Victim Blaming,Gender,and Social Media Commentary: A Randomized Vignette Study of Audience Comments on News Reports of Intimate Partner Homicide

Abstract

Public perceptions of intimate partner homicide victims are influenced by how the news media frames incidents, often perpetuating gendered stereotypes. In particular, research has found that victim blaming is common in the reporting of intimate partner homicide. However, the way the public engages with news media has changed, as social media platforms allow audiences to engage in news creation by posting comments. Despite this shift, limited research has examined the impact of gender and media frames on victim blaming comments. This study used an experimental vignette design to examine whether victim blaming comments made by Australian survey respondents (n = 537) were influenced by the gender of the offender/victim pair and the framing of a media report, controlling for respondents’ media usage, attitudes, and demographics. Survey respondents were randomly assigned to one of four vignettes presenting a news report on an intimate partner homicide, which varied by the gender of the offender/victim and media frame (victim blaming/bad offender). Respondents were asked to leave a comment as if they were on a social media platform. Analyses revealed that respondents more commonly blamed the victim where there were female offender/male victim pairs compared to male offender/female victim pairs. Respondents also more commonly blamed the victim when there was a victim blaming frame compared to a bad offender frame. Finally, the analyses showed an interactive effect of the gender of the offender/victim pair and the media frame on respondents’ comments. Respondents were more likely to blame victims when the victim was male (female offender) and there was a victim blaming frame. The insights from this study have significant implications for policy and practice. Specifically, there is a need for enhanced training and resources for media professionals, as well as the creation of safer online communities through effective comment moderation.

Keywords

intimate partner violence/abuse media and violence gender-based violence

The ways in which victims of intimate partner homicide are reported by the news media—referred throughout this paper simply as the media—can reveal underpinning cultural assumptions particularly around gender that can then influence public attitudes and responses (Fairbairn & Dawson, 2013). For instance, research commonly shows themes of direct victim blaming (e.g., negative characterization of victims) and indirect victim blaming (e.g., excusing the offender) in the reporting of intimate partner violence (e.g., Sutherland et al., 2019; Taccini & Mannarini, 2024). These narratives may impede empathy and support for those who have suffered harm (Meyers, 1994). However, the way in which news is consumed has been reshaped by the advent of social media platforms, which have shifted the potential role of audiences from silent consumers to prospective news promoters (e.g., “liking” and sharing news) and content creators (e.g., posting comments on social media) (Hille & Bakker, 2014). Despite this shift, few studies examine online comments of intimate partner homicide, especially concerning how victim blaming comments may differ across the gender of the offenders/victims¹ and whether media framing affects these comments. Using an experimental research design, this study examines the association between victim blaming comments, the gender of the offender/victim pairs, and media framing for the reporting of intimate partner homicide.

Intimate partner homicide is internationally recognized as a public health concern, having significant adverse impacts on individuals and communities (Dobash & Dobash, 2015; Miles & Bricknell, 2024; Spencer & Stith, 2020; United Nations Office on Drugs and Crime, 2023). For example, 16% of homicides in Australia involve former or current intimate partners (Miles & Bricknell, 2024). Perceptions of these incidents are often shaped by underlying cultural norms and beliefs about gender. Men and women frequently have different motivations for committing intimate partner homicide. For instance, men are often driven by factors such as separation or jealousy, whereas women are more likely to kill as a response to experiencing prior violence (Belknap et al., 2012; Dobash & Dobash, 2015). The reasons why men and women kill will likely impact on media coverage and associated cultural understandings of intimate partner homicide and, consequently, audience perceptions.

Why Do Media Frames and Online Comments Matter?

Crime is a social construction, and the media plays an important role in shaping people’s understanding of crime (Surette, 2011; Surette & Otto, 2001). Framing is the process through which journalists select certain aspects of events and topics they are covering and make those aspects more salient in their reporting (Entman, 1993). A frame enables users to categorize, label, and interpret various world events (Surette, 2011). Framing effects refers to the impact that framing has on the audience. Research regarding framing effects focuses on the media’s role in shaping public opinion, examining how differences in presentation or framing can affect how the audience perceives, understands, and remembers a certain topic or issue (Entman, 1993). Research suggests that how the media frames crime reporting plays a crucial role in shaping public opinion about crime and victims, influencing policymakers and policy, and increasing fear of victimization (Beckett & Sasson, 2003; Surette, 2011; Wlezien & Soroka, 2024).

Online comments on media stories are also now an important part of the reporting of events and issues, as comments create another (potentially incorrect) means by which public perceptions of crime and victims are shaped. By posting online comments expressing their thoughts and opinions, the audience increasingly helps shape the news (Hille & Bakker, 2014). Unlike traditional news readers, commenters create news, ideas, and public discussions rather than solely consuming information (Hille & Bakker, 2014). Thus, the responsibility of producing news no longer solely rests with news outlets. The easy accessibility, speed, and spread of social media significantly increases the risk of perpetuating gendered and harmful stereotypes and myths (Komazec & Farmer, 2020). There is evidence to suggest that individuals can have their attitudes changed simply by reading others’ comments (Anderson et al., 2014). For example, one study found that reading disapproving comments below a neutral news article could influence readers’ opinions to also be disapproving of the issue reported (Lee & Jang, 2010). Even if readers disagree with what is written in the comments, they may still believe the comments are indicative of broader public opinion (Lee & Jang, 2010). Another study found that uncivil comments led people to perceive a news story as being less credible (Waddell, 2018). These findings underscore the significant influence that comments can have on shaping public perceptions and attitudes.

In the context of gendered violence, and the ways in which victims are portrayed and discussed on social media, little is known about how gender influences the content of social media commentary. Given the gendered nature of intimate partner homicide and that the circumstances under which men and women kill differ substantially (Spencer & Stith, 2020), it is reasonable to assume that social media commentary about intimate partner homicide would also be impacted by gender. Gendered stereotypical framing of intimate partner homicide within social commentary is problematic, especially the continuation of victim blaming narratives. Homicide research has a long tradition of focusing on how outcomes of violent events might be shaped by the victim’s actions (e.g., Wolfgang, 1957). However, emphasizing victims as precipitators of their own deaths risks coming across as victim blaming. The limited research available demonstrated that both male and female victims are blamed in online comments; however, male victims are more likely to be blamed compared to female victims (Komazec & Farmer, 2020). Male victims are blamed for “deserving it,” while female victims are blamed for not leaving the relationship or for not doing something to prevent their victimization (Komazec & Farmer, 2020). While this study highlights examples of gendered blame, it is important to recognize that other factors, not addressed in this research, also influence victim blaming, such as mental health challenges or perceptions of being “provoked.” More research is needed to ascertain the impact of gender on comments relating to intimate partner homicide.

Experimental Studies

While existing research suggests that media frames can shape or change attitudes, much of it is based on inferences derived from cross-sectional studies. To truly understand the effects of framing, it is important to use experimental studies wherein certain variables (e.g., content) are held constant whereas others (e.g., frames and the gender of offender/victim) are varied. This section focuses specifically on five papers (from three separate studies) conducted in the United States that have used experimental designs. One of these focused on the framing of male offenders (Carlyle et al., 2014); one on the framing of female offenders (Bauman, 2018); and three (from the same study) compared the framing of male and female offenders (Savage et al., 2017a, 2017b; Scarduzio et al., 2017). However, none of the studies specifically examined whether the frames present in articles actually affect how and what people write in online comments on news stories. Instead, the studies focused on how news frames affect the attitudes of participants. As these studies were conducted in the United States, there could be locational and cultural differences that distinguish the findings from studies in Australia. For example, Australia has significantly fewer homicides than the United States, and there are fewer instances of homicides involving firearms (United Nations Office on Drugs and Crime, 2023). It is also important to recognize that the binary nature of the gender analysis in these experimental studies is a limitation.

