Abstract
This study examines the relationships between exposure to violence in the community, school, and family with dating violence attitudes and behaviors among 175 urban African American youth. Age, gender, state support and experiences with neglect, school violence, and community violence were the most significant predictors of acceptance of dating violence. Experiences with community violence and age were important predictors of dating violence perpetration and victimization. Findings highlight the importance of planning prevention programs that address variables affecting attitudes and behaviors of high-risk youth who have already been exposed to multiple types of violence.
Teen dating violence (TDV) among African American teens has been identified as a serious public health problem (Reed, Silverman, Raj, Decker, & Miller, 2011), especially among low-income, urban African American youth (Wilson, Woods, Emerson, & Donenberg, 2012). Studies identify divergent prevalence even among large or national studies (see review by Henry & Zeytinoglu, 2012) primarily because of differences in definitions of dating violence, method of data collection, or subject selection. For example, in a nationally representative sample recruited by telephone, Wolitzky-Taylor et al. (2008) identified an overall prevalence of “serious” dating violence (physical or sexual assault) of 1.6% with girls reporting about 3 times more dating violence than boys. However, in a national sample of more than 6,000 high school girls, Silverman, Raj, and Clements (2004) found an overall prevalence of 9.8% and a prevalence of 18% among those girls with prior sexual intercourse. Prevalence for experiencing any dating violence increased to more than 50% and 20% of the teens reported physical violence from a high-risk (urban, predominantly ethnic minority) sample when data collection was limited to those teens with a romantic partner in the past year (Lormand et al., 2013). Ethnicity is another important consideration in evaluating the prevalence of TDV. Data from the Center for Disease Control and Prevention (CDC) report rates of TDV that are highest among African American students (12.4% of boys and 11.8% of girls) closely followed by Hispanic students (CDC, 2011).
While multiple authors report that dating violence greatly affects youths’ physical and psychological well-being (Bonomi, Anderson, Nemeth, & Rivara, 2013; Reed et al., 2011; Silverman, Raj, Mucci, & Hathaway, 2001), experiencing TDV in high school is also linked to increased risk of re-victimization through intimate partner violence (IPV) later in life (Smith, White, & Holland, 2003). Wilson et al. (2012) found that for African American girls, cumulative exposure to violence increased their risk of experiencing multiple forms of victimization. Although prevalence rates may vary, Finkelhor, Turner, Ormond, and Hamby’s (2009) national study found that youth exposed to more than one type of violence had significantly higher risk of other violence exposures. Moreover, research has also found that previous exposure to violence increases the likelihood of TDV (Glass et al., 2003; Wilson et al., 2012). O’Keefe (1997a, 1997b) found that aggressive dating behaviors for both male and female high school students were significantly predicted by family, school, and community violence; likewise, Klein (2006) reported a connection between TDV and school violence. Perhaps it should not be surprising that various forms of violence are related and that exposure to violence may produce desensitization to violence (Gaylord-Harden, Cunningham, & Zelencik, 2011), which ultimately leads to an increase in other forms of violent behaviors (Ng-Mak, Salzinger, Feldman, & Stueve, 2002). Despite these connections, little is known about how experiences with family, school, and community violence lead to increased risk of various types of TDV.
This research examines the ways in which TDV attitudes and behaviors are related to various forms of violence exposure. It specifically examines how African American youths’ experiences with family (child abuse and neglect), school, and community violence relate to their attitudes about TDV and their victimization and perpetration of TDV. Based on prior research (O’Keefe, 1997b; Reed et al., 2011), we hypothesized that greater exposure to violence at home (abuse/neglect), at school, and in the community would lead to more accepting attitudes about TDV and more TDV behaviors (victimization and perpetration).
Literature Review
Child Abuse and Dating Violence
Many studies with predominately European American adolescents have found that exposure to abuse as a child is a significant predictor of future TDV perpetration and victimization (Brendgen, Vitaro, Tremblay, & Wanner, 2002; Gover, Kaukinen, & Fox, 2008; Wolf & Foshee, 2003). For example, Duke, Pettingell, McMorris, and Borowsky (2010) found that adolescents (5% African American), who had one or more adverse family violence experiences in childhood, including physical or sexual abuse or witnessing abuse among family members, were at an increased risk of perpetration of IPV, TDV, bullying, and other aggressive experiences. “For every unit increase in the adverse-events score (additional type of adverse event reported), the risk of violence perpetration increased 35% to 144%” (Duke et al., 2010, p. e784). Wolf and Foshee (2003) also found a strong correlation between male adolescents’ (18% African American) experiences with child abuse (hit by an adult in the home with the intention to hurt) and their perpetration of TDV. However, the authors found a weak correlation between experiences of abuse as a child and perpetration of TDV for females.
