Media Portrayals of Female Perpetrators of Intimate Partner Violence

Abstract

Preventing intimate partner violence (IPV) is a public health priority. An important component of designing prevention programs is developing an understanding of how media portrayals of health issues influence public opinion and policy. To better understand the ways in which media images may be informing our understanding of IPV, this study content analyzed portrayals of IPV in news media articles. Stratified media outlets were used to obtain a representative sample of daily newspapers based on their designated market areas. Researchers created constructed months using weeks from each season across a 2-year period. The first part of the study investigated quantitative differences in the coverage of female and male perpetrators (n = 395) and identified several areas where coverage differed. The second part of the study qualitatively examined coverage of female perpetrators (n = 61) to provide a richer description of such coverage. This study contributes to our understanding of female perpetrators and how these portrayals may contribute to the larger gender symmetry debate surrounding female aggressors. Implications for public health policy and research are discussed.

Keywords

domestic violence and cultural contexts domestic violence perceptions of domestic violence media and violence women offenders

In June 2008, Travis Alexander was found with a slit throat, multiple stab wounds, and a shot to the head in his home in Mesa, Arizona. His ex-girlfriend, Jodi Arias, was charged with the murder. The defense claimed that Arias was a victim of a “controlling, psychologically abusive relationship” (McLaughlin, 2013) and that she acted in self-defense. However, she was found guilty of first-degree murder in May 2013. The Arias trial and related news coverage have gained widespread national interest and also reinvigorated debate about media portrayals and public perceptions of female perpetrators of intimate partner violence (IPV).

In scholarly literature, researchers have called for investigations of gender symmetry in IPV, or studies that equally consider the experiences of male and female perpetrators and victims (see Anderson, 2005). Numerous studies have shown equal or higher rates of IPV by women, and many investigations of university students have shown equal rates between men and women (Straus, 2004). Understandably, these counterintuitive findings have aroused considerable debate. Despite this, only a small number of studies have investigated female perpetrators specifically (see Straus, 1990, 2006). Moreover, it is possible that the ways in which media portray female perpetrators may be influencing how scholarly and lay audiences view this issue. As such, this article contributes to our understanding of female perpetrators of IPV by examining how they are portrayed by news media and how these portrayals may contribute to the larger debate surrounding female perpetrators.

IPV

The Centers for Disease Control and Prevention (CDC; 2013b) describe IPV as “physical, sexual, or psychological harm by a current or former partner or spouse” that can take place “among heterosexual or same-sex couples and does not require sexual intimacy.” According to the results of the 2010 National Intimate Partner and Sexual Violence Survey (NISVS), approximately 28.5% of men and 35.6% of women have experienced rape, physical violence, and/or stalking by an intimate partner in their lifetime; 13.8% of men and 24.3% of women have experienced severe physical IPV (Black et al., 2011). Although women are victims of IPV more frequently than men, the prevalence of IPV among men nonetheless represents a significant public health concern. One out of every 14 men has been assaulted by an intimate partner (National Coalition Against Domestic Violence, 2013) and men experience 2.9 million physical assaults from intimate partners each year (Tjaden & Thoennes, 2000). In 2008, IPV accounted for 5% of all violence against males in the United States and the rate of victimization against males aged 12 or older was 0.8 per 1,000 (National Data on Intimate Partner Violence, 2013). These numbers, which are drawn from the Bureau of Justice Statistics, represent a 54% decrease since 1993, which is similar to the 53% decrease in the IPV rate for women across the same time period. However, because many people do not report IPV, these numbers are considered to underestimate the true prevalence of IPV (CDC, 2013c).

IPV victimization is associated with a variety of negative consequences ranging from physical injury and disease to mental distress and psychiatric disorders. For example, approximately 20% of men who have been physically assaulted experienced injury in their most recent victimization (Tjaden & Thoennes, 2000). Moreover, physical assaults are related to increased risk of sexually transmitted infections and central nervous system disorders. Other health consequences associated with IPV victimization among men include increased incidence of chronic physical and mental illness, depression, anxiety, low self-esteem, poorer self-reported health status, and increased risk of substance use and abuse as well as social issues such as isolation from social networks (CDC, 2013a; Coker et al., 2002; Cronholm, 2006; Reid et al., 2008). Although female victims report severe consequences of physical abuse, the negative health-related effects of psychological IPV involving abuse of power and control may be just as severe for men (Coker et al., 2002). Although much of the research surrounding IPV focuses on women, men experience many of the same negative effects and, in some cases, may experience more pronounced effects in the areas of restricted access to services and isolation given the resistance of some to see men as victims of IPV. The next section provides an overview of female perpetrators of IPV and some of the unique issues facing researchers and practitioners in this area.

Female Perpetrators

A predominant number of articles exploring IPV have focused on the experiences of male perpetrators and female victims. Consequently, some researchers in this area have called for more thorough investigations of the opposite situation—female perpetrators and male victims (Straus, 2006). In recent years, research in this area has generated increased interest and debate as more than 150 research studies have shown equal or higher rates of IPV by women (Straus, 2004). In addition, some researchers have argued that police statistics show higher incidences of male than female perpetrators because they reflect only the types of cases when the police have been notified (Kaufman Kantor & Straus, 1990; Straus, 1999). As a result, it is argued that police data exclude “at least 90% of the cases of partner assault” (Straus, 2006, p. 1087).

Despite these compelling statistics, some researchers are still reticent to investigate female perpetrators and male victims. Straus (2006) proposed that some have avoided studying female perpetrators because they have an “ideological commitment to the idea that men are almost always the sole perpetrator” (p. 1087). Other researchers have stopped publishing articles on gender symmetry because they have faced severe criticism for their work (see Straus, 1990). Conversely, Reed, Raj, Miller, and Silverman (2010) argued that numerous leading public health institutions, including the World Health Organization, frame IPV “unambiguously as rooted in the social construction of being female, both in the United States and most all other nations,” and that using a “reciprocal violence” framework to investigate IPV “ignores the world beyond our databases” (pp. 349-350). These authors contended that the reciprocal violence framework (where both partners use aggression against one another) evolves from dating violence studies that focus on adolescents and less severe violence, and argued that this line of research is considered separate from IPV or domestic violence research. However, dating violence clearly falls under the conceptualization of IPV as defined by the CDC (2013b) and not all researchers agree with the premise that dating violence is distinct from IPV, but rather would argue that dating violence is a subset of IPV. Similarly, as more forms of violence and aggression are recognized as public health issues, more researchers are investigating these issues, and from more disciplines than ever before while utilizing perspectives other than, or in addition to, feminist or gender-based approaches. While it is certainly encouraging to have so many researchers investigating interpersonal violence, the differing perspectives complicate the study of this issue.