Victim blaming frames appear to have victim blaming effects. For example, Carlyle et al. (2014) assessed the impact of news coverage of intimate partner violence by manipulating the news story (i.e., adding paragraphs containing information about the female victim and male offender). This study focused on male offenders/female victims and did not examine female offenders/male victims. A composite newspaper article reflecting typical intimate partner violence media coverage was created and modified with statements about the perpetrator and victim to alter perceived responsibility (Carlyle et al., 2014). Students were randomly assigned to read one of four news stories and completed a questionnaire. Students who were exposed to a news story that portrayed a female victim as responsible were less sympathetic to the victim, compared to students who were exposed to a news story that portrayed the victim as blameless (Carlyle et al., 2014). When a student was more sympathetic to a victim, the student indicated that they were more likely to engage in protective action to help victims of intimate partner violence, and more likely to support public health initiatives (Carlyle et al., 2014). This highlights how a frame can impact on audience perceptions of victims including influencing victim blaming attitudes. However, this study does not ascertain whether a victim blaming frame would also influence the audience to write victim blaming comments.

The effects of frames are also apparent for offenders, and such effects appear to be dependent on the gender of the audience. Bauman (2018) examined the impact of news coverage of women who had committed intimate partner homicide. Participants were randomly allocated to read one of four newspaper scenarios that varied the stereotype fit of a battered woman. They were then asked how closely the woman fit the battered woman image, her level of responsibility for the described events, and whether they perceived her as a victim or perpetrator (Bauman, 2018). The study demonstrated that female participants were likely to view female offenders as victims who acted in self-defense (i.e., indirectly blaming the victim), while male participants only viewed the offenders as victims if they fit the stereotypical image of a battered woman (Bauman, 2018). This study suggests that media frames affect attitudes differently for men and women who read news stories but fails to examine whether it affects what and how the reader writes about the news story in their own comments online.

A limited number of studies have used an experimental approach by varying a news article relating to an incident of intimate partner violence to compare the effects of framing depending on if the offender is male or female. Savage et al. (2017a, 2017b) and Scarduzio et al. (2017) all used data from the same survey that was given to participants after reading one of six intimate partner violence news stories. The news story had been varied to change the offender’s gender (male/female) and the severity of violence (weak/strong/fatal). Participants rated intimate partner violence stories with male offenders as significantly more serious than female offenders (Savage et al., 2017b). Participants also believed that male offenders were more likely to have previous involvement in intimate partner violence compared to female offenders (Savage et al., 2017a). It also appears the gender of the participant impacts on framing effects. Female participants were more likely to find higher ratings of responsibility for male offenders compared to female offenders, although male participants found male and female offenders equally responsible (Savage et al., 2017a). Similarly, other studies demonstrate that demographics (Näsi et al., 2021) and personal experience with victimization and the criminal justice system (Näsi et al., 2021; Pickett et al., 2015) can shape audiences’ understanding of an issue. This highlights the importance of using experimental research where details about participants (e.g., demographics and personal experiences) can be analyzed.

Factors Influencing Audience Perceptions

Collectively, the experimental studies show that the gender of the offender/victim and media frames can impact audience perceptions about the blameworthiness of victims and offenders. However, as stated previously, none of the studies specifically examine whether the gender of the offender/victim and the frames present in articles actually affect what people write in online comments on news stories. One Australian study that has examined this specific issue found that both the gender of the offender/victim and the frame of the media article can impact on comments (Komazec & Farmer, 2020). However, the study was not experimental and did not control for audience characteristics (e.g., age and gender), so it is not possible to ascertain whether these effects are more prominent among, for example, male commenters compared with female commenters. It is particularly important to consider whether victim blaming frames lead to victim blaming comments due to the significant detrimental impact victim blaming has on attitudes toward intimate partner violence.

Critics of the effects of framing highlight a number of considerations that need to be incorporated into research, such as that depicted in the current paper. Of particular importance to this paper is the argument that prior attitudes, characteristics, and personal experiences affect the influence of framing on audiences (Carragee & Roefs, 2004; Van Gorp et al., 2009). Specifically, the type of media a person consumes (e.g., online or traditional media) (Näsi et al., 2021), personal experience such as prior victimization (Näsi et al., 2021), prior experience with the criminal justice system (Pickett et al., 2015), and demographics including age, ethnicity, and education (Näsi et al., 2021) can impact views of crime. Therefore, in addition to media frames, a person’s prior attitudes, characteristics, and personal experiences also shape their interpretation of media. Without controlling for these individual differences, studies may overlook key variations in how framing effects manifest across different groups, leading to incomplete or skewed findings about media influence on public perceptions of crime. Thus, accounting for these variables enhances the accuracy and reliability of framing effect research. Another critique of framing effects is that it can be difficult to distinguish between the content (e.g., the facts of the case) and frames (e.g., victim blaming) of media articles (Otieno et al., 2013). Using experimental vignette designs, researchers can retain the same content while altering the media frame.

The Present Study

This study utilizes an experimental research design to examine whether victim blaming comments are influenced by the gender of the offender/victim and the frame of the article. The study also controls for media usage, attitudes, and participant demographics. This study answers the following research questions: (a) What is the prevalence of victim blaming comments?; (b) Does the gender of the offender/victim pair affect victim blaming comments?; (c) Do media frames affect victim blaming comments?; and (d) Does the gender of the offender/victim pair and media frame interact to affect victim blaming comments? While existing research is limited, we would expect the gender of the offender/victim and the media frame to impact on victim blaming comments, such that male victims and victim blaming media frames would be more likely to attract victim blaming comments.

Methodology

Participants and Procedures

Researchers recruited respondents via Facebook using a paid advertisement, which ran for approximately 1 month. Facebook uses an algorithm to determine who sees the advertisement. The advertisement appeared on Facebook users accounts who were aged 18 years and older and lived in Australia. Once a respondent clicked on the advertisement, the system redirected them to the survey. The researchers conducted this research in accordance with ethical approval granted by the Griffith University Human Research Committee (reference number: 2019/133) on February 15, 2019.

The survey was administered using the online survey tool LimeSurvey. The Facebook advertisement contained a single URL directing respondents to the survey. Respondents were randomly assigned to one of four vignettes using the random integer equation function in LimeSurvey (hidden to respondents). The vignettes featured an intimate partner homicide news article edited specifically for the current study but based on an existing online news article (ABC News, 2015). This was to avoid the vignette not feeling “real,” which is a known limitation of vignette research (Marques Nascimento Macêdo & Bispo, 2023). The existing article was edited to vary across two dimensions: gender of the offender/victim pairs (male or female) and type of media frame (bad offender or victim blaming frame). Contrasting frames were chosen to investigate whether one frame led to more victim blaming comments than the other. While a victim blaming frame puts responsibility on the victim, a bad offender frame criticizes and blames the offender for their actions. This resulted in four scenarios: (a) a bad offender frame with a male offender/female victim pair, (b) a bad offender frame with a female offender/male victim pair, (c) a victim blaming frame with a male offender/female victim pair, and (d) a victim blaming frame with a female offender/male victim pair. The vignettes can be found in Supplemental Table A1. The bad offender frame criticizes and blames the offender (e.g., a cowardly and vicious shooting), while the victim blaming frame provides excuses or sympathy for the offender (e.g., a tragic shooting). Research indicates that victim blaming extends beyond directly attributing fault to the victim; it can also involve indirectly blaming them by expressing sympathy for or excusing the offender (e.g., Meyers, 1994; Richards et al., 2011; Simington & Farmer, 2024; Sutherland et al., 2019; Taccini & Mannarini, 2024).

