![[Adobe Stock]](https://sp-ao.shortpixel.ai/client/to_webp,q_glossy,ret_img,w_750,h_375/https://www.psypost.org/wp-content/uploads/2026/11/asian-woman-criminal-750x375.jpg)
A systematic review of two decades of research indicates that women who kill their romantic partners are often evaluated through rigid gender stereotypes by the media, jurors, and judges. The findings, published in Deviant Behavior, suggest that defendants who do not conform to the expectation of a passive, helpless victim tend to face greater attributions of blame and harsher judgments. The authors argue that this reliance on simplified tropes can obscure the complex realities of domestic abuse, mental health struggles, and substance use in the courtroom.
Intimate partner homicide is the fatal killing of a current or former spouse, boyfriend, or girlfriend. Globally, about one in seven homicides is committed against a romantic partner. Men commit most of these killings, but estimates vary widely. One U.S. estimate puts men’s rate at four to five times women’s, while another found 60 to 70 women killing a partner for every 100 men.
When a woman kills an intimate partner, the act is frequently, though not always, preceded by prolonged domestic abuse or carried out in immediate self-defense. Many of these relationships involve bidirectional violence, a term describing situations where both partners engage in physical or psychological aggression. However, violence in a relationship is sometimes labeled as mutual even when one partner is primarily acting in self-defense. Misunderstandings about how mutual fighting and long-term abuse work can complicate how the legal system evaluates a woman’s actions.
Violent behavior clashes with traditional societal expectations of women as nurturing and gentle. Scholars have argued that, as a result, female offenders are often sorted into simplified clichés of being “mad” for acting out of mental illness, “sad” for being a battered victim, or “bad” for being evil or unnatural. A woman who breaks both the law and traditional expectations of femininity faces what sociologists call “double deviance,” which can lead the public and the legal system to view her as monstrous or abnormal.
Don’t miss the next big psychology finding
Every morning, PsyPost sends a short briefing on the most interesting new peer-reviewed research in psychology and neuroscience.
These characterizations are connected to benevolent sexism, a form of bias that views women with affection and protectiveness only as long as they conform to traditional, gentle roles. Researchers have argued that this protective attitude can turn punitive when women step outside those boundaries by using lethal force. In the press, coverage of women who kill after abuse frequently downplays the history of domestic violence they suffered. For example, a 2023 study of Scottish news media found that media portrayals of women who killed abusive partners were actually worsening even as understanding of domestic abuse improved, and that these problematic narratives were often shaped by courtroom remarks from judges.
To examine how these biases operate, Sally Crosland and her colleagues at the University of Huddersfield have been investigating perceptions of women who commit lethal partner violence. In a 2023 study by the same researchers, the team developed a questionnaire that identified four distinct categories of stereotypes about women who kill their partners.
Building on that work, the authors conducted a systematic review to see how these stereotypes and risk factors have appeared across the research literature over a 20-year period, noting that shifting social attitudes toward women warranted an updated look at how female defendants are characterized. The review’s concern is not whether these women should be held accountable, but whether judgments rest on the facts of each case or on assumptions about how women should behave.
A systematic review is a structured search for every study that meets a predetermined set of criteria. The researchers searched five major academic databases in August 2023 for peer-reviewed studies published in English between 2003 and 2023. To limit the variations caused by vastly different legal systems and cultural norms, they restricted their search to research conducted in the United Kingdom, the European Union, Norway, Iceland, Switzerland, the United States, Canada, Australia, and New Zealand. They also required that the studies focus on adult female perpetrators of intimate partner homicide rather than male offenders or female victims.
The initial database search yielded 682 articles, and after removing a retracted paper and duplicates, the team screened 600 studies by their titles and abstracts, narrowing the pool to 43 papers for full-text evaluation. A final sample of 24 studies met all criteria, encompassing a mix of mock-jury experiments, reviews of coroner and court records, media analyses, and case studies. One of the authors’ aims was to see whether the extent of stereotyping could be quantified. The included studies used widely different methods, and the researchers summarized the findings in prose rather than combining them statistically.
One prominent theme across the reviewed literature involved the courtroom use of Battered Woman Syndrome. This psychological concept was originally developed to explain the trauma and constant fear experienced by women subjected to prolonged domestic violence, using ideas like “learned helplessness” to explain why a person might feel unable to leave an abusive relationship. The review indicates that while introducing this framework in court can help jurors understand a defendant’s fear, it also carries the risk of reinforcing stereotypes. The review notes that the term “syndrome” can medicalize the woman’s behavior, framing her as mentally ill or irrational rather than as someone acting in reasonable self-defense.