Despite this increased research interest, female perpetrators have been studied relatively infrequently. When they are studied, feminist researchers tend to emphasize the use of violence by women in response to victimization by a male partner, suggesting that self-defense is a primary motivation for female perpetrators (Swan & Snow, 2006), while others suggest that anger and conflict resolution are the primary motivations for the use of violence among both male and female perpetrators of IPV (Straus, 2005). Recognizing the heterogeneity of female perpetrators, several researchers have developed typologies of violent women in an attempt to classify and explain women’s use of violence against intimate partners, as well as to distinguish between various types of IPV relationships (Babcock, Miller, & Siard, 2003; Dasgupta, 2002; Johnson, 2006b); however, these attempts to classify what types of female perpetrators exist are underutilized, which limits the heuristics the public has to draw on when trying to make sense of female aggressors.

Expanding on Johnson’s (1995) concept of intimate terrorism, which has previously been studied only in the context of male perpetration, Hines and Douglas (2010a) investigated the severity and characteristics of male victimization among men in a community sample and a help-seeking sample. While men in the community sample exhibited signs of common couple violence, characterized by reciprocal use of violence by both partners, men in the help-seeking sample matched the description of intimate terrorism, which is characterized by severe and disproportionate use of violence and control by one partner. In contrast to previous research suggesting that perpetration of intimate terrorism is limited to men (Johnson, 1995, 2006a), Hines and Douglas (2010b) provided evidence that females also engage in this type of severe IPV perpetration, and that the consequences of extreme IPV victimization among men are much greater than those experienced by male victims of common couple violence (Hines & Douglas, 2009, 2010a).

Many male victims feel a sense of “bewilderment” because there are no rules or guidelines for how they should respond when victimized (Cook, 1997). Furthermore, male victims face a great deal of shame—a different kind of shame than that of female victims. For example, Cook states that when interviewing male victims, the word “wimp” came up repeatedly. Similarly, some male victims believe they are “weak” because of their position as victims and feel that no one would believe them if they came forward to press charges. Even more, male victims deal with a large amount of ridicule from others. Hence, in some situations, even mentioning a “battered man” creates responses of disbelief, shock, and laughter in some audiences (Cook, 1997).

Another characteristic of male victims is the use of humor to make sense of their experiences. Male victims often use humor as a form of self-protection (Cook, 1997), whereas female perpetrators have been known to point to this humor usage and suggest it reveals that the abuse is not a problem. However, in reality, many male victims use humor because being abused is “so dangerous to him that he dares not admit it for his own well-being, his own self-esteem” (Cook, 1997, p. 53). Other research proposes that male victims may be more tolerant of abuse because they do not fear severe injury and/or they believe in ideas of chivalry (Felson, 2002; Felson, Messner, & Hoskin, 1999). In addition, some scholars have found that violent women suggest that their behavior is permissible because they believe they will not cause severe injury to male victims (Greenblat, 1983). Even more problematic, some female perpetrators are likely to engage in subsequent attacks because they find that it is easy to “get away with it” (Straus, 2006, p. 1090).

It is clear that traditional gender stereotypes are at play, though in nontraditional ways. Stereotypes characterizing women as the weaker gender are perhaps functioning to excuse women’s aggression, and ideas of masculinity constrain men’s ability to make sense of, and seek help for, abuse perpetrated by women. In fact, the lack of male victims that come forward has contributed to skepticism and disbelief toward their stories (Straus, 2006). To this end, Cook (1997) argued that the notion that there are only male perpetrators of IPV hurts all parties involved. When female perpetrators are not identified, they receive little if any assistance, support, or treatment. Even more, the argument over whether a male or a female victim is more important “serves only to detract from providing effective and helpful services for both genders” (p. 34). When victims are ignored or ridiculed, even male victims, this behavior contributes to continued abuse for all individuals in the future, irrespective of gender and what this abuse is labeled. The purpose of this article, however, is not to align with either side of the gender symmetry debate but to explore the ways in which media portrayals of female perpetrators reflect or contest the gender stereotypes that are entangled within the gender symmetry debate and how such portrayals might be contributing to public understanding of female perpetrators and influencing the policy debate surrounding the reciprocal violence approach. To this end, the next section expands upon the importance of media portrayals in shaping public opinion and policy decisions in relation to health issues.

Media Portrayals of IPV

Researchers in gender studies and health communication have been particularly concerned with the role media play in determining the public’s perceptions of IPV (Kozol, 1995; Maxwell, Huxford, Borim, & Hornik, 2000). Media portrayals of female perpetrators can influence how violent women are perceived by the public and may provide skewed representations of the potential risk of IPV and the willingness of male victims to seek help (Carlyle, Slater, & Chakroff, 2008). Of particular importance to the role of gender in IPV, Kozol (1995) argued that media representations reinforce

national ideologies through their focus on individualism and free will . . . attacking the specific problem but not larger structural forces [which] ignores crucial factors of race, class, and sexuality in ways that reproduce, rather than question, American national ideals. (p. 665)

One example of the media’s tendency to overlook underlying contributors to IPV is how the issue is framed (Carlyle et al., 2008).

Gamson (1989) defined a frame as “a central organizing idea for making sense of relevant events and suggesting what is at issue” (p. 157). To frame something thematically is to place an event in some sort of social context, for example, through the use of prevalence and incidence rates or a discussion of the underlying societal factors contributing to IPV. Without such thematic framing, media tend to focus disproportionately on the individual circumstances of an isolated IPV incident rather than how society both contributes to, and has a role in, preventing IPV. Carlyle and colleagues (2008) demonstrated that news coverage of IPV is dominated by episodic rather than thematic framing, which could be reinforcing the idea that IPV is an interpersonal rather than public health issue. While framing stories episodically and focusing on individuals can help to dramatize stories and make them more appealing, such focus can also result in stereotypical portrayals (Sotirovic, 2003). Similar to the arguments put forth by Carlyle et al. (2008), Sotirovic (2003) argued that, in focusing on extreme cases, media portrayals of criminals may lead the public to believe that such people are more different from “average” people, which, in turn, may lead to more individual explanations and attributions for social problems. As described by Lawrence (2004), individual or episodic frames limit the problems to just those individuals whereas systematic or thematic frames “invite government action” (p. 57).