Respondents were asked to read the online article and then leave an online comment in a textbox as they would on social media (“If you read this article on a social media site and was asked to make a comment, what would it be?”). Participants were subsequently asked questions about their media and social media usage, criminal justice attitudes, and demographics (see operationalizations below). A total of 639 respondents participated in the online survey, were over 18, and lived in Australia (the requirements for participation). However, providing a comment on the article (i.e., the vignette) was not compulsory. Given the focus on online commentary in the current study, respondents who did not leave a comment were excluded from the sample (n = 49). In addition, respondents were provided with an attention check question to ensure that they had read (and understood) the article. They were asked “how was the victim killed?” This attention check was included to enhance the quality of the data (Abbey & Meloy, 2017). Respondents who answered this question incorrectly were eliminated from the sample (n = 3). A total of 537 (84%) respondents wrote a comment on the article they had read, completed the whole survey (i.e., all the questions), and correctly answered how the victim was killed. This was the sample used in the current study. The final sample consisted of 136 comments pertaining to the bad offender frame with a male offender/female victim pair; 127 comments pertaining to the bad offender frame with a female offender/male victim pair; 121 comments pertaining to the victim blaming frame with a male offender/female victim pair; and 153 comments pertaining to the victim blaming frame with a female offender/male victim pair.

Measures

Victim Blaming Comments

Researchers coded respondents’ comments for victim blaming using deductive thematic analysis. A comment was coded as victim blaming if either direct victim blaming techniques (e.g., negative characterization of the victim) or indirect victim blaming techniques (e.g., excusing the offender) were present, to align with past research that emphasizes both techniques as a part of a larger pattern of victim blaming tactics (Meyers, 1994; Richards et al., 2011). Meyers (1994) highlights that there is a dichotomy of victim blame between indirect victim blaming tactics (such as excusing the offender due to mental health issues, marriage separation, alcohol use, or being out of control) and direct victim blaming tactics (such as the negative characterization of the victim by using language that “name calls” the victim). Richards et al. (2011) identify further examples of direct and indirect victim blaming. The former can include the victim’s failure to report or prosecute previous violence, and actual or suspected infidelity, and the latter the offender’s use of alcohol or drugs, and/or their mental, physical or emotional problems. To increase reliability, coding was discussed within the research team (as a form of peer scrutiny; Shenton, 2004). By regularly reviewing coding decisions, researchers were able to identify any inconsistencies or potential biases in the coding process, fostering a more objective and rigorous analysis.

Offender/Victim Pair Gender

The first dimension on which the vignettes varied was the gender of the intimate partner homicide offender/victim pair featured in the news article. The vignettes featured either a male offender/female victim pair or a female offender/male victim pair.

Media Frame

The second dimension on which the vignettes varied was the way in which the news article was framed. The vignettes were written up to present either a bad offender frame or a victim blaming frame.

Control Variables

Several control variables were included in the study. Some of these variables were constructed by the research team, while others were drawn from existing literature.

Online Media Main Source of News

Respondents were asked “what is your main source of news?.” Respondents could select either television news, radio news, print news, online news via websites, online news via social media, or other (where they could list their main source). Those who selected either online news via websites or online news via social media were coded as “1.” All other responses were coded as “0.”

Has Commented on Social Media

Respondents were asked “how often do you comment on an online news article on social media?” Respondents could choose from never, rarely, sometimes, or often. Those who selected never were coded as “0,” while all other responses were coded as “1.”

Confidence in Legal System

Respondents were asked “how much confidence do you have in the courts and legal system?” (Roberts & Indermaur, 2007). Respondents could select none, some, or a lot. Those who selected none were coded as “0,” while those who selected some or a lot were coded as “1.”

Wants Harsher Sentences

Respondents were asked if “the death penalty should be the punishment for murder” and if “people who break the law should be given harsher sentences” (Roberts & Indermaur, 2007). Each question was measured on a 7-point Likert scale ranging from strongly disagree to strongly agree. Respondents who agreed or strongly agreed with either statement were coded as “1.” All other responses were coded as “0.”

Fear of Crime

Respondents were presented with eight offense types (sexual assault, car-jacking, getting mugged, getting beaten up, being knifed or shot, getting murdered, being burglarized while at home, and being burglarized while no one is home) and asked to indicate if they were worried about this crime happening to themselves on a 4-point Likert scale (never, rarely, somewhat frequently, and frequently) (Dowler, 2003). Items were summed (possible range: 0–24), with higher scores indicating higher levels of fear of crime. The scale showed good internal consistency (α = .86).

Domestic Violence Victimization

Respondents were asked whether they had previously been a victim of domestic violence (Dowler, 2003). Those who answered yes were coded as “1,” and those who answered no were coded as “0.”

Awareness of Domestic Violence

Respondents were given 10 scenarios and asked to comment on whether these were a form of domestic violence or violence against women. An example scenario is, “one partner controls the social life of the other partner by preventing them from seeing family and friends.” These scenarios were drawn from the Australian Attitudes to Violence Against Women survey (Harris et al., 2015). The responses were coded as “1” if the respondent indicated that all the scenarios were domestic violence. The responses were coded as “0” if the respondent selected that one or more of the scenarios were not domestic violence.

Direct Contact with Criminal Justice System (Self or Family)

Respondents were asked, “have you had any direct contact with police and criminal courts yourself” and “has a close family member had any direct contact with police and criminal courts.” Responses were coded as “1” if respondents answered “yes” to either one of these questions. All other responses were coded as “0.”

Gender

Respondents were asked to indicate their gender. Responses were coded as either “0” (male) or “1” (female).²

Age

Respondents were asked to enter their age (continuous variable).

University Degree

Respondents were asked “what is your highest level of education.” Respondents could select less than year 12, year 12, certificate, diploma, bachelor’s degree, postgraduate degree, or PhD. Responses were coded as “1” if they selected either bachelor’s degree, postgraduate degree, or PhD. All other responses were coded as “0.”

Analytical Strategy

Researchers analyzed data using IBM SPSS (version 29). Researchers ran descriptive statistics to answer research question 1 (prevalence of victim blaming). To answer research questions 2 (effect of gender) and 3 (effect of media frame), researchers performed chi-square tests (using Yate’s Continuity Correction) followed by logistic regressions, the latter of which included the control variables. Before running the logistic regression models, researchers checked the data for multicollinearity (tolerance levels and correlation matrix) and influential cases (Cook’s distance).³ To answer research question 4 (interaction effects), researchers ran two logistic regression models. The first was a main effects model (both gender and media frame served as predictors), and the second was an interaction model that included the main effects and the interaction term (gender × media frame).⁴ To interpret the direction of the significant interaction effect, odds ratios were manually computed and compared for the gender of the offender/victim pairs across the type of frame used in the article. The interaction was calculated by multiplying the gender odds ratio by the interaction ratio of odds ratios: b(g1)*b(g#f) (UCLA: Statistical Consulting Group, n.d.). Nagelkerke R² and the area under the receiver operating characteristic (ROC) curve were used to examine model fit (values are presented in the results tables).

Results

RQ1: Prevalence of Victim Blaming Commentary

Table 1 presents descriptive statistics and correlations for the study variables. Importantly, the table shows that almost one-quarter of all comments (23.8%) were coded as victim blaming.

Table 1.

Descriptives and Pearson’s Correlation for Variables (n = 537).