School Violence and Dating Violence
Research is beginning to examine the connection between school violence and TDV. Although O’Keefe (1997a) found that exposure to school violence was a significant predictor of aggressive behaviors for both male and female high school students, she did not find that exposure to school violence was a significant contributor to TDV (O’Keefe, 1997b). However, in a study of 765 adolescents (42.7% African American), Schnurr and Lohman (2008) found that early exposure to and the perception of unsafe school environments coupled with exposure to violence at home increased the likelihood of perpetrating TDV among African American males. In addition, adolescents who identified with anti-social peers were more likely to perpetrate TDV regardless of ethnicity or gender (Schnurr & Lohman, 2008).
Espelage and Holt (2007), studying a sample of 7th- to 12th-grade students, reported that bully-victims (those who bullied others and were themselves victimized) reported more physical dating violence, peer sexual harassment, and emotional abuse than those students not involved in bullying. Fredland (2008) theorized that bullies, whose interactions are characterized by aggressive and dominant behaviors in peer relationships, are likely to bring the same type of interactions to romantic relationships.
Peer acceptance of violence is an important risk factor for dating violence (Klein, 2006). Schnurr and Lohman (2008) found that adolescents who identify with anti-social peers are more likely to perpetrate TDV, regardless of ethnicity. Brendgen and colleagues (2002) also found that affiliation with aggressive friends was related to a more positive attitude toward violence and, subsequently, to more frequent use of violent behavior in dating-related contexts. Vezina and Hebert (2007) report that youth who have friends exposed to violence are more likely to approve of and experience violence in their own relationships.
Community Violence and Dating Violence
Community violence is a standard term used to express violence that has been inflicted on a member or members of a community (Steinbrenner, 2010) and research finds that adolescents, especially African American adolescents, experience much violence outside the context of the home and school from neighborhood or “community” activities. In a study of a representative national sample of youth (N = 3,053), Finkelhor et al. (2009) found that 28.7% of adolescents reported experiencing community violence over their lifetimes; 42.2% of the older teens reported witnessing community assault in the previous year (Finkelhor et al., 2009). Wilson’s et al. (2012) study of 266 African American girls’ experiences with violence reported that 68% had experienced community-based violence; Gaylord-Harden and colleagues (2011) reported that 95% of African American youth in their study had been exposed to more than one crime.
Experiences of community violence negatively affect youths’ behaviors. O’Keefe’s (1997a) study of 935 high school students (13% African American) found that males, but not females, exposed to high levels of community violence had increased behavioral problems. Similarly, Sams and Truscott’s (2004) study with adolescent high school males (68% African American) found that as levels of exposure to violence in the community increased, so did adolescent violent behavior.
When looking specifically at TDV, studies suggest a relationship between community violence and TDV. Malik, Sorenson, and Aneshensel (1997) found exposure to weapons and violent injury in the community was the sole consistent predictor of TDV perpetration and victimization. TDV victimization and perpetration overlapped. Gender, generally, was an important variable related to community violence, but not dating violence, in that boys were more likely than girls to perpetrate and be victims of community violence, but girls were more likely than boys to perpetrate physical dating violence (Malik et al., 1997). Reed and colleagues’ (2011) study of adolescent males (54% African Americans) also found a correlation between TDV perpetration and community violence experiences.
Demographic Characteristics and Dating Violence
African American youth may not define some of their experiences with violence as violent. Fredland et al. (2005) suggest that African American youth view the violence they see in adults’ lives as “matter of fact” and as “normal neighborhood activities.” Decker, Raj, and Silverman (2007) found that young African American women were less likely to define forced or coerced sex as violence. McIntyre’s (2000) qualitative research on inner-city, middle school youths found that they did not consider incidents of pushing and shoving violent but rather regarded these behaviors “as a way of jostling for position” (p. 69). The teens often discussed violence with similar vocal tones as they used for conversations about sports, music, television programs, and social events. McIntyre concluded that youth seamlessly connected violence to other activities in their lives, and that this connection contributed to “normalizing violence and desensitizing the participants” (p. 74). In a study examining African American and Iraqi youths’ perceptions of violence, Black et al. (2009) found that Iraqi youth identified significantly more behaviors as violent than African American youth. Forced sexual contact was the only significant gender difference found. Surprisingly, males were more likely than females to define the behavior as violence.