In sum, frames make certain elements and perspectives more salient, especially when specific frames are presented continually. Moreover, effects of framing exist at both the individual and societal level. According to de Vreese (2005), framing affects attitudes at the individual level, and decision making and collective action at the societal level. Thus, how media frame public health issues has important implications for public understanding of the issue and public perceptions of risk, as well as support for a range of policy responses (Lawrence, 2004; Sotirovic, 2003; Yanovitzky & Bennett, 1999). Importantly, because of the limited interpersonal discourse surrounding IPV, the public’s reliance on media to understand IPV as a social issue is potentially increased. Because media have such an impact on public opinion and public policy, it is imperative to examine how media portray female perpetrators to better understand the role these portrayals might be playing in the contentious public debate surrounding gender symmetry and reciprocal couple violence. As such, this study engages in a mixed methods examination of news coverage of female perpetrators that examines quantitative differences in coverage of female and male perpetrators and a qualitative description of the overall portrayal of female perpetrators.

Method

The research in this study included the quantitative comparison of 331 news articles (270 with a male perpetrator and 61 with female perpetrator), as well as a detailed qualitative examination of the articles with female perpetrators. The sample for the study was derived from the data set developed by Carlyle et al. (2008). A brief overview of how that sample was collected is provided below and readers are directed to Carlyle et al. (2008) for more details.

Sample Selection

Stratified media outlets across the country were used to obtain a representative sample of daily newspapers based on their designated market areas (DMAs®). News outlets were sampled such that their probability of inclusion in the sample was approximately proportional to their circulation or reach; specifically, the country’s DMAs® were divided into six strata, with each stratum consisting of approximately one sixth of all U.S. households. The sample was restricted to six strata to allow for reasonable regional representation in all strata and assure reasonable homogeneity of market size within each of the strata (Long, Slater, Boiarsky, Stapel, & Keefe, 2005)].

Researchers created one constructed month each from 2002 and 2003 to avoid the problems associated with using random samples of media content (Riffe, Lacy, Nagovan, & Burkum, 1996). Each year’s constructed month consisted of one constructed week from each season to address any seasonal variation in media coverage. On each day of the constructed month in 2002, one DMA® was randomly selected from each of the six strata. The same DMAs® in 2003 as in 2002 were used because of the large effort involved in constructing the list of daily newspapers in any given DMA®. Regardless of the use of the same DMAs® in both years, the individual media outlets in a selected DMA® were randomly chosen each year.

From this sample, all articles were coded for whether the story contained mention of domestic violence as part of a larger study (see Carlyle et al., (2008) for a complete description). Because domestic violence can include acts of violence between family members other than romantic partners, this designation was further refined to include only those articles that mentioned IPV specifically (i.e., does not include stories about violence between family members other than intimate partners), resulting in a sample size of 395 articles. Physical, sexual, emotional or psychological, and economic forms of IPV were included. For the current study, the sample was further reduced by excluding the 64 articles that addressed IPV generally (e.g., community events or new laws) rather than incidences of IPV between a specific perpetrator and victim, resulting in a final sample of 331 articles.

Data Analysis and Coding

The procedure for the quantitative coding is presented in more detail in Carlyle et al. (2008). To summarize, categories including potential reasons behind attack, types of IPV, history of IPV within the relationship, perpetrator and victim demographics, relationship between perpetrator and victim, criminal and personal history information about perpetrator and victim, and the involvement of alcohol or drugs were coded at the story level by two independent coders. Intercoder reliability was assessed on 20% of the sample with an average Kappa across categories of .77.

The qualitative coding of the articles occurred in three steps. First, we carefully read and reread the articles using open coding and line-by-line coding (Lindlof & Taylor, 2011). Second, we conducted focused coding that involved separating specific types of female perpetrators into categories and looking for patterns in those categories (Charmaz, 2001; Lindlof & Taylor, 2011). During focused coding, the categories changed as we examined three key questions: (a) How are female perpetrators portrayed in the articles? (b) What reasons are provided for her violent behavior? and (c) What types of violence are female perpetrators reported as using? Furthermore, we conducted negative case analyses; purposefully looking for newspaper articles that refuted what our findings were suggesting about the media portrayal of female perpetrators and revising categories until we could not locate any more negative cases (Lindlof & Taylor, 2011).

Third, we wrote theoretical memos that described what our categories meant in relation to previous theoretical work on male victims, female perpetrators, gender symmetry, media effects, and IPV (Charmaz, 2001; Lindlof & Taylor, 2011). Finally, we uploaded electronic copies of the newspaper articles into NVivo qualitative data analysis software. Using NVivo, we coded the data a second time based on the categories we had previously created. We generated a category code book that listed the name of each category or theme, a definition of that category, and examples of the category. In addition, the NVivo node summary report was used to examine the number of times each code appeared in each article and the total number of times each code appeared in all the articles. These numbers helped us determine how female perpetrators were being presented in the articles most frequently.

Results

Quantitative Analyses

The first question asked what the important differences in news coverage of female and male perpetrators were. To answer this question, key categories reported in Carlyle et al. (2008) content analysis were analyzed for gender differences, including potential reasons behind the attack offered, types of violence, whether there was a conspirator, and the criminal and social histories of the perpetrator and victim. Because all variables are categorical, chi-square analyses were used to test all differences in coverage of female and male perpetrators, the results of which are presented next. Overall, 19.3% of articles featured a female perpetrator and 80.7% a male.

Articles with female perpetrators differed significantly in the frequency with which potential reasons behind the attack were reported as compared with articles with male perpetrators. Specifically, articles with female perpetrators were more likely to report victim infidelity, χ²(1) = 23.86, p < .01; self-defense, χ²(1) = 12.80, p < .01; emotional distress, χ²(1) = 19.27, p < .01; and money or economic, χ²(1) = 9.36, p < .01, reasons. Interestingly, there were not significant differences in the frequency with which anger, arguments, jealousy, or alcohol or drug use were reported as potential reasons behind the attack. When examining whether the article addressed the perpetrator’s history of being in an abusive relationship in general, this was reported with a greater frequency for female perpetrators, χ²(1) = 5.80, p < .05.