Variables	n (%)	M (SD)	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Victim blaming comment (yes)	128 (23.8)		—
2. Gender of the offender/victim (female/male)	274 (51.0)		.30**	—
3. Media frame (victim blaming)	280 (52.1)		.20**	.08	—
4. Online media main source of news (yes)	453 (84.4)		−.02	−.03	.00	—
5. Has commented on social media (yes)	429 (79.9)		.00	−.02	−.01	.05	—
6. Confidence in legal system (yes)	405 (75.4)		−.11	.03	.03	.00	−.17	—
7. Wants harsher sentences (yes)	332 (61.8)		.09*	.00	−.02	−.06	.10*	−.21**	—
8. Fear of crime		8.66 (4.75)	.05	−.02	.01	.08	.08	−.13**	.28**	—
9. Direct contact with CJS (yes)	373 (69.5)		−.01	−.07	−.08	−.06	.19**	−.13**	.01	.11*	—
10. Awareness of domestic violence (yes)	462 (86.0)		−.04	−.05	−.05	.05	.00	.08	.05	.05	.04	—
11. Domestic violence victimization (yes)	270 (50.3)		.02	−.01	−.02	.04	.17**	−.15**	−.02	−.04	.29**	−.01	—
12. Gender (female)	442 (82.3)		.00	−.03	.06	.04	−.01	−.04	.11*	.23**	.18**	.12**	0.01	—
13. Age		44.55 (14.11)	.03	−.05	.01	−.12	.34**	−.12**	.08	−.12**	.14**	.04	.14**	−.08	—
14. University degree (yes)	233 (43.4)		−.07	.07	.01	−.03	−.08	.18**	−.21**	−.13**	−.07	.03	−06	.04	.01

Note. CJS = criminal justice system.

p < .05 level (two-tailed). **p < .01 level (two-tailed).

RQ2: The Direct Effect of Offender/Victim Gender on Victim Blaming Commentary

Bivariate analyses showed that respondents were significantly more likely to blame the victim where the article featured female offender/male victim pairs (36.1%) compared to male offender/female victim pairs (11%) (χ² [1, n = 537] = 45.214, p < .001, φ = .295). Gender of offender/victim pair remained statistically significant while controlling for respondent characteristics and attitudes in the multivariate analyses (see model 1, Table 2). Respondents who read an article featuring a female offender (male victim) were almost five times (OR = 4.95) more likely to have written a victim blaming comment compared with respondents who read an article with a male offender (female victim), controlling for all other factors in the model. In addition, one of the control variables was statistically significant (confidence in the legal system: OR = 0.60).

Table 2.

Predictors of Offender/Victim Gender and Media Frames on Victim Blaming Comments (n = 537).

	Dependent Variable: Victim Blaming Comments
	Model 1: Offender/Victim Gender Pair					Model 2: Media Frame
Variables	B (SE)	Wald	Sig.	Exp (B)	[95% CI]	B (SE)	Wald	Sig.	Exp (B)	95% CI]
Gender and frame
Female offender/male victim (1 = yes)	1.60 (0.24)	44.26	<.001	4.95	[3.09, 7.93]
Victim blaming frame (1 = yes)						1.06 (0.22)	22.49	<.001	2.88	[1.86, 4.47]
Respondent experiences/attitudes
Online media main source of news (1 = yes)	−0.04 (0.30)	0.02	.881	0.96	[0.53, 1.71]	−0.13 (0.29)	0.20	.658	0.88	[0.50, 1.55]
Has commented on social media (1 = yes)	−0.26 (0.29)	0.76	.382	0.77	[0.43, 1.38]	−0.26 (0.29)	0.78	.376	0.77	[0.44, 1.37]
Confidence in legal system (1 = yes)	−0.52 (0.25)	4.13	.042	0.60	[0.36, 0.98]	−0.51 (0.25)	4.24	.039	0.60	[0.37, 0.98]
Wants harsher sentences (1 = yes)	0.31 (0.25)	1.57	.210	1.36	[0.84, 2.21]	0.35 (0.24)	2.09	.148	1.42	[0.88, 2.29]
Fear of crime	0.01 (0.02)	0.36	.548	1.01	0.97, 1.07]	0.01 (0.02)	0.34	.559	1.01	[0.97, 1.06]
Domestic violence victimization (1 = yes)	−0.02 (0.24)	0.01	.933	0.98	[0.62, 1.56]	−0.03 (0.23)	0.02	.897	0.97	[0.62, 1.52]
Awareness of domestic violence (1 = yes)	−0.16 (0.30)	0.27	.605	0.85	[0.47, 1.55]	−0.13 (0.30)	0.20	.653	0.88	[0.49, 1.57]
Direct contact with CJS (self or family) (1 = yes)	0.03 (0.25)	0.02	.900	1.03	[0.63, 1.70]	0.10 (0.25)	0.16	.693	1.10	[0.68, 1.78]
Respondent demographics
Gender (1 = female)	−0.01 (0.30)	0.00	.976	0.99	[0.55, 1.78]	−0.17 (0.29)	0.33	.565	0.85	[0.48, 1.50]
Age	0.01 (0.01)	0.94	.333	1.01	[0.99, 1.02]	0.00 (0.01)	0.32	.573	1.00	[0.99, 1.02]
University degree (1 = yes)	−0.31 (0.23)	1.89	.169	0.73	[0.47, 1.14]	−0.18 (0.22)	0.66	.417	0.83	[0.54, 1.29]
Constant	−1.95 (0.70)	7.78	.005	0.14		−1.39 (0.65)	4.51	.034	0.25
Model statistics
−2 Log likelihood	526.768					554.149
Nagelkerke R²	16.6%					9.6%
Model χ²	63.056					35.674
Sig.	<.001					<.001
Model prediction rate	76.0%					76.4%
ROC	.727					.673

Note. CI = confidence interval; CJS = criminal justice system; ROC = Receiver operating characteristic; SE = standard error.

RQ3: The Direct Effect of Media Framing on Victim Blaming Commentary

Bivariate analyses revealed that respondents more commonly blamed the victim when the article had a victim blaming frame compared to when the article had a bad offender frame (32.1% vs. 14.8%) (χ² [1, n = 537] = 21.291, p < .001, φ = .203). Media frame remained statistically significant when controlling for respondent characteristics and attitudes in the multivariate analyses (see model 2, Table 2). Respondents who read a victim blaming article were almost three times (OR = 2.88) more likely to have written a victim blaming comment compared with respondents who read a bad offender article, controlling for all other factors in the model. One of the control variables (confidence in the legal system) was also statistically significant (OR = 0.60).

RQ4: The Interactive Effect of Offender/Victim Gender and Media Framing on Victim Blaming Commentary

Interaction effects were examined by running two separate models, one with the main effects model only (i.e., inclusion of both offender/victim gender pair and frame; model 1, Table 3) and one with the main effects and the interaction term (model 2, Table 3). In model 1, both gender and the media frame were shown to be statistically significant (OR = 4.88 and OR = 2.82, respectively). Similar to the models presented in Table 2, one of the control variables (confidence in the legal system) was also statistically significant (OR = 0.56).

Table 3.

Interaction Effect of Offender/Victim Gender and Media Frames on Victim Blaming Comments (n = 537).

