Could Women Be as Violent as Men?

The Data Doctor answers a question about "gender symmetry" in domestic violence

Posted Jul 14, 2015

Dear Sherry,

Can you send me the piece you have about “gender symmetry”? Is there a “graphic novel” version of the paper?  I’m only kind of kidding – as wonderful as your work is, it’d be great to have a “talking points” version for those of us in the field who get “Father’s Rights” folks calling in to radio talk shows.  They always seem to have great talking points on the tips of their tongues. 

I don’t mean to be disrespectful with my question.  Perhaps that’s not your job to create such talking points or comic books based on your findings – perhaps it’s mine.


Hi Ben,

Well, I didn't quite get to the comic book version, although that is still on my fantasy list of things to do! Here's the Cliff Notes version.

For readers who may not know this, there has been a controversy raging for some time in the domestic violence field about whether women are as violent as men. Women seek more help for domestic violence from law enforcement, shelters, and other services than men. Many research studies also show that there are more female than male victims.  However, some types of surveys find gender "symmetry"—similar rates of male and female victimization.  Because the reasons for these differences were not well understood, there have been arguments about which data are "right."

I am a co-author of the most commonly used scale that finds gender symmetry, the Revised Conflict Tactics Scales, which came out in 1996.  I worked very hard on it, as did my colleagues.  It was a sincere effort to develop a good measure of domestic violence.  It has some good features, for example, the revision used gender neutral language that applies to same-sex couples as well as heterosexual couples. 

However, science is about progress and that measure (and others like it) is not perfect. One problem is that the CTS2 typcially makes males and females look similarly violent, and that is not accurate.  I no longer recommend that people use the CTS2.  The good news is that there are many other new measures that do a better job of measuring gender patterns domestic violence. Also, we are finally starting to realize that there are ways that we measure domestic violence in a lot of research that systematically under-represent the sorts of (mostly male-perpetrated) severe violence that are most concerning to law enforcement and health care professionals.

Ben mentions the Father's Rights movement.  There are a number of organizations that assert that men are unfairly depicted in research and media.  Many of these pro-symmetry groups use a debating technique known as the "straw man" argument. A "straw man" argument" involves making your opponent's argument seem more simplistic than it really is. This makes it easier to knock down the argument, just as it is easier to knock down a straw man versus a real one. [Note: historically the name for this technique has been gendered, but for the rest of the article I will refer to it as the "straw person" argument.]

The straw person argument that is often used by proponents of symmetry is to speak as if the non-symmetry position is equivalent to saying that women are never violent. This could not be further from the truth.  No one thinks that women are never violent.  

However, there is overwhelming evidence that men commit domestic violence more often than women, just as they commit all other types of violent crime more often than women. Most measures suggest there are about 2 to 4 male perpetrators for 1 female perpetrator, although some measures find even more males perpetrators.

Ways That DV Measures Systematically Under-Represent Men's Violence

Example #1:  Domestic violence homicide rates indicate about 3 to 4 men murder their partner for every female who murders their partner in the U.S. However, even this statistic systematically underrepresents the homicidal behavior of male batterers versus female batterers. Why?  It is inaccurate because our current system for tracking DV homicide only counts the actual intimate partner as the victim. We know, however, that some batterers kill not only their intimate partner (or ex-intimate partner). Batterers sometimes also kill the victim's children, extended family members, and, most commonly, the new intimate partner of women trying to move on past the abusive relationship. Perhaps most unfortunately, in the latter case the murder is put into the "acquaintance" or even "stranger" category in homicide statistics, which does not capture the reason for the homicide.

Men are responsible for almost all of these "collateral" murders.  Although it is hard to get precise estimates because of the flaws in our national surveillance of these crimes, one study of 17-years of murder in Michigan found that 100% of the intimate partner murders involving "collateral" victims were perpetrated by men (Meyer & Post, 2013). A more comprehensive understanding of DV-motivated homicide shows that abusive males are responsible for even more intimate-partner related murders than we already recognize.