Although there were not significant differences between reports of female and male perpetrators in the overarching category of physical violence, there were some differences in the reporting of specific types of physical aggression used. Specifically, males were reported as hitting, χ²(1) = 6.85, p < .01, and shooting, χ²(1) = 4.41, p < .05, with a greater frequency than females, whereas females were reported as perpetrating via stabbing with a higher frequency, χ²(1) = 19.33, p < .01. There were no significant gender differences in the frequency with which the types of sexual, emotional/psychological, verbal, and economic abuse were reported as perpetrated. Of note, the overall reporting of these forms of abuse in the sample as a whole was very low (between 0.7% and 7.6% of articles compared with 96.1% of articles reporting some form of physical abuse). There were significant differences in the reports of the type of conspirator the perpetrator had worked with, χ²(1) = 21.35, p < .01. Interestingly, males were reported as hiring their conspirator more often and females were reported as having a romantic partner as a conspirator more often.

The criminal history of female perpetrators was reported with greater frequency than for male perpetrators, χ²(1) = 6.62, p = .01, and this criminal history was more likely to be related to something other than IPV or other violent behavior. Female perpetrators were more likely to have the reported social status of someone relatively unknown, compared with men who were more often reported as sports or public figures, χ²(1) = 14.15, p < .01. There were also significant differences in the reported social status of the victims of male and female perpetrators, χ²(1) = 45.21, p < .01, with reports of female perpetrators having greater than expected sports figure victims. Females were also more likely to be characterized as socially deviant, χ²(1) = 8.33, p < .01. However, there were no significant differences in the ways male and female perpetrators were portrayed with respect to being under the influence of alcohol or drugs, history of alcohol or drug use, and history of sexual activity and infidelity. Conversely, there were significant differences in whether it was reported that the victim had a history of infidelity, with it being mentioned at a much greater rate than expected when the perpetrator was a female, χ²(1) = 34.93, p < .001.

Qualitative Analyses

While the quantitative analyses reported in the previous section offer important insights into areas of disparate coverage of female and male perpetrators, a known limitation of quantitative content analyses is the limited contextual richness of the data. Given this limitation, it is important to also qualitatively examine news coverage of female perpetrators to gain a fuller understanding of how the media describes female perpetrators and also how these portrayals may be influencing public opinion and policy. As such, this section reports the findings from the qualitative examination of the articles featuring a female perpetrator, focusing specifically on the reasons provided for her violent behavior; the specific types of violence the female perpetrators were accused of using; and how female perpetrators were portrayed in the articles in general.

Reasons for female perpetrator behavior

As the quantitative results illustrated, articles with female perpetrators were more likely to report victim infidelity, self-defense, emotional distress, and money or economic reasons, whereas there were not significant differences in reports of anger, arguments, jealousy, or alcohol or drug use as potential reasons behind the attack. These results represent where there are differences in coverage of female and male perpetrators; however, these results do not necessarily indicate the most common reasons provided for the behavior of female perpetrators or what these categories look like descriptively. Within those articles with a female perpetrator, the news stories provided six main reasons for the perpetrator’s behavior: arguments, cheating and/or affairs, money, sudden passion, alcohol or drugs, and mental illness. In the following paragraphs, we provide the prevalence of these reasons followed by exemplars of each type.

The first reason for violent behavior, depicted in 15 articles and referenced 24 times, was that the female perpetrator had been in an argument, which caused the violence. For example, one article stated, “Johnson allegedly told investigators she had argued with her husband the day he was killed.” Furthermore, another article explained, “Ruth Johnson, 50, shot Clifford Johnson, 28, three times on May 21, 2001, after they argued because she wanted a separation, she told police.” Arguments are not an uncommon reason provided for violence in cases of IPV, which makes the finding that arguments were a common reason reported for female perpetrator violence unsurprising.

The second reason given for female perpetrator behavior in the news articles involved cheating and/or affairs. Similar to the previous reason, cheating and/or affairs was provided as a reason for violent behavior in 15 articles and referenced 29 times when the perpetrator was female. For example, one article explained, “A Bronx woman who found her boyfriend in bed with another woman plunged a kitchen knife into his chest, killing him, police said yesterday.” Again, this is consistent with the quantitative analysis that revealed that the victim’s history of infidelity is mentioned at a much greater rate than expected when the perpetrator is a female.

The third reason for female perpetrator behavior, issues related to money, also occurred in 15 articles and was referenced 31 times. The articles reported that female perpetrators wanted to steal their husbands’ money (e.g., “Lallemand’s girlfriend, Lisa Toney, 45, was sentenced to life in prison in May after a federal jury convicted her of conspiring with Lallemand to steal her estranged husband’s identity, run up about US$200,000 in credit card charges in his name and kill him with the bomb”) and cash in on their life insurance policies (e.g., “Gray intentionally caused the death of all three victims for a total of US$165,000 in life insurance money”). This is also consistent with the quantitative results showing that money or economic reasons are mentioned more often for female perpetrators.

The fourth reason the articles described for female perpetrator behavior, sudden passion, was present in 11 articles and referenced 12 times. One example of sudden passion stated, “The 45-year-old dentist, who could have received a life sentence, drew a lesser penalty after jurors decided she was driven by ‘sudden passion’ to run over David Harris repeatedly in a suburban Houston hotel parking lot July 24.” Sudden passion means that the perpetrators actions were driven in the heat of passion and they did not include premeditation.

The fifth reason provided by the news articles was that the female perpetrator was under the influence of alcohol and/or drugs. This reason was provided in five articles and referenced eight times. For example, one article claimed, “Harding must also participate in outpatient alcohol counseling for the next two years. ‘I have a problem and now I’m going to take care of it.’” Another explained that the female perpetrator was a drug addict (e.g., “Prosecutors contend Rossum killed her husband to keep him from revealing to her employers at the San Diego County Medical Examiner’s office that she was a methamphetamine addict who was stealing drugs from work”). As previously mentioned, drug and alcohol abuse is a reason offered for IPV regardless of the gender of the perpetrator.