African American culture holds to a tradition of privacy (Weisz & Black, 2008). In fact, in their study of African American youth, Weisz and Black (2008) found that adolescents often indicated that TDV is a private matter and that they would not intervene if they witnessed an incident of TDV. However, boys in the study were less likely to intervene than girls (Weisz & Black). Black and Weisz (2003) also found that African American youth were reluctant to turn to law enforcement for help with TDV; however, girls were slightly more likely than boys to speak with law enforcement about abuse. Both girls and boys were more willing to approach family members over formal helping resources (i.e., hotlines, preacher, and teachers; Black & Weisz, 2003).
Although TDV occurs in a wide range of socio economic strata (SES), some studies have found higher rates of dating violence in low SES groups (Makepeace, 1987; Sigelman, Berry, & Wiles, 1984). Economic stress and poverty are viewed as important risk factors in domestic violence (for example, see Stith, Smith, Penn, Ward, & Tritt, 2004) but are not viewed as a consistent risk factor in the dating violence literature (O’Keefe, 1997b).
The literature suggests that youths’ exposure to violence in the family, school, and community is associated with TDV experiences. However, we do not know how African American youths’ experiences with the various forms of violence relate to TDV because most studies include limited numbers of African American youth. This study addresses the following questions among a sample of African American youth: (a) Do youths’ experiences with child abuse, school violence, and community violence predict their attitudes about TDV? (b) Do youths’ experiences with child abuse, school violence, and community violence predict their victimization and perpetration of TDV?
Method
To assess how African American youths’ experiences with child abuse and school and community related to their attitudes about TDV and their victimization and perpetration of TDV, we utilized data from a larger study examining the differential effects of cumulative violence and trauma exposure among African American and Iraqi refugee youth. Findings from the African American youth are included here.
The researchers obtained approval from the university’s Human Investigative Committee (HIC) prior to recruitment. Participants were recruited at school and community health fairs that were overwhelming attended by African American youth and adults. All youth who were between 11 and 17 years of age were invited to participate. A member of the research team obtained parental and youth consent at the time of recruitment. Parents were provided with detailed information on the purpose, methods, content, and potential risks of research and provided the opportunity to ask questions prior to providing consent.
Each participant took part in a day-long “retreat” at the university, which included 5 hours of data collection, interspersed with educational and recreational activities. In addition to enjoyable activities, prizes, breakfast, and lunch, the adolescents received snacks throughout the day to maintain interest and reduce participant burden. All participants received a tee-shirt and a $30 gift card at the end of the day. Participants were provided with transportation to and from the retreat. Two weekend retreat days were held within a 2-month time period in 2004 at a large urban university located in the Midwestern part of the United States. Teens were randomly assigned to one of five data collection stations that had questionnaires and assessments about trauma, various forms of violence (including TDV), social support, school safety, puberty, alcohol and drug use, mental and physical health, and demographics. The groups rotated through each station at 75-minute time intervals. Thus, the teens completed the questionnaires in varying order. Data related to community, school, family, and dating violence are reported in this study.
Data collection stations were set up throughout various buildings on campus and each small group was assigned a “group guide” (e.g., graduate or undergraduate students) who moved teens from one station to another. To insure that all teens participated in all stations, each participant received a 3 × 5 index card “passport” that was stamped after completion of each of the five stations’ data collection. When participants completed the self-report surveys in each station, they placed them in sealed opaque envelopes to maintain their confidentiality.
Trained social workers and psychologists were available throughout data collection to respond to emotional reactions to the questions. The materials given to participants at the end of data collection included appropriate resources to assist if negative feelings arose later due to the nature of the research. Data collection instruments were coded with identifiers for individual participants. Once the surveys were reviewed for responses that might warrant mandated reporting of child abuse, data were entered and cleaned, and the surveys and the data file were de-identified to protect the confidentiality of responses.