	Dependent Variable: Victim Blaming Comments
	Model 1: Main Effects Model					Model 2: Model with Interaction
Variables	B (SE)	Wald	Sig.	Exp (B)	[95% CI]	B (SE)	Wald	Sig.	Exp (B)	[95% CI]
Gender and frame
Female offender/male victim (1 = yes)	1.58 (0.25)	41.75	<.001	4.88	[3.02, 7.88]	2.73 (0.55)	24.24	<.001	15.36	[5.18, 45.56]
Victim blaming frame (1 = yes)	1.04 (0.23)	19.60	<.001	2.82	[1.78, 4.47]	2.24 (0.56)	15.92	<.001	9.39	[3.13, 28.22]
Female offender × victim blaming frame						−1.61 (0.63)	6.60	.010	0.20	[0.06, 0.68 ]
Respondent experiences/attitudes
Online media main source of news (1 = yes)	−0.02 (0.30)	0.01	.940	0.98	[0.54, 1.77]	−0.07 (0.31)	0.05	.816	0.93	[0.51, 1.70]
Has commented on social media (1 = yes)	−0.29 (0.30)	0.92	.338	0.75	[0.41, 1.35]	−0.30 (0.30)	1.01	.314	0.74	[0.41, 1.33]
Confidence in legal system (1 = yes)	−0.58 (0.26)	4.97	.026	0.56	[0.34, 0.93]	−0.60 (0.26)	5.31	.021	0.55	[0.33, 0.91]
Wants harsher sentences (1 = yes)	0.35 (0.25)	1.88	.170	1.41	[0.86, 2.32]	0.36 (0.25)	2.04	.153	1.43	[0.88, 2.34]
Fear of crime	0.02 (0.03)	0.40	.529	1.02	[0.97, 1.07]	0.01 (0.02)	0.35	.553	1.01	[0.97, 1.07]
Domestic violence victimization (1 = yes)	0.15 (0.25)	0.37	.541	1.16	[0.72, 1.88]	0.08 (0.25)	0.09	.760	1.08	[0.67, 1.75]
Awareness of domestic violence (1 = yes)	−0.11 (0.31)	0.12	.729	0.90	[0.49, 1.65]	−0.05 (0.31)	0.03	.865	0.95	[0.52, 1.73]
Direct contact with CJS (self or family) (1 = yes)	−0.01 (0.26)	0.00	.977	0.99	[0.60, 1.65]	0.04 (0.26)	0.02	.876	1.04	[0.62, 1.74]
Respondent demographics
Gender (1 = female)	−0.15 (0.31)	0.25	.616	0.86	[0.47, 1.56]	−0.12 (0.31)	0.14	.707	0.89	[0.49, 1.63]
Age	0.01 (0.01)	0.56	.455	1.01	[0.99, 1.02]	0.01 (0.01)	0.92	.338	1.01	[0.99, 1.03]
University degree (1 = yes)	−0.30 (0.23)	1.61	.205	0.74	[0.47, 1.18]	−0.32 (0.23)	1.83	.176	0.73	[0.46, 1.15]
Constant	−2.44 (0.72)	11.60	<.001	0.09		−3.42 (0.85)	16.18	<.001	0.03
Model statistics
−2 Log likelihood	505.919					498.059
Nagelkerke R²	21.7%					23.6%
Model χ²	83.904					91.764
Sig.	<.001					<.001
Model prediction rate	77.5%					78.0%
ROC	.760					.765

Note. CI = confidence interval; CJS = criminal justice system; ROC = Receiver operating characteristic; SE = standard error.

Model 2 fit remained similar to model 1 though with slight improvements (χ² = 7.861, p = .005), indicating that adding the interaction had a small improvement on the fit. As shown in model 2, the interaction term (gender × frame) was statistically significant, as were the individual predictors (gender: OR = 15.36, and frame: OR = 9.39), and one control variable (confidence in the legal system; OR = 0.55). This suggests that the gender of the offender/victim pair does moderate the relationship between the framing and the likelihood of the respondent making a comment that blamed the victim.

To interpret the direction of the significant interaction effect, odds ratios were manually computed and compared for the gender of the offender/victim pairs across the type of frame used in the article (see “Analytical Strategy” section). Results revealed that when a bad offender frame was used in the vignette, the odds of the outcome being a victim blaming comment were 15.36 times greater for female offender/male victim pairs compared to male offender/female victim pairs. However, when a victim blaming frame was used in the vignette, the odds of the outcome being a victim blaming comment was only 3.07 times greater for female offender/male victim pairs compared to male offender/female victim pairs. These findings suggest that male victims are more commonly blamed compared to female victims when either frame (bad offender or victim blaming) was used.

Discussion

The key finding of this study is that victim blaming commentary on intimate partner homicide news media reporting observed in social media and similar places is not a random phenomenon. Factors about the homicide (the gender of the offender and the victim) and the news story (how it is framed) impact on whether the public writes victim blaming commentary. Overall, this study found that almost a quarter of comments victim blame which is consistent with previous research (e.g., Whiting et al., 2019). In terms of gender, respondents less commonly blamed the victim in cases of female victims and male offenders, a finding that aligns with the limited research available (Komazec & Farmer, 2020). Given the victim blaming rhetoric facing many women of intimate partner violence (at least historically), this finding seems to suggest a possible change in community attitudes toward violence against women, with more culpability placed on the offender’s behavior than the victim’s.

Instead, victim blaming was more prominent in cases with male victims and female offenders. This might partly be explained by increased recognition that at least some of women’s violence in intimate relationships are acts of self-protection (Peterson, 1999). However, even in cases without provocation, male victims may still be blamed for their deaths. In the current study, blaming the male victim was observed even in vignettes that did not contain any victim blaming framing and where the female offender was portrayed as “bad.” In Western contemporary society, traditional stereotypes of men as strong, assertive, independent, brave, and powerful remain prominent. Perpetuations of such stereotypes may lead male victims to be reluctant to seek help due to a fear of being perceived as weak or vulnerable (Bates et al., 2019). Ultimately, this can result in under-funded services for male victims (Bates et al., 2019); male victimization receiving less attention from researchers, the criminal justice system, and the media (Komazec & Farmer, 2020); and lower societal concerns about male victimization (Andersen, 2013).

Another significant contribution of this paper is the finding that media frames influence victim blaming comments. Very few studies to date have explicitly tested whether reading a news story framed in a particular way makes a person more likely to respond to that story using similar themes in their commentary (e.g., Komazec & Farmer, 2020). Frames shape how the audience understands and remembers a topic (Entman, 1993). In other words, the more often the media portrays victims of crime in a specific manner, the greater the likelihood that the public will perceive victims in that same light. The effects of framing matter, since incorrect framing of victims and crime can lead to distorted perceptions of crime, victims, offenders, and the criminal justice system (Beckett & Sasson, 2003). Even further, media audiences not only use but also actively produce ideas and news, which means that their commentary is also influential in shaping public opinion about victims and crime (Gearhart et al., 2022; Lee & Jang, 2010). In the context of intimate partner homicide, this is important because it can lead to further victim blaming and misconceptions about the nature and context of such violence.

This paper found clear evidence of framing effects. Respondents (i.e., commenters) more commonly blamed the victim in comments when there was a victim blaming frame compared to a bad offender frame. Furthermore, the current study demonstrated that the gender of the offender/victim pair and the media frame also interacts to influence respondents’ comments. Respondents are more likely to blame victims when the victim is male (female offender) and there is a victim blaming frame. The results of this study clearly suggest that media framing can have impacts beyond simply shaping people’s attitudes about topics but also on how they communicate with others around them. In today’s online world, where opinions spread fast and misinformation is common (Del Vicario et al., 2016), it is important to understand the flow-on effects of media framing.

