the total sample’s results, except that for girls, neither gender-role beliefs (r = .067, p = .16) nor attitudes justifying boy-on-girl or girl-on-boy violence (r = .065, p = .08; r = .074, p = .056, respectively) were associated with dating violence perpetration (total scale). For subscales, attitudes justifying boy-on-girl and girl-on-boy violence were correlated with girls’ perpetration of controlling behavior (r = .087, p = .033; r = .094, p = .022, respectively); and gender-role beliefs were correlated with girls’ perpetration of sexual violence (r = .08, p = .046). Because correlation analyses show that the gender-role beliefs and attitudes justifying boy-on-girl and girl-on-boy violence were not associated with either victimization (total scale for all samples) or girls’ perpetration (total scale), the current study conducted the following multiple regressions on boys’ perpetration only (n = 106).

Three hierarchical multiple regression analyses examined whether gender-role beliefs and attitudes justifying violence were significant contributors to the frequency of male-perpetrated dating violence after other potential related factors had been controlled for. Dependent variables were physical violence frequency (M = 5.1; SD = 12.7; ranging from 0 to 76; skewness = 3.7; Kurtosis = 15.2), sexual violence frequency (M = 1.4; SD = 5.2; ranging from 0 to 33; skewness = 4.7; Kurtosis = 22.6), and controlling behavior frequency (M = 4.4; SD = 9.9; ranging from 0 to 75; skewness = 4.9; Kurtosis = 29.9). Control variables in the regression analysis are as follows: age (M = 16; SD = 1.7), victimization frequency (M = 17.2; SD = 25.7), and CHI scale scores (M = 64.2; SD = 13).

In each hierarchical multiple regression analysis, control variables were entered into the regression model as the first block (i.e., Model 1). Gender-role beliefs and boys’ attitudes justifying boy-on-girl and girl-on-boy violence were entered into the regression model as the second block (Model 2).

The results show that boys’ attitudes justifying boy-on-girl violence turned out to be the most significant predictor (according to the Beta value) of physical violence perpetration (β = .351) and sexual violence perpetration (β = .358), when all variables were entered simultaneously in Model 2. This one particular variable predicts an additional 11.8% unique variance in boys’ actual physical-violence behaviors, beyond the 11.7% variance explained by control variables. Therefore, the hierarchical regression analyses support the hypothesis that attitudes justifying boy-on-girl violence predict the frequency of boys’ perpetration of physical and sexual violence. However, attitudes justifying boy-on-girl and girl-on-boy violence did not predict boys’ controlling behavior.

The results also show a significant effect of gender-role beliefs on sexual violence perpetration. Boys’ endorsement of traditional gender roles, along with both boys’ attitudes justifying boy-on-girl violence and boys’ attitudes against girl-on-boy violence, predict the frequency of boys’ actual sexual-violence behavior, after other risk factors were controlled for. These three variables together predict an additional 26.4% unique variance in boys’ sexual-violence frequency, beyond the 13.5% variance explained by control variables. Therefore, the hierarchical regression analyses support the hypothesis that gender-role beliefs are a significant predictor of the frequency of boys’ sexual violence perpetration after other potential risk factors are controlled for. However, gender-role beliefs were not a significant predictor of either the frequency of boys’ physical violence perpetration or the frequency of boys’ controlling behavior perpetration.

Among the control variables, the frequency of a boy’s dating violence victimization (total scale) was a significant predictor of the frequency of boy’s physical violence perpetration. In addition, hostility turned out to be a significant predictor of both sexual violence and controlling behavior perpetration.

Table 4.

Predictors of Boys’ Perpetration Frequency (N = 106)

	Physical Violence			Controlling Behavior			Sexual Violence
	β	t	p	β	t	p	β	t	p
Model 1
Age	.171	1.62	.108	.171	1.72	.089	.132	1.27	.208
Victimization Scale	.282	2.67	.009**	.20	.21	.837	−.024	−.23	.818
Hostility	.052	.49	.624	.445	4.47	.000***	.359	3.44	.001**
	R² = .117			R² = .205			R² = .135
	F_{(3, 82)} = 3.64*			F_{(3, 83)} = 7.12***			F_{(3, 82)} = 4.26**
Model 2
Age	.172	1.70	.092	.150	1.55	.124	.082	.91	.365
Victimization Scale	.238	2.35	.021*	.003	.03	.979	−.044	−.50	.619
Hostility	−.045	−0.42	.673	.363	3.58	.001**	.223	2.37	.020*
Gender-role Beliefs	.058	.54	.590	.188	1.81	.075	.254	2.65	.010*
Attitudes Justifying Boy-on-girl Violence	.351	3.00	.004**	.173	1.56	.123	.358	3.50	.001**
Attitudes Justifying Girl-on-boy Violence	−.187	−1.72	.090	−.088	−.86	.392	−.270	−2.85	.006**
	R² = .235			R² = .288			R² = .399
	F_{(6, 79)} = 4.05***			F_{(6, 80)} = 5.39***			F_{(6, 79)} = 8.73***
	R² change = .118			R² change = .083			R² change = .264
	F change_{(3, 79)} = 4.06**			F change_{(3, 80)} = 3.11*			F change_{(3, 79)} = 11.55***

Note: Bold font represents Significant values.