Participants
The mean age of the participants in this study was 14.1 years (SD = 1.40) among the 175 African American youth (69 males and 106 females). Table 1 reports characteristics of the sample. The majority (56.1%) of the sample was in middle school (6th through 8th grade), whereas the remaining 43.9% were in high school (9th through 12th grades). No gender differences emerged by age or grade. Youth in the study had experienced challenges in school. More than 20% (25.1%) of the youth had failed a subject in school and 80.7% had also been suspended or expelled from school.
Sample Characteristics.
Note. N = 175.
Instruments
Child abuse
Child abuse was measured using the Conflict Tactics Scale–Parent Child (CTSPC). The CTSPC measures child treatment/maltreatment, and includes scales for nonviolent discipline, psychological aggression, physical assault, discipline counts, neglect, and sexual abuse (Straus, Hamby, Finkelhor, Moore, & Runyan, 1998). Among the available scales, we examined overall physical violence, neglect, sexual abuse, and overall child abuse. Overall physical violence, neglect, and sexual abuse were each calculated by summing incidents that youth experienced. Derived from the Conflict Tactics Scale (CTS), the CTSPC is a widely used and validated measure. Internal consistency reliability for this sample was computed for the Overall Child Abuse Scale (α = .67) and deemed adequate.
School violence
To assess violence in the schools, we used the Major Unsafe subscale of the Safe and Responsive Schools Survey (SRSS). The SRSS includes 45 items that assess students’ perceptions of school climate and safety. Excellent internal consistency estimates (α = .83 to .95) have been reported (Skiba & Peterson, 2002). In the analyses reported below, the 4-item Major Unsafe subscale was used as an indicator of school violence. Items on the Major Unsafe Scale include “Physical fighting or conflicts happen regularly at school” and “I have seen a knife at school (not including a cafeteria knife).” The participants rated statements on a 5-point scale (1 = strongly disagree and 5 = strongly agree). A total score was calculated with a higher score indicating more “unsafe” or less safe responding. Adequate internal consistency reliability was obtained for the Major Unsafe subscale (α = .60).
Community violence
We measured community violence using a shortened version of the Survey of Exposure to Community Violence (SECV), which measures whether a child has been a victim of, or witness to, any of 21 violence-related events (Richters & Saltzman, 1990). Assessment of violence can be divided into minor and major violence-related events. In this study, we looked at youth being a victim of violence and witnessing major violence-related events. Community violence victimization score was calculated by summing the number of acts of violence youth witnessed and were victimized by. The SECV has been widely used in studies of community violence exposure (Berman, Kurtines, Silverman, & Serafini, 1996; Fitzpatrick & Boldizar, 1993; Kliewer, Lepore, Oskin, & Johnson, 1998; Martinez & Richters, 1993). Adequate internal consistency reliability was obtained for the SECV Major Violence-Related Events subscale (α = .71).
Attitudes about TDV
Attitudes about TDV were assessed using 8 items out of the 12-item Attitudes About Aggression in Dating Situations Scale (AADS; Slep, Cascardi, Avery-Leaf, & O’Leary, 2001). The participants were asked to rate eight vignettes on a scale of 1 (strongly agree) to 5 (strongly disagree), which included both females and males perpetrating violence. For example, the participants rated the acceptability of Celina’s behavior in the vignette: “David blocks the doorway so Celina can’t leave the room. Celina shoves him out of her way.” Attitudes about TDV were calculated by summing items on the AADS subscales. Higher score values represent more acceptance of TDV. Slep and colleagues (2001) reported acceptable psychometric properties including test–retest and construct validity. Acceptable internal consistency for this modified version was obtained in the current sample (females perpetration of violence α = .67; male perpetration of violence α = .76; total perpetration of violence α = .75).
TDV behaviors
Dating violence behaviors were assessed using a 26-item scale, developed using the items contained in the CTS (Neidig, 1986; Straus, 1979). This 26-item scale includes 13 aggressive or violent behaviors. For each of the 13 aggressive behaviors, respondents were asked to report the number of times they perpetrated and were victimized by each behavior. Respondents rated the number of times they perpetrated and were victimized by each behavior on a 4-point scale (never, once, twice, or three or more times). For example, respondents rated the number of times “I threw something at my dating partner that could hurt” and the number of times “My dating partner threw something at me that could hurt.” Constructed scales from this measure were TDV perpetration and TDV victimization, and Total TDV (perpetration and victimization combined). TDV behavior scores were calculated by summing the number of experiences with victimization, perpetration, and overall. Internal consistency reliability for the current sample was very good for the Total TDV Scale (α = .87) and the Perpetrator Score (α = .79); it was adequate for the Victimization Scale (α = .69).