The sixth and final reason for female perpetrator behavior, mental illness, was present in three articles and referenced three times. For example, one article stated, “According to court records, she failed to participate in required psychological counseling and perform 320 hours of community service at the required rate of 10 hours per month.”

Types of violence used by female perpetrators

In all of the articles, female perpetrators were described as using physical violence. There were no examples of psychological or sexual violence presented. As noted earlier, physical violence dominated coverage regardless of the gender of the perpetrator. Quantitatively, males reported hitting and shooting with a greater frequency than females, whereas females reported a higher frequency of perpetrating via stabbing. However, despite these quantitative differences, there was a wide range of types of physical violence that female perpetrators were reported as using which are detailed below.

Twelve articles stated that the female perpetrator killed the victim by running them over with a car, 7 articles reported that the female killed the victim with a gun, 5 articles reported killed by stabbing with a knife, 3 articles reported killed by poisoning, 2 articles reported conspiracy to kill, and 1 article reported killing with a pipe bomb. In articles that reported violence that did not result in the death of the victim, the female perpetrators were most commonly described as stabbing with a knife (7 articles) or pushing (2 articles), while biting, grabbing, hitting, and hitting with an object were all reported in 1 article, respectively. The finding that 12 articles reported the female perpetrator killed the victim by running them over is potentially due to a few high-visibility cases that were covered multiple times in national news. Therefore, the incidence of running a victim over with a vehicle is likely less prevalent overall for female perpetrators. Fourteen articles explained that the female perpetrator had allegedly killed someone but did provide any descriptions of the types of violence used. Similarly, 5 articles reported that the female perpetrator attacked or assaulted someone but did not provide any more description about the type of violence.

The portrayal of female perpetrators

In the news articles, female perpetrators were portrayed in eight common ways. These included having an accomplice, engaging in abnormal behavior, feeling sadness and/or remorse for their behavior, being wives of sports figures, being described with nicknames given by the media, holding traditional female jobs, working as exotic dancers, and having a history of abuse.

In 16 separate articles and 32 specific references, female perpetrators were described as having an accomplice in their violent behavior. In other words, any articles that explained that the female perpetrator did not act alone were placed in this category. For example, one article explained, “Lynn Ryan Kilroy, who got five years probation last fall after pleading no contest to charges of trying to hire a hit man to kill her millionaire husband.” While this article explained the use of a hit man as an accomplice, other female perpetrators worked with their current boyfriends. For example,

Josephine Gray enlisted the help of each successive husband and boyfriend to commit murder on her behalf, in 1974, in 1990, then 1996, authorities said in court documents. The second and third victims were suspected of killing the husbands who came before them.

This is consistent with the quantitative results that showed a conspirator was more likely to be a romantic partner when the perpetrator was female.

Female perpetrators were portrayed in 15 articles and referenced 24 times, as being what we labeled “abnormal,” which included a range of behaviors and classifications of the female perpetrators that deviated from normal behavior (e.g., she was involved in voodoo). The abnormal category was quite prevalent, occurring in 25% of the total articles. Furthermore, the articles in the abnormal category described behavior that was unique and sensational. One example explained that “prosecutors allege Rossum staged a suicide scene, scattering rose petals around him to mimic a scene from her favorite film, ‘American Beauty,’ then placed a wedding photo nearby.” Of course, staging a crime scene could be argued as typical behavior; however, the use of rose petals and the copycat nature of the behavior appears to be abnormal. In addition, there were numerous articles that described the behavior of Josephine Gray, a woman accused of killing three of her previous husbands. For example, “Authorities said Gray escaped prosecution for years by using threats of voodoo spells and violence to keep potential witnesses silent. Searches of Gray’s home found voodoo dolls resembling Gray’s relatives pierced with needles.” The use of voodoo makes this specific case abnormal. Moreover, these abnormal representations could engender the notion that only abnormal women engage in this type of violent behavior and “regular” women do not. This also aligns with the quantitative analyses indicating that female perpetrators are more likely to be characterized as socially deviant. Female perpetrator behavior was classified as socially deviant if it was described in the article as being abnormal, extreme, and if the behavior broke a societal norm. For example, covering a body with rose petals after the individual is deceased breaks a societal norm about not desecrating dead bodies. In addition, marrying multiple husbands and killing each one for their life insurance money breaks a societal norm of demonstrating care and compassion for one’s spouse or partner.

Female perpetrators were depicted as feeling sad or remorseful for their behavior in 12 articles and referenced 16 times. This is interesting because, similar to stereotypes that women are emotional and men are rational, women were depicted as sad or remorseful 20% of the time. One example that explains this behavior stated,

Harris burst into sobs as the penalty phase began with her 17-year-old stepdaughter confiding on the witness stand that she had attempted suicide since her father’s death last July. The teen was seated next to Harris in the car when her father was run down. “I’m sorry, Lindsey! I’m sorry, baby!” Harris cried.

In this example, Clara Harris is apologizing to her stepdaughter in court for allegedly murdering the child’s father. The comments and behavior described in the article clearly explain that Clara Harris is feeling remorseful about her actions. In another example, a “tearful” perpetrator stated in court, “I really didn’t mean for any of this to happen.” Both of these examples portray female perpetrators as individuals who feel sorry for their actions.

In eight articles, female perpetrators were labeled as wives of professional sports figures. Indeed, the victims of female perpetrators were more likely to be sports figures than victims of male perpetrators. One article explained, “Tawny Kitean agreed Wednesday to a plea bargain on spousal abuse and battery charges alleging she attacked her husband, St. Louis pitcher Chuck Finley.” Another series of articles discussed an alleged attack by a wife of an National Football League player. For example, claims were made such as “suspended Browns running back William Green and his fiancée Asia Gray continued yesterday to dispute charges that Gray stabbed Green with a kitchen knife at their Westlake home.” These portrayals reveal that female perpetrators are not just from low class economic situations, but in fact some are in affluent economic situations married to high profile men.

Six articles and 11 references described the female perpetrators with nicknames. In other words, these articles used nicknames that described the female perpetrator in ways related to her alleged crime. Several articles referred to Josephine Gray as the “Black Widow.” For example, “Gray’s attorney, Michael Citaramanis, said at the December hearing that the ‘Black Widow’ nickname given to his client was unfair.” Josephine Gray was labeled as the black widow because black widow spiders kill their mates and she was accused of murdering three of her previous husbands. In another example of nicknames, Clara Harris, a woman accused of running over her husband with a Mercedes, was called the “Mercedes Killer.”