Of course, more things impact how and what people post online than simply how the article they just read was framed. Critics of the research on framing effects highlight that both media frames and personal experience shape the audience’s interpretation of media (Carragee & Roefs, 2004; Van Gorp et al., 2009). Given this critique, this study accounts for prior attitudes and personal experience by surveying the respondents and using an experimental approach to control for audience demographics and attitudes. Although media frames (or the gender of the offender/victim pairs) had the largest effect size, this paper found that another variable also statistically increased the likelihood of victim blaming. Respondents who lacked confidence in the criminal justice system were more likely to write victim blaming comments. Research has found that individuals have varying degrees of confidence in Australia’s criminal justice system (Moore, 2020). Individuals who lack confidence in the legal system may believe that it is ineffective or biased. This is a cause for concern, particularly given the correlation between victim blaming comments and individuals with lower levels of trust in the legal system. Given that personal experience also shapes audience’s interpretation of media (Carragee & Roefs, 2004; Van Gorp et al., 2009), it is interesting that other factors controlled for in this study were not statistically significant. In particular, prior research showed that the gender of the respondent impacts on framing effects (Bauman, 2018; Savage et al., 2017a); however, this study did not find a statistical difference for the gender of the respondent. Given the high proportion of respondents identifying as female in the current study (over 80%), further research is needed that oversamples males to achieve a more balanced gender ratio.

Implications for Policy and Practice

The findings of this research have important implications for policy and practice. This study shows that the media influences social media commentary, which highlights the importance of the media being mindful about how they frame intimate partner homicide. The Australian Press Council (2014), as well as government departments and other organizations (e.g., Our Watch), set out advisory guidelines relating to family and intimate partner violence, which are neither mandatory nor enforceable. One way to encourage adherence to media guidelines is to provide training and education for media professionals, especially since some journalists may be unaware of the media guidelines or how to implement them (Easteal et al., 2022; Sutherland et al., 2017). Education could occur in university curricula, through short courses for media professionals, or through on-the-job training (Sutherland et al., 2017). However, extending training to every media professional may be an expensive endeavor. An economical alternative is to have a designated intimate partner violence reporter who receives training on how to adhere to the media guidelines of family and intimate partner violence (Easteal et al., 2022). Unfortunately, there has been a shrinking of specialist reporters in recent years, indicated that wider cultural shift is needed to change media practices across the industry (Easteal et al., 2022). Shifting public perceptions of intimate partner violence to more closely align with reality also requires public education from government departments and educational organizations. For example, public campaigns can also be a powerful tool for enhancing public education on intimate partner homicide, helping to shift perceptions by recognizing harmful narratives like victim blaming (Wang, 2016).

Another implication stemming from this research is the importance of creating safe online communities. Around one-in-four comments were victim blaming in the current study, a result that is concerning given online comments can affect readers’ perceptions of the issues discussed (Gearhart et al., 2022). To address this, it is crucial to consider strategies such as comment moderation, which plays a key role in fostering secure online spaces (Gorwa et al., 2020). Research has highlighted various moderation approaches, such as engaging in dialogue, implementing pre- or post-moderation measures, closing comments at certain times, restricting discussion topics, modifying or removing content, and enforcing sanctions against rule violators (Ihlebæk & Krumsvik, 2015; Ruiz et al., 2011). However, moderation is not a simple task. Gillespie (2010) notes that no moderator is neutral, as moderation involves value-based judgments. These values are often detailed in guidelines or rules that moderators use to enforce sanctions or interventions (Kalsnes & Ihlebæk, 2021). Moderators must balance preventing harmful comments with maintaining freedom of expression (Karlina, 2021).

Future Research and Limitations

This paper adds significant knowledge in terms of understanding that the gender of offender/victim pairs and media frames impact victim blaming commentary. Nevertheless, some limitations warrant discussion. One challenge was the use of the experimental vignette design where respondents were asked to leave a comment as they would on social media. This approach was chosen as it allows researchers to use the same intimate partner homicide incident while altering the media frame. It also permits researchers to control for media usage, attitudes, and participant demographics. However, the experimental design did not provide respondents with the same “real world” Facebook experience, meaning that respondents could not see other comments and participate in dialogue as they would be able to in real life. Therefore, future experimental approaches could be amended to encourage more authentic interactions by allowing an ongoing dialogue between commenters. For example, a Facebook post could be simulated where respondents can see other comments and can contribute their own comments in turn. Providing a simulation of a Facebook post that is truer to life may enable respondents to provide more realistic comments. Furthermore, a limitation of using Facebook for participant recruitment is the potential for sampling bias, as the platform’s users may not represent the broader population. Future research could diversify recruitment methods by incorporating multiple platforms.

This paper has demonstrated that gender matters. However, it did not explore how alternative aspects of offender/victims’ cultural and social identity (such as race and social status) shape social media commentary. Further investigation is necessary to delve deeper into the intersection of gender and other factors in online media framing and its impact on audience commentary. For feasibility reasons, this paper also focused solely on pairs that were presented as male/female. This is because male/female pairings continue to be the most commonly reported relationship in media articles in the context of intimate partner violence. However, research suggests differences in how the media portrays intimate partner violence occurring in same-sex relationships compared to male/female pairings (Estes & Webber, 2021). Individuals who are attracted to the same sex or other gender identities face damaging stereotypes and prejudices regarding their sexual orientations (Lehavot & Lambert, 2007). Therefore, additional research is needed to examine how the media and audience commentary portray intimate partner homicide occurring in same-sex relationships and those with diverse gender identities and sexual orientations. Furthermore, the current study did not control for differences in motivation between male and female offenders. Future research could incorporate controls for gender-specific factors in offending, such as the role of prior victimization and cultural norms. This paper also used binary coding (0 or 1) to ensure consistency when coding variables. Future research could expand on this by exploring more nuanced coding frameworks to capture the complexity of variables.

Conclusion

This study utilized an experimental research design to examine whether victim blaming comments are influenced by the gender of the offender/victim and media frames. Very few studies have compared how the gender of offender/victim pairs influences social media commentary. Moreover, despite the influence of media frames on public perceptions of crime, there has been very little research specifically investigating whether media frames directly influence victim blaming comments. This study used an experimental vignette design to address these gaps. First, the analyses showed that respondents more commonly blamed the victim where there were female offender/male victim pairs compared to male offender/female victim pairs. Second, the analyses revealed that respondents more commonly blamed the victim when the article had a victim blaming frame compared to a bad offender frame. Finally, the analyses showed that the gender of the offender/victim pair and the media frame interacts to influence respondents’ comments. Respondents are more likely to blame victims when the victim is male (female offender), and there is a victim blaming frame. Therefore, this paper has demonstrated that the gender of the offender/victim pair and media frames impact on victim blaming comments. The findings from this paper have important implications for policy and practice. They highlight the necessity for improved training and resources for media professionals, as well as the importance of establishing safer online communities through comment moderation.

Supplemental Material

sj-docx-1-jiv-10.1177_08862605251322816 – Supplemental material for Victim Blaming, Gender, and Social Media Commentary: A Randomized Vignette Study of Audience Comments on News Reports of Intimate Partner Homicide

Supplemental material, sj-docx-1-jiv-10.1177_08862605251322816 for Victim Blaming, Gender, and Social Media Commentary: A Randomized Vignette Study of Audience Comments on News Reports of Intimate Partner Homicide by Emily Wright, Li Eriksson and Christine E. W. Bond in Journal of Interpersonal Violence

Footnotes

Data Availability

The dataset involves sensitive data (e.g., personal opinions) and are, therefore, not publicly available. Aggregate-level data can be provided on request to the first author.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research and/or authorship of this article.