Demographic data
At one of the five stations, we collected data on the teens’ age and gender, as well as their families’ receipt of state financial support, including whether they received a free school lunch at school or whether their families received food stamps. If they responded positively to any of these questions (free school lunch, support check, and/or food stamps), we identified the participant as receiving state support.
Data Analyses
Checks were performed for missing and out-of-range data, and for deviations from normality. Initial analyses included an examination of descriptive statistics for all predictor and outcome measures. Following this, multivariate analyses were conducted examining the relationship between age, gender, state support (as a proxy for SES), community, school, and child abuse with attitudes about TDV (male and female perpetration of TDV) and TDV behaviors (perpetration, victimization, and total). Six individual regression analyses examined the above measures on attitudes about TDV (female perpetration of TDV, male perpetration of TDV, and total perpetrated of TDV) and on TDV behaviors (TDV perpetration, TDV victimization, and TDV total experiences).
In the regression analyses, all violence exposure variables (with the exception of community violence) and the three demographic indicators (age, state support, and gender) were correlated to examine the inclusion of any variable into the regression model. This was done for each one of the six regression equations. Any variable that was even modestly related to each of the six dating violence outcomes (p < .25) was included. All variables in each regression equation were entered simultaneously into the regression equation. Both the zero-order correlation and the “β” statistic, the standardized regression coefficient, are shown in the tables for each dating violence outcome.
Results
Descriptive Analysis
Table 1 provides sample characteristics for the 175 child participants and the proxy measures for the economic status of their families. The majority of youth (78.9%, n = 138) reported that their families received state support. More specifically, 73.7% (n = 129) youth reported receiving free lunches; 45.7% (n = 80) of the youth reported their families received food stamps, and 25.1% (n = 44) of the youth reported their families receiving welfare checks. Less than 20% (n = 34) of the youth reported that their families received all three types of state support.
Table 2 displays the descriptive results for independent variables (child abuse, school violence, and community violence) and dependent variables (TDV attitudes and TDV behavior-perpetration and victimization).
Descriptive Statistics for Community and School Violence Exposure and Child Abuse, and TDV Attitudes and Behaviors.
Note. TDV = teen dating violence.
# events.
Higher scores represent more accepting attitudes of TDV.
p < .05. **p < .01. ***p < .001.
Child abuse
More than one third of the youth (37.9%; n = 66) reported they had experienced severe physical abuse. No gender differences were found, χ2 = 1.62, p = .203. Nor were any gender differences found in the total number of abusive events experienced, t(129) = 1.54, p = .127, or overall physical violent events, t(130) = 0.85, p > .05. Six (4.4%) of the youth reported being a victim of sexual abuse and 31.1% reported being the victim of acts of neglect. Child neglect was experienced equally by boys and girls; 31% indicated yes for both gender groups (χ2 = 0.01, p = .935). Although not statistically significant, girls (7%) reported more sexual abuse that boys (0%; Wilcoxen W = 3,300, Z = −1.89, p = .058).
School violence
School violence was examined using the Major Unsafe subscale of the SRSS. Males reported more school violence-related activities than females on the SRSS (M = 2.95, SD = 1.0 for males; and M = 2.63, SD = 0.9 for females), t(161) = 2.15, p < .05, d = .34.
Community violence
Many of the youth (39.1%, n = 68) experienced and witnessed (94.8%, n = 166) acts of community violence. A significantly higher percent of males (51.5%, n = 90) than females (31.1%, n = 54) experienced community violence, χ2 = 7.20, p < .001. There were no gender differences for witnessing violence, males = 54.4% and females = 63.2%, χ2 = 1.33, p > .05. We used the Victimization and Witness scales to report one measure of community violence.