Female perpetrators were also presented as having a variety of occupations. Five articles and nine references described female perpetrators as having traditional female jobs, including “daycare provider” and “kindergarten teacher.” Interestingly, in these articles, the general reaction from friends and neighbors about the female perpetrator behavior was disbelief. That is, the female perpetrator’s behavior was described as out of character and unbelievable because she seemed “normal” and held a traditional job. However, four articles and five references labeled the female perpetrator’s occupation as an exotic dancer. For example, “An exotic dancer has been charged with trying to hire a hit man to kill her boyfriend and his mother.” In these cases, where the female was employed in what Ashforth and Kreiner (1999) have referred to as “dirty work,” or jobs that are socially, physically, or morally tainted by society, there was less of a response of disbelief in the behavior. This also aligns with the quantitative findings that female perpetrators are viewed as more socially deviant and thus, more likely to engage in work that is considered tainted.

Finally, female perpetrators were portrayed in four articles as being involved in a relationship that had a history of abuse or as retaliating against abusive behavior from their partners. One example stated, “Gately reportedly told police her husband had abused her and she shot him with a .357-caliber handgun as he slept.” And another explained, “Police said they had answered previous domestic calls at the Holmes Place address, but Marva Rose said her family never dreamed the problems between her son and Shannon had deteriorated to that extent.”

Discussion

The study presented here investigated the ways in which female perpetrators of IPV are portrayed by the news media. The first part of the study quantitatively compared the coverage of female and male perpetrators and identified several areas where coverage differed. The second part of the study qualitatively examined coverage of female perpetrators to provide a richer description of this coverage. The following sections describe the key differences in coverage and discuss how such coverage corroborates or challenges various arguments within the gender symmetry debate, including gender stereotypes.

News Coverage of IPV

In the quantitative examination of news coverage, several important differences emerged in relation to the gender of the perpetrator. When examining the potential reasons behind the act of IPV presented in news stories, articles with female perpetrators were more likely to report reasons based on victim infidelity, self-defense, emotional distress, and money. These particular reasons relate to several gender stereotypes and themes within the gender symmetry debate. Specifically, women are often portrayed as being overly emotional and acting out in violence “in the heat of passion.” The tendency to characterize women’s violence as coming from a place of out of control emotions is also evident in Harris, Palazzolo, and Savage’s (2012) examination of gender differences in perceptions of perpetrators of IPV. The authors found that emotional stereotypes were present in many of the perceptions of why perpetrators commit IPV and that women were more often described as “overly emotional.” Such reasoning can serve to excuse violence in the same ways that “boys will be boys” rationalizations excuse some forms of male aggression.

Another idea that pervades the gender symmetry debate is that women are aggressive in response to male violence (Dutton, 2006), or that the couple is engaging in reciprocal violence. There is a large body of research supporting the prevalence of reciprocal couple violence (e.g., Straus, 2004), and whether an aggressive female has been the victim herself of domestic violence is often speculated upon. Looking at the long history of violence against women, this is not an unreasonable speculation. However, it is also indicative of a tendency to try to explain women’s violence, whereas men can often be understood as engaging in violence because it is their nature. Similarly, when the perpetrator was a female, her criminal history was reported more frequently than for male perpetrators. This could also indicate a tendency to explain her aggressive behavior by establishing that she is not a “typical” woman but rather one with criminal tendency. This corresponds to the framing of female aggressors as abnormal and is consistent with Sotirovic’s (2003) argument that individual-focused media portrayals of criminals may lead the public to believe that such people are more different from “average” people, which, in turn, may lead to more individual explanations and attributions for social problems. A female perpetrating IPV is counter to the stereotypes many hold of women; hence, to counter any identification people might feel with the woman, it is important to portray her as abnormal to realign our perceptions of the world. This incongruence between stories of female aggressors and perceptions of what IPV looks like is reflected in the debate over whether reciprocal couple violence and dating violence are really IPV. Because the idea of a female perpetrating violence against a male partner is so counter to many of the assumptions underlying research on violence against women (e.g., patriarchy), some researchers are perhaps more comfortable labeling the act something other than IPV than reformulating their ideas of IPV to include such scenarios. As this discussion illustrates, media portrayals of female perpetrators of IPV are both similar to, and different from, portrayals of male perpetrators.

When looking more closely at some of the similarities in the coverage of female and male perpetrators, two notable patterns in the coverage of IPV emerge. First, although the rates of sexual, emotional or psychological, verbal, and economic abuse reported were similar for both genders, the overall rates at which these forms of IPV were reported was a small fraction compared with physical violence, which comprised approximately 96% of the overall sample. This skewing of coverage toward only physically violent forms of IPV helps perpetuate the misconception that violence has to be physical to be considered IPV. Second, there were no significant differences in the ways male and female perpetrators were portrayed with respect to being under the influence of alcohol or drugs. However, as noted by Carlyle et al. (2008), substance use in the context of IPV is significantly underreported by news stories as compared with epidemiological data. Hence, although these statistics show some equality in the ways in which male and female perpetrators are portrayed, the portrayals themselves are problematic.

The findings of this research both support and refute the gender symmetry debate. In support of the debate, the findings reveal that men and women perpetrate violence against their partners. It is impossible to determine whether the amount of violence that men and women perpetrate is similar because our findings are based only on violence that the news stories covered rather than the actual amount of violence that has occurred. However, our findings do show that women engage in violence—and it is not always in self-defense—a finding that supports the idea there are female perpetrators who may just be violent by nature. Female perpetrators were violent for a variety of reasons including emotional distress, money, and infidelity. In addition, the news stories reported stabbing more often for female perpetrators. Stabbing a victim is an extremely personal and violent choice of weapon because the perpetrator must be very close to the victim. In other words, the findings show that females can be just as violent and in some cases perhaps even more violent than male perpetrators—challenging the gender symmetry notion that men and women perpetrate violence at equal levels.

