Next steps in campaign strategies to reduce teen dating violence: Examining media campaigns through the lens of “boy culture”

Abstract Research frames applied to dating violence media campaigns traditionally have not emphasized teenaged boys' self-efficacy relative to the pressures they feel to conform to a social code of masculinity. Focus group research was conducted on at-risk African American teenaged boys to explore their reactions to existing dating violence campaigns directed to men and to boys in their age group. The constructs of self-efficacy and response efficacy were applied as frameworks to guide the investigation. The findings challenge traditional approaches to dating violence campaigns where the central message is based on reactive directives that instruct the target on what not to do. The results suggest that current campaigns miss the opportunity to provide desired skills that educate at-risk teen perpetrators about proactive steps for managing relationship stress. The results also support the need for developing specific scripts for modeling positive partner interaction and identify paths for building on the self-efficacy and response efficacy of the targeted audience.

ABOUT THE AUTHORS Cynthia R. Morton is an Associate Professor of Advertising at the University of Florida. Her research interest is in message strategy, communication effects, and audience persuasion in social marketing contexts. Her research is concentrated in the areas of social issue promotion and health communication, with an emphasis on shifting audience attitudes and beliefs to benefit their present state.
Troy Elias is an Assistant Professor in the Advertising sequence at the School of Journalism and Communications at the University of Oregon. His research interest is in identity and social influence, with an emphasis on race and ethnicity, environmental science communication, and communication technology. His research examines the impact of appealing to consumer's identities and its effect on American culture.
The present research is the result of an intersection between the authors' interests and a need to discover new opportunities to reach a specific media audience more effectively.

PUBLIC INTEREST STATEMENT
Campaigns to curb teenaged dating violence commonly direct messages about how to recognize or get help for relationship abuse toward teenage girls. In contrast, campaigns that might provide pro-social scripts to teenaged boys at risk of perpetrating dating violence are relatively few. While a scholarly research has substantiated the influence of family history, environment, peer pressure, and media on boys' beliefs about gender roles in dating, there is little research that investigates "boy culture"-i.e., the unwritten social rules ascribed to being an accepted member of the reference group-as an additional consideration to predict dating violence. This study examined at-risk African American teenaged boys' attitudes and beliefs about dating and their reactions to existing campaigns against dating violence. Qualitative research findings support the need for campaigns that deliver messages that not only promote pro-social behaviors in dating, but also are acceptable to boy culture.

Introduction
Teen dating violence is defined as "any physical, sexual, or psychological/emotional violence within a dating relationship, as well as stalking. It can occur electronically and may occur between a current or former dating partner" (Centers for Disease Control, 2016). Dating violence is demonstrated by a pattern of abusive behavior whereby one partner exerts a combination of power and control over the other partner (National Domestic Violence Hotline, 2017), in essence coercing the latter to submit to the whims of the former using negative reinforcement tactics. Specific categories of dating violence measured in research include physical violence, psychological abuse, sexual coercion, and cyber dating abuse (Zwieg, Dank, Lachman, & Yahner, 2013). Survey data from a 2011 CDC report (Black et al., 2011) noted that "one in three adolescent girls is a victim of abuse" at the hands of a dating partner, and approximately 9% of high school students affirmed that they have been the target of a deliberate hit, slap, or physical harm from a dating partner, which totals nearly 1.5 million high school students nationwide (CDC, 2006;Davis, 2008;Grunbaum et al., 2004).
While the topic of teen dating violence may conjure older adolescents (e.g., aged 15 to 19) acting on their perceptions of relationships, extensive research expands the scope of exploration to pre-adolescent groups aged 10 and 12 years. The CDC's (Black et al., 2011) findings about youth risk behavior are based on self-reported data from adolescents between ninth and twelfth grades. Zwieg et al. (2013) conducted a survey of over 5,600 youth in grades 7 to 12. Their findings concluded that 20.5% teens in a relationship were subjected to some form of physical dating violence, and 25.7% of those sampled indicated psychological dating abuse. Simon, Miller, Gorman-Smith, Orpinas, and Sullivan (2010) studied perceived dating norms on a sample of 5,404 sixth-grade students and found reports of physical aggression toward a boy-or girlfriend reported by both girls and boys in this age cohort.
Research associates key demographic and socioeconomic indicators with the risk of dating violence. Minority youth, particularly boys, are more likely to be associated with dating violencewhether as perpetrators or victims-than white youth, though the research is inconclusive about the indices among various racial and ethnic categories (e.g., black/African American, Asian/Pacific Islander, and Hispanic/Latino) (Ackard, Neumark-Sztainer, & Hannan, 2003;Foshee et al., 2008;Halpern, Oslak, Young, Martin, & Kupper, 2001;Henry & Zeytinoglu, 2012). Family exemplars of partner interaction and peer influence (Arriaga & Foshee, 2004;Laporte, Jiang, Pepler, & Chamberland, 2011) are among the influencers predictive of behaviors that perpetuate ongoing patterns of intimate partner violence into adulthood and across family generations (Powers & Kerman, 2006;Wolf & Foshee, 2003). Additional parent-related risk factors to adolescent dating violence risk include the absence of parental references-a father figure for males or a mother figure for females, low parental education levels (i.e., less than high school), and parent-centered family crises, such as divorce, life-threatening illness, or death (Halpern et al., 2001).