Ethical Consideration and Consent to Participate

This research was conducted in accordance with ethical approval granted by the Griffith University Human Research Ethics Committee (reference number: 2019/133) on February 15, 2019. Respondents were asked to provide informed consent (by ticking a box) prior to beginning the survey. The survey preamble indicated the voluntary nature of participation, stating that respondents were not required to respond to every question and could withdraw from the survey at any point. The survey was completed anonymously, and contact information for support services related to intimate partner violence was provided to respondents.

ORCID iDs

Emily Wright

Li Eriksson

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Emily Wright, PhD, has undertaken research at Griffith University on the topic of gender and intimate partner homicide in online media. Her main research interests are crime and media, intimate partner homicide, and audience interaction with the media.

Li Eriksson, PhD, is a senior lecturer at the School of Criminology and Criminal Justice at Griffith University and member of the Griffith Criminology Institute. Her research examines developmental and contextual predictors of lethal and non-lethal violence, including the role of coercive control, adverse childhood experiences, offending history, and substance use.

Christine E. W. Bond, PhD, is a professor and dean (learning futures) in the Office of the Deputy Vice-Chancellor (Education) at Griffith University as well as a member of the Griffith Criminology Institute. Her research focuses on justice responses to domestic and family violence, as well as bail and sentencing.

References

Abbey

J. D.

Meloy

M. G.

(2017). Attention by design: Using attention checks to detect inattentive respondents and improve data quality. Journal of Operations Management, 53(1), 63–70. https://doi.org/10.1016/j.jom.2017.06.001

ABC News . (2015, May 15). Woman dies, man in custody after double shooting on suburban road in Hobart. https://www.abc.net.au/news/2015-05-14/woman-dies,-husband-in-custody-after-double-shooting-in-hobart/6470372

Andersen

T. H.

(2013). Against the wind: Male victimization and the ideal of manliness. Journal of Social Work, 13(3), 231–247. https://doi.org/10.1177/1468017311410002

Anderson

A. A.

Brossard

Scheufele

D. A.

Xenos

M. A.

Ladwig

(2014). The “nasty effect”: Online incivility and risk perceptions of emerging technologies. Journal of Computer-Mediated Communication, 19(3), 373–387. https://doi.org/10.1111/jcc4.12009

Australian Press Council. (2014). Standards of practice. https://www.presscouncil.org.au/standards/statement-of-principles

Bates

E. A.

Klement

K. R.

Kaye

L. K.

Pennington

C. R.

(2019). The impact of gendered stereotypes on perceptions of violence: A commentary. Sex Roles, 81(1), 34–43. https://research.aston.ac.uk/files/44512382/Commentary_final.pdf

Bauman

S. K.

(2018). Battered women who kill: Stereotype influence through the media [Master’s thesis, The University of North Dakota]. UND Scholarly Commons. https://commons.und.edu/cgi/viewcontent.cgi?article=3166&context=theses

Beckett

Sasson

(2003). The politics of injustice: Crime and punishment in America. SAGE Publications.

Belknap

Larson

D.-L.

Abrams

M. L.

Garcia

Anderson-Block

(2012). Types of intimate partner homicides committed by women: Self-defense, proxy/retaliation, and sexual proprietariness. Homicide Studies, 16(4), 359–379. https://doi.org/10.1177/1088767912461444

10.

Carlyle

K. E.

Orr

Savage

M. W.

Babin

E. A.

(2014). News coverage of intimate partner violence: Impact on prosocial responses. Media Psychology, 17(4), 451–471. https://doi.org/10.1080/15213269.2014.931812

11.

Carragee

K. M.

Roefs

(2004). The neglect of power in recent framing research. Journal of Communication, 54(2), 214–233. https://doi.org/10.1111/j.1460-2466.2004.tb02625.x

12.

Del Vicario

Bessi

Zollo

Petroni

Scala

Caldarelli

Stanley

H. E.

Quattrociocchi

. (2016). The spreading of misinformation online. Proceedings of the National Academy of Sciences of the United States of America, 113(3), 554–559. https://doi.org/10.1073/pnas.1517441113

13.

Dobash

R. E.

Dobash

(2015). When men murder women. Oxford University Press.

14.

Dowler

(2003). Media consumption and public attitudes toward crime and justice: The relationship between fear of crime, punitive attitudes, and perceived police effectiveness. Journal of Criminal Justice and Popular Culture, 10(2), 109–126. https://www.researchgate.net/publication/285675137

15.

Easteal

Blatchford

Holland

Sutherland

(2022). Teaching journalists about violence against women best reportage practices: An Australian case study. Journalism Practice, 16(10), 2185–2201. https://doi.org/10.1080/17512786.2021.1886866

16.

Entman

R. M.

(1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51–58. https://doi.org/10.1111/j.1460-2466.1993.tb01304.x

17.

Estes

M. L.

Webber

G. R.

(2021). “More closeted than gayness itself”: The depiction of same-sex couple violence in newspaper media. Journal of Interpersonal Violence, 36(1–2), 712–734. https://doi.org/10.1177/0886260517736271

18.

Fairbairn

Dawson

(2013). Canadian news coverage of intimate partner homicide: Analyzing changes over time. Feminist Criminology, 8(3), 147–176. https://doi.org/10.1177/1557085113480824

19.

Fohring

(2018). What’s in a word? Victims on ‘victim’. International Review of Victimology, 24(2), 151–164. https://doi.org/10.1177/0269758018755154

20.

Gearhart

Coman

I. A.

Moe

Brammer

(2022). Facebook comments influence perceptions of journalistic bias: Testing hostile media bias in the COVID-19 social media environment. Electronic News, 17(1), 3–18. https://doi.org/10.1177/19312431221103127

21.

Gillespie

(2010). The politics of ‘platforms’. New Media & Society, 12(3), 347–364. https://doi.org/10.1177/1461444809342738

22.

Gorwa

Binns

Katzenbach

(2020). Algorithmic content moderation: Technical and political challenges in the automation of platform governance. Big Data & Society, 7(1), 1–15. https://doi.org/10.1177/2053951719897945

23.

Harris

Honey

Webster

Diemer

Politoff

(2015). Young Australians’ attitudes to violence against women: Findings from the 2013 National Community Attitudes towards Violence Against Women Survey for respondents 16–24 years. Victorian Health Promotion Foundation. https://www.vichealth.vic.gov.au/sites/default/files/SurveyReport_YoungPeople-attitudes-violence-against-women.pdf

24.

Hille

Bakker

(2014). Engaging the social news user: Comments on news sites and Facebook. Journalism Practice, 8(5), 563–572. https://doi.org/10.1080/17512786.2014.899758

25.

Ihlebæk

K. A.

Krumsvik

A. H.

(2015). Editorial power and public participation in online newspapers. Journalism, 16(4), 470–487. https://doi.org/10.1177/1464884913520200

26.

Kalsnes

Ihlebæk

K. A.

(2021). Hiding hate speech: Political moderation on Facebook. Media, Culture & Society, 43(2), 326–342. https://doi.org/10.1177/0163443720957562

27.

Karlina

(2021). News site commenting policy and its ethical implications. Jurnal Peradaban, 1(1), 29–47. https://journal.paramadina.ac.id/index.php/jpb/article/download/499/224

28.

Komazec

Farmer

(2020). Family violence homicide in Australia: The effect of victim/offender gender on social media commentary. Journal of Gender Studies, 30(1), 4–17. https://doi.org/10.1080/09589236.2020.1754177

29.

Lee

E. J.

Jang

Y. J.

(2010). What do others’ reactions to news on Internet portal sites tell us? Effects of presentation format and readers’ need for cognition on reality perception. Communication Research, 37(6), 825–846. https://doi.org/10.1177/0093650210376189

30.