In sum, the ways in which news coverage of male and female perpetrators is similar lend support to the idea that perhaps violence perpetrated by females on male partners is indeed a form of IPV, whereas the differences reveal many of the arguments central to the gender symmetry debate. Media contribute to the public’s understanding of what constitutes IPV as well as what a victim and perpetrator “look like.” When news media provide systematic representations of IPV, it is likely that public opinion will be likewise limited. In focusing on the potential role of media content in shaping public opinion, the study presented here shifts the study of female perpetrated IPV to the factors shaping the gender symmetry debate, rather than the debate itself. By better understanding the interplay of media, public, and policy agendas, perhaps violence prevention researchers can find a way to restructure the debate that moves forward the public health agenda of violence prevention across all relationships in productive ways.

Strengths and Limitations

One of the primary strengths of this investigation is the mixed methods approach used. As scholars, our understanding of male aggressors is far more advanced than our understanding of female aggressors. Because of this knowledge gap, it was important to look beyond simple gender differences in frequency of coverage of IPV and provide a richer description of how female aggressors are portrayed. Such rich description helps inform the modification of typologies of male aggressors into ones that accurately describe female aggressors. The sample investigated is another strong point of the piece. The use of a nationally representative sample that accounts for both seasonal and regional variations in news coverage enhances the generalizability of the findings.

Although the findings are primarily descriptive, this study represents an imperative step forward in our understanding of female aggressors. The focus of most IPV research is on male aggressors and this study not only examined differences in coverage by gender of the perpetrator but also provides a richer understanding of how female aggressors are portrayed in news stories. The media portrayals of female perpetrators may or may not align with actual female perpetrator experiences. However, given the public’s reliance on media portrayals to form opinions, such an understanding is an important component of addressing numerous issues within the gender symmetry debate.

As with any study, there are limitations to note. First, the sample content was limited to news media. Although news media are an important source of information, people also learn from entertainment and other forms of media and future studies should investigate the portrayal of female aggressors in these mediums. Second, the news stories analyzed here represent only those covered by the media and, as Carlyle et al. (2008) demonstrated, news coverage of IPV can be skewed in comparison to epidemiological data on IPV. Because there is a considerable controversy surrounding female aggressors of IPV, it is possible that the coverage that is present in the media is even more skewed toward extreme or sensational cases. Future research should compare news coverage of IPV with police and hospital data at the community level to better understand which types of stories are being covered and, perhaps more importantly, which are not.

Conclusion

This study content analyzed news media coverage of female perpetrators of IPV. The findings suggest that news coverage may be colluding with, or unintentionally contributing to, several aspects of the gender symmetry debate, including drawing upon stereotypes of women as emotional and needing assistance, whether in the form of a weapon or accomplice, to inflict harm on men. The findings also highlight important differences between media coverage of male and female perpetrators of IPV, and future research should investigate whether these portrayals are reflective of actual epidemiological differences. By increasing our understanding of how female aggressors are portrayed in news media, researchers can now turn to studies that examine the effects of this coverage and its implications for public and policy support as discussed herein.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research reported here was funded in part by National Institute of Alcohol Abuse and Alcoholism Grant AA10377 awarded to Michael D. Slater, Principal Investigator.

Author Biographies

Kellie E. Carlyle (PhD, MPH, the Ohio State University) is the Program Director in the Department of Social and Behavioral Health in the School of Medicine at Virginia Commonwealth University, Richmond, VA. Her research examines the role of communication in health promotion and disease prevention across a variety of contexts, with a particular interest in the promotion of healthy intimate relationships. Much of her research focuses on the design, implementation, and evaluation of public health campaigns. Within these processes, she focuses primarily on designing theoretically based messages and understanding the role of the media in shaping public understanding of health issues.

Jennifer A. Scarduzio (PhD, Arizona State University) is an assistant professor in the Department of Communication at Lamar University, Beaumont, TX. Her research and teaching span the areas of organizational, interpersonal, and health communication centering on emotion, wellness, sensemaking, and identity. She has explored issues of sexual harassment, intimate partner violence, work–life balance, and emotional expression of employees in nonprofit and for-profit organizations.

Michael D. Slater (PhD, Stanford University) is a Social and Behavioral Sciences Distinguished Professor in the School of Communication at the Ohio State University, Columbus. His research has focused on media effects on beliefs and behavior and persuasion processes, primarily but not exclusively in the domain of public health (e.g., substance abuse and cancer prevention). Other current interests include new media and health information seeking, the persuasive effects of entertainment narratives and other message design factors, longitudinal processes of message genre/channel selection and influence, and methodology and research design.

References

Anderson

K. L.

(2005). Theorizing gender in intimate partner violence research. Sex Roles, 52, 853-865.

Ashforth

B. E.

Kreiner

(1999). How can you do it? Dirty work and the challenge of constructing a positive identity. Academy of Management Review, 24, 413-434.

Babcock

J. C.

Miller

S. A.

Siard

(2003). Towards a typology of abusive women: Differences between partner-only and generally violent women in the use of violence. Psychology of Women Quarterly, 27, 153-161.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M. T.

. . .Stevens

M. R.

(2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary Report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Carlyle

K. E.

Slater

M. D.

Chakroff

J. L.

(2008). Newspaper coverage of intimate partner violence: Skewing representations of risk. Journal of Communication, 58, 168-186.

Centers for Disease Control and Prevention. (2013a). Intimate partner violence: Consequences. Retrieved from http://cdc.gov/ViolencePrevention/intimatepartnerviolence/consequences.html

Centers for Disease Control and Prevention. (2013b). Intimate partner violence: Definitions. Retrieved from http://cdc.gov/ViolencePrevention/intimatepartnerviolence/definitions.html

Centers for Disease Control and Prevention. (2013c). Intimate partner violence factsheet. Retrieved from http://cdc.gov/violenceprevention/pdf/IPV_factsheet2012-a.pdf

Charmaz

(2001). Grounded theory. In Emerson

R. M.

(Ed.), Contemporary field research (pp. 335-352). Prospect Heights, IL: Waveland Press.

10.

Coker

A. L.

Davis

K. E.

Arias

Desai

Sanderson

Brandt

H. M.

Smith

P. H.

(2002). Physical and mental health effects of intimate partner violence for men and women. American Journal of Preventive Medicine, 23, 260-268.

11.

Cook

P. W.

(1997). Abused men: The hidden side of domestic violence. Westport, CT: Praeger.

12.

Cronholm

P. F.

(2006). Intimate partner violence and men’s health. Primary Care: Clinics in Office Practice, 33, 199-209.

13.