Purpose of the study
Traditionally, teen dating violence campaigns aimed at males have been framed authoritatively, directing potential perpetrators to not pursue violent solutions to domestic conflicts. This approach assumes the most basic of communications goals-awareness and exposure-are sufficient to generate pro-social outcomes and that mere attention as a metric is reasonable for determining the message's effectiveness. In fact, the assumption that targeted audiences [in this case boys and young men] will process the message and integrate its advisory into their behavior at the appropriate opportunity is incongruent with the expansive programmatic approaches to educate, if not re-socialize, at-risk males about alternative responses to violence. Yet, in the presence of alternative theoretical frameworks, the means for evaluating message opportunities relative to the world that teen boys are challenged to negotiate could result in more measurable outcomes of success. This research explores the opportunities to make campaign messages more effective among one segment of teenagers-African American male teens and tweens, aged 12 −17. In doing so, the findings seek to advance recommendations for improved message resonance and outcomes in a direction of pro-social change.

Literature review
Much has been written about boy culture, the "boy code" (Pollack, 2000) and the challenges boys face to live up to societal standards of masculinity. Wiseman (2013) notes that the socialization boys receive from society about appropriate roles for manhood is limited in scope and greatly abbreviated relative to the roles that girls are allowed. "For most boys, the goals of being ethical and honorable, while valued, are vague….His experiences in groups will influence how hard he tries in school, how he presents himself, [as well as] his level of respect for women and girls" (Wiseman,p. 26).
Personality scripts for boys suggest that they must hold themselves emotionally detached, even if it means sacrificing the esteem of others to do so (Pollack, 2000;Wiseman, 2013). Mueller, Jouriles, McDonald, and Rosenfield (2013) examined teenagers' beliefs about the acceptability of violence and concluded that respective explicit and implicit cognitions about violence play a role in justifying aggression in relationships. In this way, not only does "conscious and effortful thought" (e.g., Anderson & Bushman, 2002;Crick & Dodge, 1994;Jourlies, McDonald, Mueller, & Grych, 2012) influence their way of processing violence, but also feelings and actions grounded in experience (Greenwald & Banaji, 1995). Harding (2009) suggested that the effects of neighborhood socialization of younger adolescents by older peer role models also reinforce a complicity toward violence. Violence was found to be a strategy utilized by adolescent boys from disadvantaged neighborhoods not only to insure protection from older adolescent peers when guarding against victimization and threats but also when maneuvering romantic relationships. The study concluded that the code set for socialization dictates acceptable behaviors for all interactions, including that between males and females.
Research also acknowledges the importance of the neighborhood's collective efficacy-that is, a community's "cohesiveness and residents' willingness to intervene for the common good" (Sampson, Morenoff, & Gannon-Rowley, 2002)-on adolescents' response to dating violence, whether as perpetrator, victim, or bystander (Reed, Silverman, Raj, Decker, & Miller, 2011). Jain, Buka, Subramanian, and Molnar (2010) concluded that, for males, collective efficacy is a moderator to the risk of perpetrating dating violence in low-to mid-level disadvantaged neighborhoods; however, socioeconomically poor neighborhoods [with higher gang presence or violence in general] "increased the risks of young males perpetrating dating violence" (p. 1742). Henry and Zeytinoglu's (2012) research concludes that the effect of neighborhood socialization makes the risk of dating violence more salient for African American teens. Ironically, teen males' reactive and hyper-masculine coping strategies also could become a cause for further alienation by adults who might otherwise be in position to intervene with positive role modeling to adolescent males (Cunningham, Swanson, & Hayes, 2013).
Gender-based stereotypes about "manhood" and masculinity are made even more prevalent in the images portrayed in media (Gillum, 2002). Media socialization reinforces a forceful and physically dominant male persona that is emblematic of society's template for boys, and "support-[s] a positive association between exposure to media violence and aggression" (Escobar-Chaves et al., 2005). Manganello (2008) conducted a meta-analysis of the extant research on the relationship between media use and teen dating violence. The investigation established that all genres of media influence dating violence risk, including movies, video games, music lyrics and videos, and entertainment programming on television and proposes a framework to future researchers for standardizing the exploration across categories.
Within the context of discussion about the influence of individual cognitions, peer socialization, and neighborhood efficacy is research that substantiates the role bystanders have in reversing the pattern of teen dating violence. Baynard, Plante, and Moynihan (2004) note that a bystander intervention approach "frames sexual violence as a community issues" incumbent on all members to mitigate the opportunity for and perpetuation of sexual assault and violence. From the perspective of third parties, the choice not to intercede on behalf of the victim when they witness dating violence makes them an accomplice to the assault. McMahon (2015) examined bystander literature in a meta-analysis of 59 articles and found five major themes associated with bystander action or inaction: social norms, the prevailing sense of community, the degree of pro-social modeling available, policies, and the physical environment. The collective efficacy invested in any one or more of the factors can either mitigate or exacerbate the problem. Azjen's (1985Azjen's ( , 1991 theory of planned behavior has been widely used to explain the relationship between an individual's perceived behavioral control in a situation and the outcomes expected to result from these attitudes (see Basen-Engquist & Parcel, 1992;McCabe & Killackey, 2004;Slater, 1999). The theory posits that degree of control will range between total control and total lack of control, depending on the interaction between internal variables (i.e., skill, emotions, abilities, etc.) and external variables (i.e., situation, environmental influence, etc.). A concept central to the theory is that of self-efficacy, defined as the individual's belief in his ability to perform a particular behavior in a given situation, as well as the outcome one can expect from such behaviors (i.e., outcome expectancy) (Bandura & Adams, 1977).

Theoretical foundation
Self-efficacy has been found to mediate the interaction between the individual's knowledge and his actual performance of the behavior (Rostosky, Dekhtyar, Cupp, & Anderman, 2008) and can be attributed to a behavioral intervention's reason for success or failure. In the context of teen dating violence campaigns, the utility of campaigns that attempt to curb partner violence is evidenced in their ability to educate at-risk audiences about alternative pro-social strategies for coping with relationship stress. Research supports the premise that the coping strategies advanced will be more effective when at-risk individuals hold stronger beliefs about their ability to successfully enact such strategies (Edwards, Mattingly, Dixon, & Baynard, 2014). The at-risk target's assessment of the need for behavior modification is increased further if his response efficacy to messages promoting behavior change is also strong (Bandura, 1986). Response efficacy represents the degree to which a proposed response to a problem will be effective; response efficacy influences, and is influenced, by self-efficacy (Witte & Allen, 2000), and is associated with intentions to change behavior, as well as the behavior shift itself (Floyd, Prentice-Dunn, & Rogers, 2000).
In addition, the frame that accompanies a persuasive message can either advance or hinder response efficacy effectiveness if its application to a message fails to resonate with the audience's needs. Framing involves "[selecting] some aspects of a perceived reality and [making] them more salient in a communicating text, in such a way as to promote a particular problem definition, casual interpretation, moral evaluation, and/or treatment recommendation for the item described" (Entman, 1993, p. 52). A frame provides a schema both for presenting and for comprehending an issue (Entman, 1993). The way a message is framed in media with respect to tone (positively or negatively), voice ("us" or "them"), or lens (individual, community, or societal) can change the audience's opinions about the importance of that issue and the prioritization of merits associated with it (Gamson & Modigliani, 1989;Kinder, 1983;Scheufele, 1999).

Research questions
For teenaged boys at risk or dating violence behavior, the initial intent to change one's behavior so that it results in long-term benefit may be determined in part by the relationship between self-efficacy and response efficacy as presented in the message framing of a particular issue. When self-efficacy is high, the assessment of response efficacy to solutionbased messages is expected to be high as well; when self-efficacy is low, response efficacy toward solution-based messages is predicted to be low. The interaction between these constructs not only can influence how an individual processes messages, but also the message's effects on one's intent to adopt the message's recommended behavior. From this four research questions were advanced to guide the investigation: (1) What are teen boys' attitudes toward manhood and what criteria are used to define it? Are there expectations about gender roles?
(2) What perceptions do teen boys have about healthy and unhealthy relationships?
(3) What are their attitudes and opinions toward existing teen-targeted dating violence campaigns?
(4) What beliefs, if any, exist about bystander (or third-party) responsibilities to intimate partner violence?

Method
Qualitative research was conducted on behalf of a state-funded local domestic abuse network in the southeastern United States. Focus groups were conducted with African American teenaged boys aged 12 to 17 to investigate the research questions. Participants were recruited from a zip code in northeastern Florida, an area identified as a high-risk area based on police records that tracked incidents of domestic violence. Teen participants were recruited from two sites of the local Boys & Girls Club with assistance from the organization's area director. All participants were required to have a parent or guardian's written consent before the he could be part of the study. The incentive for research participation was $20.
A discussion guide was developed that consisted of 21 questions. The questions were framed in various formats according to their utility in generating the most elaborative responses from the focus group participants. Examples of the types of questions used to facilitate discussion include direct and factual, picture sort, grand tour scenarios, and hypothetical interaction (Davis, 2012). The question categories addressed at-risk teenaged boys' (1) perceptions of masculinity (e.g., Describe your idea of a "real man."); (2) perceived gender roles in relationships (e.g., What is the boy's role in a relationship? What is the girl's role?); (3) beliefs about their role as a bystander to dating violence; (4) general attitudes about media's influence on teens like themselves; and (5) attitudes toward specific dating violence campaigns.
The focus groups were held on-site at the Boys & Girls Club locations in conjunction with the afterschool program. Two African American adult males were trained to facilitate the discussion using the approved protocol. All groups were recorded to aid in the generation of transcripts after the discussions were completed. Each discussion was approximately an hour to 1-1/2 h in duration.

Research stimuli
Participants were asked about three specific dating violence campaigns-Choose Respect, Men of Strength, and That's Not Cool. Each of the promotional campaigns was part of a comprehensive outreach program to reduce the rate of intimate partner violence. In general, campaign promotional materials targeted all middle-and high school aged students across the United States using a media mix of posters, digital media (e.g., sponsored web sites and text messages), and video public service announcements. In contrast, the Men of Strength campaign specifically directed its message to teenaged boys and young adult men. Participants were shown poster ads from the three campaigns and asked to discuss what they liked and disliked about each.
The Choose Respect ad campaign was sponsored by the CDC and aimed its messages to adolescents aged 11 to 14. The program's purpose is to promote healthy teen relationships by educating them on pro-social relationship coping strategies, as well as about ways to prevent dating abuse before it starts (CDC, 2017). The campaign's promotional materials are archived in the Public Health Foundation's (PHF) Learning Resource Center (Public Health Foundation: Learning Resource Center, 2018). The images displayed individuals in shadow graphic form, so that the central character was indistinguishable in terms of race or ethnicity. The copy was presented in "resolution alternative" format whereby two alternatives-one negative and one positive-for resolving the problem is introduced prominently on the ad.
The Men Can Stop Rape-Men of Strength organization was founded in 1997 with a mission to advance "positive solutions to engaging men as allies" in the effort to end men's violence against women. The organization's outreach is to boys and young men aged 11 to 22 (mencanstoprape. org, 2011). Its campaign messages intentionally challenge traditional notions of masculinity and accountability to women and girls. The two ads presented in the study featured a young adult male with a female partner. The copy presented thoughts from the male's perspective and feel of the ads was bold and masculine (http://www.mencanstoprape.org/Public-Awareness/). The That's Not Cool (TNC) initiative aims to increase awareness and reduce dating violence among young people by focusing on unhealthy relationships both online and offline (www.thats notcool.com, 2001) and what youth communities can do to stop it. The TNC campaign gives specific emphasis to digital media, including social media and mobile text messaging, as tools for coercion, stalking, and harassment.

Findings
A total of 25 African American males aged 12 to 17 participated in the research. The average size of the discussion group was six participants. Focus group responses were transcribed, and themes identified within the specified theory frameworks. Saturation, defined as the point at which no new information emerged on the discussion topics, was reached across six discussion groups (Lindlof, 1995). Following is a discussion of the findings organized by topic addressed in the research.

Attitudes toward manhood & role models
Participants were asked to define what a "real man" is through a picture sort activity that asked them to select the pictures that they believed represented the term. The pictures were selected from stock photos and included in the sort represented an array of male images. The males pictured ranged in age between adolescent to young adult and illustrated males in a variety of contexts, including visuals with a solitary male, with other male peers, with female peers, in task situations (doing homework), or in home and family contexts. Participants explained their selections with descriptive words that captured their perceptions about qualities associated with real men.
The images that were consistently selected across all groups tended toward pictures that portrayed an older male with younger males-whether that is interpreted as a father with his sons or with younger siblings; images of male/female closeness; and images of hanging out with other males. The pictures of men portrayed as caring role models or caring partners, as well as pictures that displayed ease with oneself or one's peers resonated most strongly with participants. Other pictures that generated discussion showed teenaged boys in confident and independent, if not solitary, situations.
When asked to summarize in three words how they might describe a real man, words used to characterize the selections were "family-centered," "comfort," "caring," "love," "responsible," "courageous," "brave," "lively," "ordered," "positive," "pride," and "joy." Pictures that presented males as aggressive or anti-social were downgraded as masculine role models. The feedback reinforced that participants interpreted the concept of "real man" according to the responsibility a man takes on, as well as the support he shows toward his partner. The idea of edginess or hardcore traits often associated with the age group's perceptions is present in cases where it seems appropriate. Edginess that was inappropriate or nonsensical was viewed with disapproval according to the code under which the participants operated.
Overwhelmingly, participants tended to agree with the idea that music videos and music contributed to teens' perceptions about how boys should treat girls. Yet, their role models also varied widely, with participant responses ranging from parents and close family friends to television characters and celebrities. Furthermore, the traits admirable in the role models selected seemed grounded in the stability of the person's perceived lifestyle.

Gender role expectations
Participants were asked to offer their opinions about the specific role, if any, that boys and girls have in a relationship. All research participants across all discussion groups unanimously agreed that both did. Generally, the participants tended to believe the boy's role was that of protector, supporter, and even provider. Some stated the male's role in terms of what he owed to the relationship-truth, fidelity, and financial support. Participants with a more traditional vision about relationships ascribed to the idea that females should take on a supportive role in the relationship. One participant (aged 16) said, "The only job, the only role I like my girlfriend to do for me is to cook and clean and keep everything right." Equity between the male and female was preferred by a subset of participants, noting that females share as much of the financial burden [of dating] as the male and should have as much, if not more, say in the relationship.
The participants identified open communication, trust, fidelity to one another, honesty, and mutual respect as important qualities that make a relationship healthy. The group believed that poor communication (yelling, screaming, etc.), hitting, and other forms of violence were unhealthy, though they recognized that challenges were also part of relationships. In the words of one participant, "…relationships have ebb and flow, and that there will be good points and bad points."

Dating violence campaign attitudes
Focus groups were presented with ads from three dating violence campaigns. For each campaign discussed, participants were asked what they liked and disliked about the ads, as well as their opinion about their effectiveness to adolescent boys like them.

Men of strength
The headline of the ads shown for the Men of Strength campaign read, "My strength is not for hurting," and the copy reinforced men's respect for women in the context of relationships and sexual interaction. For example, one ad read, "So when I wanted her, I asked her-and took no [all previous words capitalized] for an answer." Participants who felt positively about the campaign liked the visual presentation, as well as the fact that the ad approach made boys value the girl's perspective and push back from what he might want rather than using force or abuse to get what he wants. One participant interpreted the campaign's meaning to be "a guy can't always have what he wants." Participants generally liked the headline, "My Strength Is Not for Hurting," and said the campaign would get their attention.
However, some participants expressed skepticism about the man's level of frustration if the woman actually did say no. One participant suggested that the resolution offered would create even greater tension for the couple because of the woman's refusal to be intimate. The direct link between the headline and the assertion about rape was also confusing, particularly to the younger aged participants who interpreted the ad's message to mean that the man was getting ready to commit rape, but [he] decided against it. They did not seem to understand that the two individuals in the ad were in a relationship or about the concept of consensual sex between partners. In this way, the ad's concept needed further explanation before some participants could fully grasp the copy's context. Thus, while this campaign resonated with adolescent boys based on aspired characteristics of a real man, its direct link to rape may alienate this group, which may view the negative outcome as an extreme relative to a man's intent to show his partner affection.

Choose respect
Two ads were presented to the groups from the Choose Respect campaign. One ad presented a boy sitting alone considering how to cope with anger at a dating partner. The alternative choices presented as opportunities for problem resolution were: (1) "She made me mad so I pushed her"; and (2) "She made me mad so we talked about it." A second ad from the campaign offered the alternatives of either defaming or respecting the girl's reputation.
The campaign was well received for its colorful graphics. Participants consistently commented on the visual presentation and the alternative behaviors suggested in the copy. In addition, most found it very approachable and believed that it would get the attention of other youth in their age range. Moreover, they readily comprehended the choice scenario posed in the two ads presented and liked the idea of being offered behavioral alternatives for addressing relationship problems. The groups did speculate on precursory acts that led to the featured decision, and one person suggested that communicating-whether to the female partners or to supportive male friendswas inauthentic, stating "nobody [is] going to tell their friend she's cool or something…it's none of their business if she's cool or not cool." This suggested that the act of talking openly about one's fledgling relationship to other boys might be taboo to this audience, and that response efficacy for the ad could be less effective as presented. Nonetheless, their suppositions did not detract from their favorable impressions about the ad. In contrast, the group also discussed a scenario that could go both ways.
Unlike any of the other campaigns, the Choose Respect approach framed the alternative resolutions scripts that adolescent boys could model when managing their own relationship frustrations. The feedback reinforced the fact that they could put themselves in the situation and enact the resolution as scripted in the ad. This unique aspect could also be attributed to the reason why the campaign was so positively received. The fact that the feature character was displayed using abstract graphics also avoided message rejection that might result if the source was deemed "not like me."

That's not cool
Of all the campaigns presented to the sample, the That's Not Cool ads shown were the most difficult for participants to interpret. Although they found the graphic approach and ad context relevant, the paradox implied in the copy approach led to questions and confusion. One ad presented to the groups showed a trophy and was framed with the copy, "Congrats! With that last text message you've achieved stalker status." The stalker status reference was confusing to several participants, who struggled to understand the nature of the reference. Few participants comprehended its meaning quickly and required further conversation to get its relevance. The second ad posed a rhetorical question in the headline: "When does caring become controlling?" Similarly, the fact that the ad relies on the reader to understand the use of figurative language made it less effective for this audience. However, the look of the campaign-graffiti-style visuals and big fonts-resonated youthfulness, which was received.

Bystander responsibility in dating violence situations
Participants were presented with a scenario in which they were asked to imagine being witness to an intimate partner violence event between two kids at their school and asked what they would do if they saw open abuse from a male perpetrator on a female victim. The question was presented in alternate forms to reflect differing levels of acquaintanceship between the participant and the individuals involved (i.e., complete strangers or a friend to one or both individuals).
Several participants cited the moral code that it is not appropriate to hit a girl as the reason why for their intervention. For situations where they either did not know either person, or where the male partner was a friend, the intervention would mainly consist of talking to the boy in order to get him the "chill out." However, their emotional response to the situation intensified if they were personal friends to the attacked female. Reactions to the questions reinforce the idea that adolescent boys believe it is their responsibility to intercede, mainly for the purpose of protecting the female from further abuse, whether they know her or not. One participant made it clear that his intervention was conditional; he would intervene if he were certain that he could avoid physical injury. Another participant commented that he would do nothing because the situation was "[their] business, not mine." When presented with a scenario where a close male friend was heard verbally disparaging a girlfriend, or admitting to having abused his girlfriend, participants were less committal about taking action versus when the perpetrator was a stranger. Some of the teens said they would tell their friend to stop, while others had little or no feedback to this question. The inconsistency in reactions to the latter scenario appears to substantiate research about boy code (Harding, 2009) and community efficacy (Edwards et al., 2014), where needs for inclusion and community acceptability stand as barriers to dating violence intervention. The weak feedback from participants could also suggest that the act of witnessing an intimate partner violence event is more motivating to male bystanders than a verbal recount of behavior unseen.