Lehavot

Lambert

A. J.

(2007). Toward a greater understanding of antigay prejudice: On the role of sexual orientation and gender role violation. Basic and Applied Social Psychology, 29(3), 279–292. https://doi.org/10.1080/01973530701503390

31.

Leisenring

(2006). Confronting “victim” discourses: The identity work of battered women. Symbolic Interaction, 29(3), 307–330. https://doi.org/10.1525/si.2006.29.3.307

32.

Marques Nascimento Macêdo

N. M.

Bispo

M. d. S

. (2023). Qualitative vignettes drawing on real cases as method in organizational research. Canadian Journal of Administrative Sciences/Revue Canadienne des Sciences de l’Administration, 40(2), 188–198. https://doi.org/10.1002/cjas.1692

33.

Meyers

(1994). News of battering. Journal of Communication, 44(2), 47–63. https://doi.org/10.1111/j.1460-2466.1994.tb00676.x

34.

Miles

Bricknell

(2024). Homicide in Australia 2022–23 (Statistical Report No. 46). Australian Institute of Criminology.

35.

Moore

(2020). Public confidence in the New South Wales criminal justice system: 2019 update. Crime and Justice Bulletin, 227, 1–38.

36.

Näsi

Tanskanen

Kivivuori

Haara

Reunanen

(2021). Crime news consumption and fear of violence: The role of traditional media, social media, and alternative information sources. Crime & Delinquency, 67(4), 574–600. https://doi.org/10.1177/0011128720922539

37.

Otieno

Spada

Renkl

(2013). Effects of news frames on perceived risk, emotions, and learning. PLoS One, 8(11), e79696. https://doi.org/10.1371/journal.pone.0079696

38.

Peterson

E. S.

(1999). Murder as self-help: Women and intimate partner homicide. Homicide Studies, 3(1), 30–46. https://doi.org/10.1177/1088767999003001003

39.

Pickett

J. T.

Mancini

Mears

D. P.

Gertz

(2015). Public (mis) understanding of crime policy: The effects of criminal justice experience and media reliance. Criminal Justice Policy Review, 26(5), 500–522. https://doi.org/10.1177/0887403414526228

40.

Richards

T. N.

Gillespie

L. K.

Smith

M. D.

(2011). Exploring news coverage of femicide: Does reporting the news add insult to injury? Feminist Criminology, 6(3), 178–202. https://doi.org/10.1177/1557085111409919

41.

Roberts

Indermaur

(2007). Predicting punitive attitudes in Australia. Psychiatry, Psychology and Law, 14(1), 56–65. https://doi.org/10.1375/pplt.14.1.56

42.

Ruiz

Domingo

Micó

J. L.

Díaz-Noci

Meso

Masip

(2011). Public sphere 2.0? The democratic qualities of citizen debates in online newspapers. The International Journal of Press/Politics, 16(4), 463–487. https://doi.org/10.1177/1940161211415849

43.

Savage

M. W.

Scarduzio

J. A.

Harris

K. L.

Carlyle

K. E.

(2017a). A mixed method study of intimate partner violence: How sex and severity impact perpetrator prior involvement, abnormality, and responsibility. Violence and Victims, 32(5), 897–918. https://doi.org/10.1891/0886-6708.VV-D-16-00100

44.

Savage

M. W.

Scarduzio

J. A.

Lockwood Harris

Carlyle

K. E.

Sheff

S. E.

(2017b). News stories of intimate partner violence: An experimental examination of participant sex, perpetrator sex, and violence severity on seriousness, sympathy, and punishment preferences. Health Communication, 32(6), 768–776. https://doi.org/10.1080/10410236.2016.1217453

45.

Scarduzio

J. A.

Carlyle

K. E.

Harris

K. L.

Savage

M. W.

(2017). “Maybe she was provoked” Exploring gender stereotypes about male and female perpetrators of intimate partner violence. Violence Against Women, 23(1), 89–113. https://doi.org/10.1177/1077801216636240

46.

Shenton

A. K.

(2004). Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22(2), 63–75. https://doi.org/10.3233/EFI-2004-22201

47.

Simington

A. L.

Farmer

(2024). Australian news media portrayals of filicide: depictions of mental health and the flawed mother trope. Crime, Law and Social Change, 82, 563–588. https://doi.org/10.1007/s10611-024-10158-6

48.

Spencer

C. M.

Stith

S. M.

(2020). Risk factors for male perpetration and female victimization of intimate partner homicide: A meta-analysis. Trauma, Violence, & Abuse, 21(3), 527–540. https://doi.org/10.1177/1524838018781101

49.

Surette

(2011). Media, crime, and criminal justice: Images, realities, and policies (4th ed.). Wadsworth.

50.

Surette

Otto

(2001). The media’s role in the definition of crime. In Henry

Kleck

Lanier

M. M.

Adler

M. J.

Farr

K. A.

Gertz

Gibbons

D. C.

Gould

L. C.

Greer

Hagan

(Eds.), What is crime?: Controversies over the nature of crime and what to do about it (pp. 148–163). Rowman & Littlefield.

51.

Sutherland

Easteal

Holland

Vaughan

(2019). Mediated representations of violence against women in the mainstream news in Australia. BMC Public Health, 19(1), 502. https://doi.org/10.1186/s12889-019-6793-2

52.

Sutherland

Simons

Blatchford

(2017). News media and the primary prevention of violence against women and their children (emerging evidence, insights and lessons). Our Watch. https://media-cdn.ourwatch.org.au/wp-content/uploads/sites/2/2019/11/07040030/Emerging-evidence-guide-media-AA.pdf

53.

Tabachnick

B. G.

Fidell

L. S.

(2013). Using multivariate statistics (6th ed.). Pearson Education.

54.

Taccini

Mannarini

(2024). News media representation of intimate partner violence: A systematic review. Sexuality Research and Social Policy. Advance online publication. https://doi.org/10.1007/s13178-023-00922-z

55.

UCLA: Statistical Consulting Group. (n.d.). Deciphering interactions in logistic regression. https://stats.oarc.ucla.edu/stata/seminars/deciphering-interactions-in-logistic-regression/

56.

United Nations Office on Drugs and Crime. (2023). Global study on homicide 2023. UNODC. https://www.unodc.org/unodc/en/data-and-analysis/global-study-on-homicide.html

57.

Van Gorp

Vettehen

P. H.

Beentjes

J. W

. (2009). Challenging the frame in the news: The role of issue involvement, attitude, and competing frames. Journal of Media Psychology, 21(4), 161–170. https://doi.org/10.1027/1864-1105.21.4.161

58.

Waddell

T. F.

(2018). What does the crowd think? How online comments and popularity metrics affect news credibility and issue importance. New Media & Society, 20(8), 3068–3083. https://doi.org/10.1177/1461444817742905

59.

Wang

(2016). Factors influencing attitude toward intimate partner violence. Aggression and Violent Behavior, 29, 72–78. https://doi.org/10.1016/j.avb.2016.06.005

60.

Whiting

Dansby Olufuwote

Cravens-Pickens

J. D.

Banford Witting

(2019). Online blaming and intimate partner violence: A content analysis of social media comments. The Qualitative Report, 24(1), 78–94. https://doi.org/10.46743/2160-3715/2019.3486

61.

Wlezien

Soroka

(2024). Media reflect! Policy, the public, and the news. American Political Science Review, 118(3), 1563–1569. https://doi.org/10.1017/S0003055423000874

62.

Wolfgang

M. E.

(1957). Victim precipitated criminal homicide. The Journal of Criminal Law, Criminology, and Police Science, 48(1), 1–11. https://doi.org/10.2307/1140160

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB