Unintended Effects OF A Domestic Violence Campaign PDF

Preview

Summary

This paper reports findings from a media intervention designed to raise awareness about domestic violence and to prevent abuse. The campaign included four television advertisements, one billboard, a poster, and newspaper ads. A quasi-experimental, in-field approach was used to assess campaign effect...

Description

UNINTENDED EFFECTS OF A DOMESTIC VIOLENCE CAMPAIGN Sarah N. Keller, Timothy Wilkinson, and A. J. Otjen ABSTRACT: This paper reports findings from a media intervention designed to raise awareness about domestic violence and to prevent abuse. The campaign included four television advertisements, one billboard, a poster, and newspaper ads. A quasi-experimental, in-field approach was used to assess campaign effects. The study analyzes the effects of campaign exposure on awareness, perceived threat, and efficacy. Results obtained using analysis of covariance indicate that male and female attitudes were significantly different following campaign exposure. Women (but not men) increased their perceived severity of domestic violence and awareness of services from pretest to posttest. The campaign appeared to have effectively reached women. By contrast, male attitudes moved in the opposite direction from pre- to posttest, or did not change. A comparison of means confirmed these findings and also showed that women increased their belief in the effectiveness of services. The authors draw comparisons between male and female reactions and offer some suggestions for new strategies to address audience beliefs about domestic violence.

A nascent body of literature has documented the occurrence of unintended effects in health communication and, recently, a great deal of attention has been paid to the potential for audiences to react unpredictably to fear appeal messages (Cho and Salmon 2007; Witte, Meyer, and Martell 2001). Specifically, Witte, Meyer, and Martell (2001) propose that audiences who feel unable to change their current behaviors may respond with denial, defense-avoidance, or reactance (undermining a messenger) to health messages that elucidate the risks of their current behavior in an attempt to promote health behavior change. Simultaneously, a growing body of literature documents the gender-specific nature of domestic violence with recommendations leading toward tailoring interventions by gender. This study describes the development of a campaign designed to increase the perceived threat of abuse and the corresponding reactions of men and women. Given previous research indicating widespread audience denial of abuse, the campaign described here focused on increasing the public’s perception of domestic violence as a severe community problem. Offenders frequently minimize the severity of their offense or completely deny incidents of abuse. Most (83%) convicted offenders in one study felt that their situation was blown out of proportion and that police made the incident sound worse than it really was (Henning, Jones, and Holdford 2005). Health educators have long indi-

cated that increasing an audience’s perception of a health risk is the first step in promoting preventive action (Witte, Meyer, and Martell 2001). In spring 2006, students and faculty at a state university introduced the “Open Your Eyes” mass media campaign to increase perceived severity of abuse, perceived response efficacy of domestic violence hot lines, and awareness of services (Keller and Otjen 2007). Because the campaign used negative and highly emotionally charged ads and employed gender stereotypes to confront a highly sensitive issue, it might have triggered some backlash reactions. One purpose of this paper is to look at the different effects of the campaign on men and women and to explore why such differences might have occurred. A second purpose is to examine the overall effectiveness of the campaign using components of the Health Belief Model as an analytical tool (Becker et al. 1977). First, we review prior domestic violence interventions, unintended effects of health communication and relevant behavior change communication theories. Next, we describe the campaign project and outline results. This is followed by a detailed discussion of findings, the drawing of conclusions, acknowledgment of limitations, and suggestions for future research.

Sarah N. Keller (Ph.D., University of North Carolina at Chapel Hill) is an associate professor in the Department of Communication and Theatre, Montana State University Billings.

Domestic Violence Interventions

Timothy Wilkinson (Ph.D., University of Utah) is a professor at the College of Business, Montana State University Billings. A. J. Otjen (Ph.D., University of Missouri at Kansas City) is an associate professor at the College of Business, Montana State University Billings.

BACKGROUND

Domestic violence interventions typically take one of three primary forms: shelters for survivors, hot lines for survivors and perpetrators, or information services for both survivors and perpetrators (Gosselin 2010). Responses to domestic violence have focused to date primarily on interventions after a problem has been identified and harm has occurred. A particular Journal of Advertising, vol. 39, no. 4 (Winter 2010), pp. 53–67. © 2010 American Academy of Advertising. All rights reserved. ISSN 0091-3367 / 2010 $9.50 + 0.00. DOI 10.2753/JOA0091-3367390404

54 The Journal of Advertising

challenge for those working in domestic violence prevention involves finding strategies for informing both survivors and perpetrators about available services. Often, neither audience is reached until the issue has entered the legal system. There are, however, new domestic violence prevention strategies emerging, and prevention approaches from the public health field can serve as models for further development of these strategies (Wolfe and Jaffe 1999). Among these are social marketing campaigns aimed at reaching both perpetrators and victims of abuse in order to educate individuals about the severity of the issue and the availability of services. The goal of these campaigns is to empower victims to escape abusive situations and to encourage perpetrators to seek assistance in changing their behaviors. Through public service announcements and advertisements, such campaigns typically provide information regarding warning signs of domestic violence as well as community resources for victims and perpetrators. A review of intimate partner violence prevention programs from 1991 to 2001 found that most were clinic based, and of the few interventions involving public education, few had been systematically evaluated. Most interventions targeted at women attempted to increase physician screening for victims or refer clients to counseling and shelters. Of the interventions designed for male batterers (either alone or with partners), not many have been associated with effective outcomes (Wathen and MacMillan 2003). One comprehensive public education campaign, developed by the Family Violence Prevention Fund (FVPF) in collaboration with the Advertising Council, employed television advertisements to deliver the message that there is no excuse for domestic violence, and to make referrals to local domestic violence services. Results of a telephone survey evaluation from 1994 to 1996 showed significant decreases over the two years in the number of people who said they (1) did not know what to do about domestic violence, (2) did not believe it was necessary to report incidences of domestic violence, (3) felt that it was no one else’s business when a husband beats his wife, and (4) believed that the media exaggerated the problem of domestic violence (Wolfe and Jaffe 1999). Unintended Effects of Health Communication Health communication campaigns, as outcomes of and inputs into the social process, can create both unintended and intended effects (Cho and Salmon 2007). Starting with the notion of noise (Shannon and Weaver 1949), communication scholars have long been mindful of the potential of a communication outcome to deviate from intention. Campaign designers, as senders of messages, can only control inputs while outcomes are influenced by multiple forces (Schramm 1961). Westley and MacLean (1957) observed that, unlike interpersonal settings, mass communication limits feedback

from the receiver, minimizing opportunities to detect audience reactions. Here, we present a typology of potential unintended effects of health communication and then discuss the dissonance and boomerang effects that appear to have occurred in our advertising campaign. Unintended or Excluded Audiences It is typically recommended that evaluators of health communication campaigns assess changes only in intended audiences. Doing so, however, may not provide a complete or objective understanding of campaign effects. In fact, the third-person effect hypothesis (Davidson 1983) is based on this very phenomenon. Confining evaluation to the scope of the planner’s intended audience will not provide a proper understanding of the function that health communication campaigns play in society. Messages intended for a subset of the population might produce misunderstanding among members of the nontargeted subset. For example, breast cancer prevention messages emphasizing the need for women with a family history of breast cancer to have regular mammography created a false sense of security among women who did not have a family history (Lerman et al. 1990). Health communication campaigns, perhaps due to their manifest intent to promote social good, have been underexamined for their potential to cause undesirable effects. Boomerang Effects Health messages have been associated with boomerang effects. Research has reported that exposure to fear appeals resulted in intentions to increase smoking (Rogers and Mewborn 1976), drinking (Kleinot and Rogers 1982), and a self-reported increase in unsafe sex (Witte 1992). Witte, Meyer, and Martell (2001, p. 27) suggest that when a person’s perceived threat begins to exceed their perceived sense of self-efficacy, the person will focus on how frightened he or she feels and try to eliminate fear through denial (e.g., “It’s not that big a problem” or “It’ll never happen to me”), defense-avoidance (e.g., “This is just too scary; I’m simply not going to think about it”), or reactance (e.g., “They’re just trying to manipulate me; I’m going to ignore them”). Culpability Health messages may also increase an individual’s guilt about a poor health behavior, because most health messages are based on the presumption that individuals have the potential to change their behaviors. Individuals taking action to prevent and manage their risk might be celebrated. Individuals who feel incapable of changing their risk behaviors may experience an elevated sense of guilt, which may, in turn, reduce their