Example #2. Better questions produce survey results that match arrest, homicide, and other data that find more male perpetrators. For example, every year, the Centers for Disease Control and Prevention (CDC) administers a large, school-based survey on a variety or risk behaviors, including teen dating violence called the Youth Risk Behavior Survey (YRBS; Vagi et al., 2015). For many years the answer to the YRBS dating violence question seemed to support the gender symmetry position, with approximately 10% of males and 10% of females reporting being victimized by a dating partner. However, recently they changed the measurement. They re-worded the item on physical victimization and added an item on sexual victimization. Because the defining characteristic of intimate relationships is their intimacy, it is extremely important to include sexual victimization in the operationalization of dating or domestic violence . These two changes in the YRBS dramatically altered the results. Now, the overall rate of dating violence victimization is approximately 20% for females and 10% for males. Further, more female than male victimization was found for each category of violence: physical victimization only, sexual victimization only, and the combination of the two.

Violence is intentionally causing unwanted harm. Unfortunately, many of our measures do not assess intent, do not assess wantedness, and do not assess harm. Thus, all sorts of events, ranging from accidents to pillow fights, sometimes get reported and scored as violence. Without assessing the intent to cause unwanted harm, even acts such as tackling in football or pushing someone out of harm's way could be scored as violence. New research especially suggests that horseplay and joking around are often mistakenly included in responses to many checklists that are supposed to measure violence. New tools fix this problem.  My new scale, the Partner Victimization Scale (PVS) produced a lifetime DV rate of 34.1% for women and 18.7% for men in a large community sample, bringing self-report data more in line with what we know about non-fatal DV from other sources, such as reports to law enforcement, arrests and helpseeking.

Looking to the future

The good news is that, for the first time in 40 years, real alternatives, such as the new CDC items or other new scales, are available that will help bring survey measures into line with other sources of information.  We need to fix this problem, just like it is important that when two medical tests give a different result, such as a mammogram and a biopsy, that we have clear ways of reconciling these differences.  Hopefully, these new measures will help at long last to provide a scientific conclusion to the long-running arguments about gender symmetry versus asymmetry and help everyone in the field recognize that the best description of domestic violence is a pattern of moderate gender asymmetry. Perhaps even more importantly, these more accurate measures will help us better develop and evaluation prevention and intervention. For example, understanding that female victims of serious battering may fear not only for their own lives but also the lives of their children and other loved ones is important to designing better safety planning. More sensitive and specific assessment will also hopefully help us better understand which prevention programs are best for reducing domestic violence and eventually reducing the burden that domestic violence places on too many families and on societies across the world.

A Note on the Hoffman Report and APA's role in permitting psychologist involvement in torture.

This has been a very difficult week for the field of psychology after the release of the Hoffman Report finding that, for many years post 9/11, the American Psychological Association colluded with the Department of Defense to find ethical loopholes that would allow psychologists to participate in waterboarding and other torture of terrorist detainees.  Prominent psychologists produced undisputed rulings that sleep deprivation and other techniques were not torture. 

It is thanks to psychological science that we know that sleep deprivation and other psychological and behavioral techniques are every bit as harmful as physical violence.  This makes it all the more sad to see psychology misused in this way.  Many senior people at APA have already departed. 

It is going to take some time to sufficiently digest and address this report, but APA and the field more generally must find a way to ensure that psychologists are not involved in torture.  I congratulate those who persisted in pursuing this issue, including Jean Maria Arriago and Stephen Reisner and James Risen.  It is a sad day for psychology but I hope the light of the truth will help us find a way forward.

Ask the Data Doctor appears on Tuesdays.  I respond to questions about research, therapy, and policy.  Have a question?  Send an email to or

Further reading:

Hamby, S. (2014). Battered women’s protective strategies: Stronger than you know.  New York:  Oxford University Press.

Hamby, S. (online first). Self-report measures that do not produce gender parity in intimate partner violence: A multi-study investigation. Psychology of Violence.

Hamby, S. (2014). Intimate partner & sexual violence research:  Scientific progress, scientific challenges, and gender. Trauma, Violence, and Abuse.

Hamby, S. (2009).  The gender debate about intimate partner violence:  Solutions and dead ends.  Psychological Trauma, 1(1), 24-34.