Dasgupta

(2002). A framework for understanding women’s use of nonlethal violence in intimate heterosexual relationships. Violence Against Women, 8, 1364-1389.

14.

de Vreese

C. H.

(2005). News framing: Theory and typology. Information Design Journal + Document Design, 13, 51-62.

15.

Dutton

D. G.

(2006). Rethinking domestic violence. Vancouver, Canada: University of British Columbia Press.

16.

Felson

R. B.

(2002). Violence and gender reexamined. Washington, DC: American Psychological Association.

17.

Felson

R. B.

Messner

S. F.

Hoskin

(1999). The victim-offender relationship and calling the police in assaults. Criminology, 37, 931-947.

18.

Gamson

A. W.

(1989). News as framing: Comment on Graber. American Behavioral Scientist, 33, 157-161.

19.

Greenblat

C. S.

(1983). A hit is a hit is a hit. Or is it? Approval and tolerance of the use of physical force by spouses. In Finkelhor

Gelles

R. J.

Hotaling

G. T.

Straus

M. A.

(Eds.), The dark side of families (pp. 235-260). Beverly Hills, CA: Sage.

20.

Harris

K. L.

Palazzolo

K. E.

Savage

M. W.

(2012). ‘I’m not sexist, but . . .’: How ideological dilemmas reinforce sexism in talk about intimate partner violence. Discourse & Society, 23, 643-656.

21.

Hines

D. A.

Douglas

E. M.

(2009). Women’s use of intimate partner violence against men: Prevalence, implications, and consequences. Journal of Aggression, Maltreatment & Trauma, 18, 572-586.

22.

Hines

D. A.

Douglas

E. M.

(2010a). A closer look at men who sustain intimate terrorism by women. Partner Abuse, 1, 286-313.

23.

Hines

D. A.

Douglas

E. M.

(2010b). Intimate terrorism by women towards men: Does it exist? Journal of Aggression, Conflict and Peace Research, 2, 36-56.

24.

Johnson

M. P.

(1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and Family, 57, 283-294.

25.

Johnson

M. P.

(2006a). Conflict and control gender symmetry and asymmetry in domestic violence. Violence Against Women, 12, 1003-1018.

26.

Johnson

M. P.

(2006b). Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12, 1003-1018.

27.

Kaufman Kantor

Straus

M. A.

(1990). Response of victims and the police to assaults on wives. In Straus

M. A.

Gelles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8145 families (pp. 473-487). New Brunswick, NJ: Transaction Publishers.

28.

Kozol

(1995). Fracturing domesticity: Media, nationalism, and the question of feminist influence. Signs: Journal of Women in Culture and Society, 20, 646-667.

29.

Lawrence

R. G.

(2004). Framing obesity: The evolution of news discourse on a public health issue. Harvard International Journal of Press/Politics, 9, 56-86.

30.

Lindlof

T. R.

Taylor

B. C.

(2011). Qualitative communication research methods (3rd ed.). Thousand Oaks, CA: Sage.

31.

Long

Slater

Boiarsky

Stapel

Keefe

(2005). Obtaining nationally representative samples of local news media outlets. Mass Communication and Society, 8, 299-322.

32.

Maxwell

K. A.

Huxford

Borum

Hornik

(2000). Covering domestic violence: How the O. J. Simpson case shaped reporting of domestic violence in the news media. Journalism & Mass Communication Quarterly, 77, 258-272.

33.

McLaughlin

E. C.

(2013, May). Haven’t been following the Jodi Arias trial? Read this. Retrieved from http://www.cnn.com/2013/05/04/us/jodi-arias-primer

34.

National Coalition Against Domestic Violence. (2013). Male victims of violence. Retrieved from http://www.ncadv.org/files/MaleVictims.pdf

35.

National Data on Intimate Partner Violence. (2013). 2012 Data on IPV. Retrieved from http://opdv.state.ny.us/statistics/nationaldvdata/natlipvstats.html

36.

Reed

Raj

Miller

Silverman

J. G.

(2010). Losing the “gender” in gender-based violence: The missteps of research on dating and intimate partner violence. Violence Against Women, 16, 348-354.

37.

Reid

R. J.

Bonomi

A. E.

Rivara

F. P.

Anderson

M. L.

Fishman

P. A.

Carrell

D. S.

Thompson

R. S.

(2008). Intimate partner violence among men: Prevalence, chronicity, and health effects. American Journal of Preventive Medicine, 34, 478-485.

38.

Riffe

Lacy

S. R.

Nagovan

Burkum

(1996). The effectiveness of simple random and stratified random sampling in broadcast news content analysis. Journalism & Mass Communication Quarterly, 73, 159-168.

39.

Sotirovic

(2003). How individuals explain social problems: The influences of media use. Journal of Communication, 53, 122-137.

40.

Straus

M. A.

(1990). The National Family Violence Surveys. In Straus

M. A.

Gelles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8145 families (pp. 3-16). New Brunswick, NJ: Transaction Publishers.

41.

Straus

M. A.

(1999). The controversy over domestic violence by women: A methodological, theoretical, and sociology of science analysis. In Arriaga

Oskamp

(Eds.), Violence in intimate relationships (pp. 17-44). Thousand Oaks, CA: Sage.

42.

Straus

M. A.

(2004). Prevalence of violence against dating partners by male and female university students worldwide. Violence Against Women, 10, 790-811.

43.

Straus

M. A.

(2005). Women’s violence toward men is a serious social problem. In Loseke

Gelles

Cavanaugh

(Eds.), Current controversies on family violence (2nd ed., pp. 55-77). London, England: Sage.

44.

Straus

M. A.

(2006). Future research on gender symmetry in physical assaults on partners. Violence Against Women, 12, 1086-1097.

45.

Swan

S. C.

Snow

D. L.

(2006). The development of a theory of women’s use of violence in intimate relationships. Violence Against Women, 12, 1026-1045.

46.

Tjaden

Thoennes

(2000). Extent, nature, and consequences of intimate partner violence: Findings from the National Violence Against Women Survey (Publication No. NCJ 181867). Washington, DC: U.S. Department of Justice. Retrieved from https://www.ncjrs.gov/pdffiles1/nij/181867.pdf

47.

Yanovitzky

Bennett

(1999). Media attention, institutional response, and health behavior change: The case of drunk driving, 1978-1996. Communication Research, 26, 429-453.