TDV attitudes
The total dating violence attitude mean score for both males and females was 3.36 (SD = 0.83) indicating moderate approval for physical TDV. Youth were more accepting of vignettes involving female perpetration of TDV (M = 3.77, SD = 0.94) than male perpetration of TDV (M = 2.94, SD = 1.08), t(173) = 9.23, p < .001, d = .70. Males (M = 3.29, SD = 1.18) were significantly more accepting of vignettes involving male perpetration of dating violence, t(172) = 3.54, p < .001, d = .51, than females (M = 2.71, SD = 0.95).
TDV behaviors
Girls reported significantly more total experiences with TDV than boys (male M = 3.04, SD = 5.91; female M = 6.41, SD = 12.19; t = 2.12, p < .05, d = .35). Females perpetrated significantly more TDV than their male counterparts (male M = 1.43, SD = 3.12; female M = 3.85, SD = 7.02; t = −2.67, p < .01, d = 44).
Multivariate Analyses
Multivariate analyses were conducted examining the relationships between age, gender, state support (as an indices of SES), and experiences with child abuse, school violence, and community violence with attitudes about TDV (male perpetration of violence, female perpetration of violence, and total perpetration of violence), and TDV experiences (victimization, perpetration, and total experiences). Results are reported in Tables 3 and 4.
Regression Analyses Predicting Attitudes About Female Perpetration of TDV, Male Perpetration of TDV, and Total Perpetration of TDV.
Note. “—” indicates that the variable was not entered into the equation (r > .25). TDV = teen dating violence.
# events.
p < .05. **p < .01. ***p < .001.
Regression Analyses Predicting TDV Victimization, Perpetration, and Total Experiences.
Note. “—” indicates that the variable was not entered into the equation (r > .25). TDV = teen dating violence.
# events.
p < .05. **p < .01. ***p < .001.
TDV attitudes
Results from the regression analyses confirmed that TDV attitudes were significantly related to teen gender and age, the family’s state support, child neglect, and community violence (see Table 3). Older teens, β = .18, p < .01, were more accepting of female perpetration of TDV than younger teens. Males reported more acceptance of male perpetration of TDV than females, β = −.30, p < .001; youth who had experienced more community violence, β = −.19, p < .05, reported more acceptance of male perpetration of TDV. Overall, teens who experienced more community violence, β = .20, p < .05, were more accepting of TDV perpetration (by males and females). Teens who reported less state support were more accepting of TDV perpetration (by males and females), β = −.16, p < .05.
TDV behavior (victimization and perpetration)
Teens who experienced higher rates of community violence were victimized by and perpetrated more acts of TDV. Community violence predicted TDV victimization, β = .39, p < .001; perpetration, β = .31, p < .01; and total dating violence experiences, β = .36, p < .001. In addition, age predicted total TDV behaviors, β = .24, p < .01, and the perpetration and victimization of TDV (see Table 4). Older teens were more likely to perpetrate TDV, β = .24, p < .01, and be victimized by TDV, β = .20, p < .05, than younger teens.
Discussion
Findings from the study help distinguish between the effects of exposure to several types of violence on TDV attitudes and behaviors. Although exposure to some types of violence affects both acceptance of violence and experiences of TDV, other variables predicted only attitudes or experiences.
We hypothesized that greater exposure to violence at home (abuse/neglect), at school, and in the community would lead to more accepting attitudes about TDV and more TDV behaviors (victimization and perpetration), and we found that African American youth in this sample experienced high levels of violence in their communities, schools, families, and relationships. Males experienced more violence in their schools and communities than females; females perpetrated and were victimized by more acts of violence in their dating relationships. For both genders, but for males more than females, TDV was an accepted social interaction, especially when perpetrated by females. Gender only predicted attitudes about male perpetration of TDV. Gender may have only predicted male perpetration because female perpetration of violence is more widely accepted by both girls and boys because the consequences of female versus male violence perpetration is often less severe (Molidor & Tolman, 1998). However, females may be less accepting of male perpetration of violence than males because they are often the victims of the violence and male peer groups often endorse a variety of justifications for male perpetration (Black & Weisz, 2004).
Youths’ experiences with violence in the community predicted accepting attitudes about TDV. These findings support prior literature (Klein, 2006; Thompson & Massat, 2005; Wolf & Foshee, 2003; Youngstrom, Weist, & Albus, 2003) in finding a relationship between increased exposure to various forms of violence and the acceptance of violence in dating relationships. As found in previous studies (Black et al., 2009; Fredland et al., 2005), African American youth may be accepting of a certain amount of violence in a relationship as normal.