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

Discussion

The present research findings add to the growing knowledge on adolescent dating violence by extending research to the understudied populations of Chinese adolescents. Research results reveal the high prevalence of dating violence among Chinese adolescents with dating experiences: more than one fourth of participating adolescents (27.3%) reported that they had perpetrated dating violence, and more than one third of participating adolescents (39%) reported that they had experienced related victimization at the hands of their dating partners. The current study’s finding of a high prevalence of dating violence among Chinese adolescents is consistent with previous Western studies and suggests that aggression is very common in adolescent dating relationships (Barter, 2009).

This study has also examined the relationships among gender-role beliefs, attitudes justifying violence, and the different types of dating violence frequency among Chinese adolescents. Correlation analyses show that adolescents’ adherence to traditional gender roles was correlated with the adolescents’ attitudes justifying boy-on-girl violence. However, the significance of these variables as predictors of dating violence varies among different types of dating violence for male adolescents, which suggests that each type of dating violence has its own unique predictors. Regression analyses show that, regarding sexual violence, boys perpetrated more sexual violence against their dating partner if they (a) adhered to traditional gender-role beliefs, (b) believed that boy-on-girl violence was justifiable, or (c) believed that girl-on-boy violence was not justifiable. These findings are congruent with previous research indicating that adolescents endorsing traditional gender-role beliefs would be more predisposed to perpetrate sexual dating violence than individuals who hold relatively equalitarian gender-role beliefs (Sears et al., 2007).

For physical violence, present results also show that boys’ attitudes justifying boy-on-girl violence is the most significant predictor of physical violence perpetration. This finding is consistent with previous research demonstrating that attitudes endorsing violence as justifiable under certain circumstances are associated with the perpetration of physical dating violence (Jackson, 1999; Lewis & Fremouw, 2001). It should be noted that this study measured only the attitudes justifying physical violence (i.e., hitting dating partners) and did not measure attitudes justifying sexual violence or controlling behavior. Another significant predictor of boys’ perpetration of physical violence was the boys’ experience as victims of dating violence; in other words, some dating violence perpetrators might be the recipients of dating violence themselves, and there might be mutual violence in adolescent couples. This result is consistent with prior research indicating that, among adolescents, experiencing violence victimization is a significant predictor of perpetrating violence (O’Keefe, 1997).

Contrary to social role theory, the results of this study’s regression analyses show that gender-role beliefs were not a significant predictor of the frequency of boys’ perpetration of physical violence. This finding is consistent with studies conducted with Chinese Americans, indicating that gender-role beliefs are not associated with physical IPV (Yick & Agbayani-Siewert, 2000). In the present study’s correlation analyses, gender-role beliefs were significantly correlated with boys’ perpetration of physical violence. However, in the regression analyses, gender-role beliefs became a nonsignificant predictor, probably because other variables such as attitudes justifying boy-on-girl violence have more power in predicting boys’ perpetration of physical violence.

Controlling behavior is seldom researched by studies on adolescent dating violence. However, this study reveals that controlling behavior is the most prevalent type of violence perpetrated by male adolescents (19.4%). This study also shows that boys’ hostility is the only significant predictor of boys’ controlling behavior. In other words, boys with relatively high levels of hostility displayed more controlling behavior toward their dating partners than did boys with relatively low levels of hostility. Hostility was also a significant predictor of boys’ perpetration of sexual violence. Previous research reported that anger was adolescents’ primary reason for using physical, psychological, and sexual violence (O’Keefe, 1997). Future research can explore more potential predictors of controlling behavior.

Present results show that gender differences characterized various types of dating violence prevalence. More boys than girls reported perpetrating sexual violence and reported experiencing, as victims, physical violence. This finding is consistent with previous research indicating that reports of being a victim of physical violence are more common among male than female adolescents (Simon, Miller, Gorman-Smith, Orpinas, & Sullivan, 2010) and that boys are more likely to report committing acts of sexual violence against a dating partner than are girls (Hickman et al., 2004; O’Keefe, 1997). These aforementioned findings are consistent with the present study and prior U.S. studies, showing that it is more acceptable for a female to hit a male partner than the reverse (O’Keefe, 1997; Simon et al., 2010).

This study found no significant correlation among gender-role beliefs, attitudes justifying girl-on-boy violence, and violence perpetration and victimization for female adolescents. This absence of a significant correlation here is not consistent with some studies’ findings showing that dating violence victimization is related to traditional gender-role beliefs for females (Foshee et al., 2004). One plausible explanation for this inconsistency is that gendered roles have a more prescriptive influence on males’ reactions to a female’s deviation from her expected role in Chinese culture than in other cultures. This cultural trait may be especially pronounced because traditional Chinese culture is heavily colored by patriarchal values. Further research is needed to unravel the factors predicting girls’ violence perpetration and victimization in dating relationships among Chinese adolescents.

Practical Implications

The ultimate purpose of the current empirical study on dating violence is prevention and intervention. According to the present findings, prevention and intervention programs may be developed at both individual and societal levels for male adolescents. On an individual level, prevention programs for physical and sexual violence may focus on challenging traditional gender-role beliefs and changing attitudes and myths that might otherwise tend to support dating violence. Educators and practitioners may help adolescents learn alternatives to violence when the adolescents think that acts of violence against their dating partners are justifiable under certain circumstances. Prevention of and intervention in male adolescents’ controlling behavior toward their dating partners should focus on the boys’ hostility. Educational and therapeutic programs can be geared toward helping boys learn to identify their hostile feelings and to appropriately express them in a nonviolent way. Educators and parents can express to adolescents that anger is a feeling while violence is a behavior. Everyone has a right to what he or she feels. However, it is unacceptable to respond to these angry feelings in an abusive manner (Miles, 2005).

On the societal level, public advocacy through mass media and educational courses is essential to increase public and adolescents’ awareness of dating violence, with the goal of reducing the prevalence rate of dating violence perpetration and victimization. Efforts should also be made to alter the representation of traditional gender stereotypes in adolescent-targeting types of media.

Limitations

Use of self-report measures to collect data from middle and high school students might generate some limitations for this study. First, the self-report method itself is subject to errors in recall, intentional false responses, and traumatic-abuse memory inaccessibility, thereby possibly reducing data reliability and validity. However, Foshee et al. (2004) argued that self-reports of adolescent dating violence are the only reasonable options for obtaining individual measures of adolescent dating abuse, because few incidents of adolescent dating violence are witnessed by third parties. Second, although this study employed a large probability sample, contrasting sharply with shelter samples or small community samples, these data apply only to students who were attending school during the survey and, therefore, are not representative of all persons in this age group. Third, the survey is cross-sectional in nature, and therefore, causal relationships cannot be analytically identified from among the variables.

This study has many strong points despite these limitations. The survey response rate was high and the random sample came from the general population as opposed to selected groups in multiple Chinese societies, thus increasing the generalizability of study results. Moreover, this study examined not only physical violence but also sexual violence and controlling behaviors among a previously unstudied population, and this widened scope strengthens our understanding of dating violence across cultures.

Conclusion

This is the first study to examine adolescent dating violence among the Chinese societies of Taiwan, Hong Kong, and China. Although beliefs about gender parity have altered tremendously in Chinese societies since ancient times, the present results disclose that traditional gender-role beliefs remain encoded in some adolescents’ belief systems, which influence attitudes justifying violence and actual violent behaviors in adolescents’ intimate relationships. More research and prevention programs are definitely needed in Chinese societies to combat future relationship violence among adults by preventing it in today’s adolescents.

Footnotes

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research project was approved and funded by the National Science Council of Taiwan (NSC 96-2412-H-002-009) and National Taiwan University.

References

Banyard

V. L.

Cross

(2008). Consequences of teen dating violence. Violence Against Women, 14, 998-1013.

Barter

(2009). In the name of love: Partner abuse and violence in teenage relationships. British Journal of Social Work, 39, 211-233.

Chan

K. L.

Straus

M. A.

(2008). Prevalence and correlates of physical assault on dating partners: A comparison between Hong Kong and the U.S. samples. Open Social Science Journal, 1, 5-14.

Foshee

V. A.

Benefield

T. S.

Ennett

S. T.

Bauman

K. E.

Suchindran

(2004). Longitudinal predictors of serious physical and sexual dating violence victimization during adolescence. Preventive Medicine, 39, 1007-1016.