In a 2003 mock-jury experiment of 316 students included in the review, researchers found that presenting a battered woman’s experiences through a post-traumatic stress disorder framework increased sympathy and credibility, though that added sympathy did not necessarily lead to more lenient verdicts. The PTSD framing also led jurors to view her behavior as a sign of mental instability, which the review notes may fit an insanity plea better than a self-defense claim. In a separate mock-jury experiment with 618 undergraduates, defendants described as having resisted their abusers in the past, rather than fitting the typical passive profile, received more guilty verdicts.
How jurors assign blame also depends heavily on the specific circumstances of the killing and the history of the relationship. In one mock-jury experiment, participants viewed women as more culpable if the prior abuse was emotional rather than physical. In another, they judged women more harshly if the killing happened while the partner was sleeping rather than during an immediate confrontation. In that second experiment, involving 171 participants, men were more likely than women to see the defendant as mentally stable and aware of her actions when she waited until her abuser was asleep, which the review interprets as a sign of greater perceived culpability.
Across the sample as a whole, whether the victim was asleep predicted guilty verdicts. Self-defense laws generally require an immediate threat, but the review notes that women in long-term abusive relationships may perceive danger as imminent even between attacks, a dynamic jurors may not understand.
Biases also appeared in actual courtroom sentencing. One reviewed study analyzed judges’ remarks across 26 Canadian sentencing decisions between 1974 and 2006 for women who killed their abusive partners. The analysis indicated that when a woman received a prison sentence, judges often minimized the violence she had suffered by pointing to mutual fighting or substance use, framing the relationship as a simple two-way conflict rather than abuse.
The review found support for the continued presence of the “mad,” “bad,” and “sad” stereotypes in both the courtroom and the press. Defendants who do not fit the “mad” or “sad” narratives are often labeled as “bad” and treated as cold, calculating, or monstrous. One included paper, a legal analysis of two Australian murders committed by lesbian couples, argued that queer women are especially likely to be portrayed as “legal monsters” or unnatural figures in court and the news, overshadowing evidence of their mental health struggles. The victims were not the women’s partners, but the review included the paper because it addressed how lesbian defendants are stereotyped.
The reviewed studies also showed that not all women who kill their partners do so in self-defense. In one review of 276 spousal homicide coroner files in Quebec, only 28 percent of the female offenders had a documented history of experiencing violence from the victim. Other motives identified across the 24 studies included revenge, arguments, separation, practical reasons, and sexual jealousy.
News coverage tends to focus on unusual cases. One reviewed analysis of 37 newspaper articles found that the press rarely reported on what the researcher called “normal” homicides, which typically involved unemployed couples, domestic disputes, and mutual drug or alcohol use. Because the press mostly covers unusual killings that seem to require a dramatic explanation, research that relies on news coverage may overestimate how often the “mad” and “bad” tropes apply to everyday cases, the author of that analysis suggests.
Substance abuse emerged as a major factor in the reviewed cases, frequently affecting both the perpetrator and the victim. In the same study that examined news coverage, 57 percent of the official case reports involved mutual alcohol or drug use, while a small study of 12 female offenders found that one-third had a history of alcohol abuse. Even so, judges in the Canadian sentencing study sometimes pointed to alcohol to minimize the abuse a woman had experienced, and other analyses suggested that a woman’s drinking can be seen as an individual problem that places her outside the image of a deserving victim.
There are a few things to keep in mind when interpreting these findings. Because the review summarized 24 studies with very different designs, ranging from small legal case studies to student mock-jury experiments, it could not measure how often stereotyping happens or how strongly it sways real juries. Although the included papers were published between 2003 and 2023, much of their underlying evidence is older, including Canadian sentencing decisions dating back to 1974, newspaper articles from 1978 to 2002, and a murder case from 1989. The review also found some evidence that attitudes are shifting over time, and noted that studies based on news coverage may overstate how often extreme stereotypes apply.
In addition, many of the official police, court, and coroner records analyzed in the underlying studies were incomplete and lacked formal data on whether the defendants had post-traumatic stress disorder or other mental illnesses at the time of the offense. The review was also restricted to English-language studies from Western nations, where legal definitions of self-defense and cultural attitudes toward gender roles may differ from other parts of the world.
Future research could compare how different cultures and legal systems treat female defendants, and gather more complete data on underrepresented groups, such as sexual and gender minority women. The authors suggest that developing better educational tools for legal professionals and expert witnesses could also help ensure that courtroom decisions are based on the facts of each case rather than outdated stereotypes.
The study, “Female Perpetrators of Intimate Partner Homicide: Stereotyping and Factors. A Systematic Review of Twenty Years of Literature, 2003–2023,” was authored by Sally Crosland, Calli Tzani, and Maria Ioannou.