Age was an important variable in the study. Experiencing community violence increases with age. Older youth were more accepting of female perpetration of TDV than younger youth. Older youth were also more likely to experience TDV victimization and perpetration. One possibility is that older youth had more time to experience and hear about multiple types of violence and, therefore, were more likely to view violence as common, acceptable behavior (Ng-Mak et al., 2002). This finding suggests that practice interventions with youth should consciously develop age appropriate responses.
Receiving state support predicted attitudes about TDV, but not youths’ experiences with TDV. A somewhat unexpected finding was that teens who reported less state support were more accepting of TDV. This analysis did not examine variables that might explain this finding. However, one possibility is that parents from low-income neighborhoods who were not receiving state aid were working long hours and were less able to supervise their teens or spend time processing violent experiences with them (Lindstrom Johnson, Finigan, Bradshaw, Haynie, & Cheng, 2013). Therefore, these adolescents might be more likely to perceive many kinds of violence as normal, including TDV (Ng-Mak et al., 2002).
There were some differences in the variables that predicted attitudes versus those that predicted TDV victimization and perpetration, supporting the literature that asserts that attitudes do not necessarily predict behavior (Glass et al., 2003; Wilson et al., 2012). Experiences with community violence were an important predictor of youths’ victimization of TDV, their perpetration of TDV, and their overall experiences with TDV. Age was another important predictor of TDV perpetration and victimization, with older youth having more combined TDV victimization and perpetration experiences than younger youth. This finding should not be surprising, as older youth have had more time than younger youth to experience relationships and violence within those relationships. Findings of gender differences in risk factors are common in the TDV literature (Foshee, Linder, MacDougall, & Bangdiwala, 2001; Renner & Whitney, 2012). In the current study, gender only predicted attitudes about male perpetration of TDV.
Limitations and Conclusion
The major limitation of this study is its use of a convenience sample and cross-sectional analyses. It is possible that families with the most social connections were the ones in their neighborhood to learn about the study. The families had to be able to get their adolescent children to the bus pick up site and had to have enough trust in the university and the researchers to send the youth to the university for an entire day. Therefore, these youth may come from families that are not representative of the community. However, given the high rates of community violence, child abuse, school problems, and receipt of state support that these youth reported, they were far from “elite.” Considering the previous lack of research combining all of the violence exposure variables included in this study, valuable information can still be gained by studying this convenience sample.
Another limitation might result from conducting a whole day of data collection at a university. The youth might have been intimidated by the unfamiliar, academic atmosphere, and some might have quickly tired of completing surveys and answering questions. Therefore, youth might have minimized the exposure to violence that they experienced. In addition, other risky behaviors were not examined as possible correlates of the violence exposure, such as substance abuse. In addition, we used partial scales to minimize the number of instruments youths had to complete, and thus, conceptual equivalence may be questioned.
Furthermore, the requirement of parental consent could be a limitation of this study. Teen participants’ forthrightness about TDV experiences may have been lessened due to the perception that their parents might discover disclosed information. Although the participants were informed of the confidential nature of the study and implications of confidentiality, any parental involvement may have caused some teens discomfort in fully disclosing details about their dating violence exposure.
This study is an important step toward understanding the factors that best predict TDV perpetration and victimization among African American youths. Further research needs to examine whether urban African American youth are most affected by cognitions, tolerance, or traumatic emotions that might lead to accepting attitudes and increased perpetration of TDV. Research is also needed to focus on the development of TDV prevention programs that address specific risk factors for youth who have experienced other types of violence.
As prevention programming spreads to many settings that adolescents encounter, teens attending these programs may resemble those in this study—teens who have already been exposed to several different types of violence. Therefore, TDV intervention programs should address the complex effects of child abuse, school violence, and community violence. Interventions with urban teens experiencing high levels of violence and with increased risk of TDV may need to be more focused on the realities of teen life in urban settings rather than relying on more generalized interventions, which may not effectively speak to teens dealing with urban social issues.
Footnotes
Authors’ Note
The research materials related to this article can be accessed through the first and second authors.
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) declared receipt of the following financial support for the research, authorship, and/or publication of this article: The authors received a grant from the Children’s Bridge at Wayne State University to conduct the research.