Winter 2010 55

ability or motivation to change (Guttman 1997; Minkler 1999; Ryan 1976). Models of behavior change postulate that in addition to the motivation to change, individuals must also have the internal resources to carry out the recommended behavior (Ajzen 1991; Strecher et al. 1986). Therefore, when individuals desire to change but lack the needed abilities or necessary environmental supports, they can end up in psychological distress (Festinger 1957). This intransigence may be particularly relevant in circumstances of domestic violence that typically occur in long-term, enmeshed psychological and family relationships that are not easily ended. Johnson and Leone (2005) describe two types of abuse perpetrators: the Intimate Terrorism abuser is typically male, very controlling, and resistant to change, whereas the Situational Couple abuser could be either gender, wants to change, and can even be treated. Men typical of the Intimate Terrorists described by Johnson and Leone would likely experience feelings of dissonance when confronted with messages urging prevention of domestic violence. Such dissonance could lead to boomerang effects. Social Reproduction Health communication may also reinforce existing social distributions of knowledge, attitudes, and behaviors. An evaluation of the television program All in the Family indicated that the program, designed to dispel ethnic biases, reinforced existing biases among some viewers (Vidmar and Rokeach 1974). Similarly, the Open Your Eyes campaign might have simultaneously educated those who were already aware of the need for prevention and education while reinforcing stigmas among those who were in favor of the status quo, and possibly antagonizing those who perceived the campaign as anti-male. Social Norming Campaigns, via social norming, can render individuals vulnerable to shame and isolation. Social norms demand compliance and forge conformity. A person who is out of alignment with implicit and explicit social expectations becomes even more vulnerable to shame and isolation in a campaign atmosphere (Piers and Singer 1953). For those who are able to adapt to majority social norms, normative messages may lead to a healthier lifestyle. Those who cannot or will not conform are marginalized; the attitudes of the nonmarginalized toward the marginalized can become increasingly negative (Goffman 1963). Gender-Specific Responses Only a few studies look at gender-specific responses to advertising. Ford and LaTour (1996) present a model in which

purchase or boycott intention is a function of both gender role resistance and negative corporate image. A comparison can be made with this study insofar as purchase intent can be equated with increased awareness (or use) of domestic violence services, and negative corporate image can be equated with the intransigence of commonly held domestic violence–related myths, which the campaign attempted to debunk. Studies of gender portrayals in advertising rarely link gender portrayals to gender-specific effects. Women are typically shown in advertising as passive, submissive, deferential, unintelligent, shy, dreamy, gentle, likely to be manipulated, and helpless. In contrast, men have often been portrayed as constructive, powerful, dominant, autonomous, and achieving (Artz, Munger, and Purdy 1999; Browne 1998). Sexist advertising has been associated with sexual harassment, violence against women, negative self-evaluations, distorted body images, eating disorders, and stereotyped perceptions of, and behavior toward, men and women (Gulas and McKeage 2000; Lavine, Sweeney, and Wagner 1999). Audiences that frequently view television portrayals of violent sex, objectified women, and irresponsible men may gradually become cultivated to adopt similar beliefs about sex in the real world. This recognition that popular media may serve as a poor vehicle for sexual socialization should not prevent media from being used as a positive health educator. However, if gender stereotypes in advertising increase stereotypical behaviors, then it is possible that media depictions of men as violent and women as provocative or submissive may encourage abusive behavior in real life. THEORY AND HYPOTHESES The goals of many domestic violence awareness campaigns, including the one discussed here, are to call attention to the severity of abuse, increase perceived susceptibility to the risk of abuse, educate potential victims about support services, and change public acceptance of myths that tacitly blame victims or deny abuse. Similar beliefs are laid out by the Centers for Disease Control and Prevention as important targets for public policy, domestic violence prevention, and public health campaigns (CDC 2006). In the current campaign, attitudinal objectives were based on both formative research and what has been called the “grandmother” theory of health behavior change (Becker et al. 1977). The Health Belief Model (HBM) suggests that individuals weigh the potential benefits of a recommended response against the psychological, physical, and financial costs of an action before considering change. Becker et al. (1977) believed that a combination of perceived susceptibility (e.g., “I am at risk for X”) and severity (e.g., “X is a serious problem that causes harm”) provides the motivation for action, while the perceived cost/benefit ratio provides the pathway for action.

56 The Journal of Advertising

Similarly, the Heuristic-Systematic Model (HSM) postulates that individuals are more likely to change their behaviors in response to a message when their comprehension of an issue is high (Eagley and Chaiken 1984). According to HSM, comprehension mediates persuasion and can influence persuasion by its impact on attention to and comprehension of message content. Systematic processing is conceptualized as an analytic orientation in which people take all (or most) relevant information into account in forming their attitudinal judgments. It implies, as does the Elaboration Likelihood Model’s “central route,” that attitudes are mediated by recipients’ understanding and cognitive elaboration of incoming information (Petty and Cacioppo 1986). Building on this theoretical background, the campaign was designed to increase audience understanding of the threat and prevalence of domestic violence, as well as increase belief in the effectiveness of emergency response services (e.g., hot lines and services). In this paper, both perceived severity of a health risk and perceived response efficacy of a solution are conjectured to contribute to comprehension. Although the rates of abuse are high, the percentage of victims who access support services is consistently reported to be low (Carlson and Worden 2005). Of 240 respondents in this study who reported knowing someone who was assaulted, only 28 (11%) reported contacting local domestic and sexual violence centers. Health educators have also made clear that audience awareness of services is a preliminary step to increasing use. Therefore, domestic violence hot line numbers were referred to in every ad (Appendix 1). This leads to the first hypothesis:

Witte, Meyer, and Martell (2001), who built on HBM, argue that the greater the threat perceived, the more motivated individuals are to evaluate the efficacy of the recommended response. Witte, Meyer, and Martell argue that it is essential for the threat to be counterbalanced with a strong positive message that will instill in audiences a belief that there is an effective solution. For people who feel unable to change their behaviors, perceptions of threat become perceptions of fear (Witte, Meyer, and Martell 2001). At such a point, people shift to a fear control process, bypassing all thoughts of the threat. Thus, a repeated theme of the campaign was to recommend that people call a hot line, with the intention of increasing audience perceptions that an effective solution was available. The concept of response efficacy (e.g., “Will I get accurate and useful information from calling a domestic violence hot line?”) was promoted through cues to action for hot lines and one television ad that demonstrated a phone call to services. Therefore, we hypothesize that: H3: Campaign exposure increased beliefs about the response efficacy of domestic violence services.

According to HBM, individuals appraise hazards by determining whether they think the threat is severe and whether they think they are susceptible to it. Extensive denial of abuse uncovered by the formative research conducted for this campaign (focus groups with men and women and in-depth interviews with survivors) further prompted campaign organizers to focus on increasing public recognition of the severity of domestic violence (Henning, Jones, and Holdford 2005). Message comprehension, which, in this case, is partly comprised of perceived severity of a threat, is considered to be an important prerequisite for systematic processing (Eagley and Chaiken 1984). Greenwald and Leavitt (1984) suggest that persuasive impact and attitudinal effects are greater when message analysis, comprehension, and motivation for systematic processing are high. Thus, this campaign specifically targeted audience perceived severity and comprehension of the threat. We hypothesize that:

Many scholars have argued that domestic violence results from larger gender inequalities in society. Viewed through this lens, the perpetration of abuse by men against women can be conceptualized as the interpersonal enactment of larger cultural forces (Connell 1995). In particular, the physical and economic power of men, achieved through a gendered division of labor, combine with women’s powerful emot...