Conclusion
The purpose of this research was to investigate opportunities to make dating violence campaign messages more effective to African American male teenagers. The findings from this study support the findings from previous research on the tension between doing what is right and doing what conforms to perceptions about what is masculine. For the participants, "real men" and the role for males in relationships support boy world characterizations of masculinity attributed to stereotypic beliefs about men, women, and their "proper place" in relationships (Cunningham et al., 2013;Spencer, Fegley, Harpalani, & Seaton, 2004). Not only did the participants ascribe conventional attributes to the concept of what the term "real man" implied, but most also associated male and female roles in a relationship with the most traditional values. The idea of man as protector and defender of females, regardless of one's level of acquaintance, was a common theme expressed across the group, even if it meant resolving a conflict with violence. Indeed, there seemed to be a code of understanding that respect for self was indirectly related to one's respect for others.
However, the findings also suggest that teenaged boys are conflicted when placed in circumstances where layers of nuance test their reactions to dating violence situations. For some, close friendships are grounded in solidarity, and respecting the actions others, no matter what the circumstance, means minding one's own business and not getting involved. Others stand by a chivalry code, which prods the individual to intervene every time in order to keep a situation from getting out of control regardless of the circumstance, and despite the fact that such action may run afoul of in-group solidarity. Still, the idea of not risking one's own safety or security in the process of helping someone else, whether male or female, also emerged as a barrier to bystander intervention. This was reinforced when the boys were asked how they would talk to a male friend suspected of dating violence. The underlying implication suggest that, even though role modeling the idealized real man is aspirational, finding appropriate opportunities to explore these roles without threat of harm remains a challenge for this group of teenaged boys.
A clear vision exists about the criteria associated with a healthy relationship and the teens sampled could describe their ideas with words and examples. However, relationships modeled in celebrity culture through popular media genres, especially music videos, also were identified as important influences to teenaged boys who seek examples for how men should behave in relationships. The tension between aspirational relationships and those presented in entertainment media challenges teen boys' self-efficacy in situations that require them to respond to issues of conflict with their partner. The evidence suggests they lack the tools necessary to negotiate such circumstances with nonreactive coping strategies (Cunningham et al., 2013).
Though the teens expressed intrigued with advertising campaigns aimed at them, they also were skeptical of media messages that told them what they believed they already knew. The "don't" frame applied to dating violence campaigns were typical of many they had seen before, so their rationale to ignore them altogether seemed sound because in general there was no new information. In this study, participants responded favorably to ad executions that were clear in their visual presentation and copy message, as well as those that left little room for interpretation. Seeing couples portrayed implied "healthy" togetherness and seeing alternative strategies for problem resolution provided practical scripts for making healthy relationship interaction. In contrast, copy techniques that used any form of play on words (e.g., metaphor, pun, or rhetorical question) over-complicated the message in such a way that compromised response efficacy to the message among those who demonstrated high self-efficacy.

Implications for teen-targeted campaigns
The findings from the current study offer implications for the development of future dating violence campaigns, as well as for future research investigation. Campaign messages against dating violence seem to often presume that adolescents need to understand what constitutes the behavior (i.e., awareness) more than what to do when faced with such circumstances. However, this research determined a need for communication agents to advance message strategies in such a way that gives the target tools for managing the pressure to react violently, as well as scripts to better communicated with dating partners and/or at-risk peers more effectively. Clearly, one's understanding of what makes relationships healthy does not necessarily translate into the skills needed to exercise such strategies when relationship pressures arise. Yet, an increase in self-efficacy improves the target audience's response efficacy to anti-violence messages. At-risk teen boys become more interested in mimicking pro-social behaviors demonstrated in dating anti-violence messages as confidence in their ability to implement those behaviors builds. Therefore, opportunities exist to reformulate dating violence campaigns away from "don't do it" frames and toward frames that present concrete alternatives for behavior and communication with dating partners.

Limitations
A limitation of this research is that it focused on at-risk African American teenaged boys identified in a specific, small-sized metropolitan locale. The findings may not reflect African American teenaged male attitudes in large urban settings or in other geographic regions, so future research should explore the similarity in attitudes in other contexts.

Recommendations for further study
Future research is recommended to better understand the relationship between self-efficacy and response efficacy in light of the current research findings. Indeed, campaigns formulated with a focus toward socialization and/or relationship coping skills and which model pro-social scripts and behaviors for responding to stressful situations are the next step for curbing dating violence.