Harper

D. W.

Voigt

(2007). Homicide followed by suicide: An integrated theoretical perspective. Homicide Studies, 11, 295-318.

Hickman

L. J.

Jaycox

L. H.

Aronoff

(2004). Dating violence among adolescents: Prevalence, gender distribution, and prevention program effectiveness. Trauma, Violence, & Abuse, 5(2), 123-142.

Huang

T. Y.

Wang

W. C.

(2005). Development of the coping strategy of the Adolescent Dating Conflict Inventory. Journal of Education & Psychology, 28, 469-494. (in Chinese)

Jackson

S. M.

(1999). Issues in the dating violence research: A review of the literature. Aggression and Violent Behavior, 4, 233-247.

Lewis

S. F.

Fremouw

(2001). Dating violence: A critical review of the literature. Clinical Psychology Review, 21(1), 105-127.

10.

Mahlstedt

D. L.

Welsh

L. A.

(2005). Perceived causes of physical assault in heterosexual dating relationships. Violence Against Women, 11, 447-472.

11.

Miles

(2005). Ending violence in teen dating relationships: A resource guide for parents and pastors. Minneapolis, MN: Augsburg Books.

12.

O’Keefe

(1997). Predictors of dating violence among high school students. Journal of Interpersonal Violence, 12, 546-568.

13.

O’Keefe

(1998). Factors mediating the link between witnessing interparental violence and dating violence. Journal of Family Violence, 13(1), 39-57.

14.

Pflieger

J. C.

Vazsonyi

A. T.

(2006). Parenting processes and dating violence: The mediating role of self-esteem in low- and high-SES adolescents. Journal of Adolescence, 29, 495-512.

15.

Reitzel-Jaffe

Wolfe

D. A.

(2001). Predictors of relationship abuse among young men. Journal of Interpersonal Violence, 16, 99-115.

16.

Rudman

L. A.

Glick

(2008). The social psychology of gender: How power and intimacy shape gender relations. New York, NY: Guilford Press.

17.

Sears

H. A.

Byers

E. S.

Price

E. L.

(2007). The co-occurrence of adolescent boys’ and girls’ use of psychologically, physically, and sexually abusive behaviors in their dating relationships. Journal of Adolescence, 30, 487-504.

18.

Shen

A. C.-T.

(2008, March). Coping strategies of dating violence: A national study of Taiwanese college students. Paper presented at the International Family Aggression Society Conference, Preston, UK.

19.

Silverman

J. G.

Raj

Mucci

L. A.

Lorelei

Hathaway

J. E.

(2001). Dating violence against adolescent girls and associated substance use, healthy weight control, sexual risk behavior, pregnancy, and suicidality. Journal of the American Medical Association, 286, 572-579.

20.

Simon

T. R.

Miller

Gorman-Smith

Orpinas

Sullivan

(2010). Physical dating violence norms and behavior among sixth-grade students from four U.S. sites. Journal of Early Adolescence, 30, 395-409.

21.

Smith

P. H.

White

J. W.

Holland

L. J.

(2003). A longitudinal perspective on dating violence among adolescent and college-age women. American Journal of Public Health, 93, 1104-1110.

22.

Spence

J. T.

Hahn

E. D.

(1997). The Attitudes Toward Women Scale and attitude change in college students. Psychology of Women Quarterly, 21, 17-34.

23.

Spence

J. T.

Helmreich

R. L.

(1978). Masculinity and femininity: Their psychological dimensions, correlates and antecedents. Austin: University of Texas Press.

24.

SPSS (2007). SPSS 16.0.1 for Windows. Chicago, IL: SPSS Inc.

25.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The Conflict Tactics Scales. Journal of Marriage and the Family, 41, 75-88.

26.

Straus

M. A.

(2008, March). Prevalence and effects of mutuality in physical and psychological aggression against dating partners by university students in 32 Nations. Paper presented at the International Family Aggression Society Conference, Preston, UK.

27.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.

28.

Sugarman

D. B.

Hotaling

G. T.

(1998). Dating violence: A review of contextual and risk factors. In Levy

(Ed.), Dating violence: Young women in danger (pp. 100-118). Seattle, WA: Seal Press.

29.

Weng

C. Y.

Lin

I. M.

Lue

B. H.

Chen

H. J.

Y. C.

Cheng

Y. R.

(2008). Development and psychometric properties of the Chinese Hostility Inventory: Short form. Psychological Testing, 55, 463-487. (in Chinese)

30.

Yick

A. G.

Agbayani-Siewert

(2000). Dating violence among Chinese American and White students: A sociocultural context. Journal of Multicultural Social Work, 8 (1/2), 101-129.