Are Gay & Lesbian Couples as Violent as Straight Couples?

The Data Doctor answers a question on the dark side of marriage.

Posted Jul 07, 2015

Dear Advice Doctor:

I hope the summer is not too hot in Sewanee. We are having a heat spell here, which is the high 80s with no tradewinds. I suppose we cannot complain. I was just in Phoenix where it was 114 in the daytime, and 95 at midnight. And I don't care what they say about dry heat - hot is hot.

I was wondering if I could get your thoughts on the NISVS study results for LGB respondents. I am doing some work with the WNBA (fun!!) and one of the trainers wanted to cite the survey findings that LGB respondents were as or more likely to have been victims of IPV. I have always had a problem with this notion that the incidence of lesbian and gay male same-sex violence is as frequent as in straight couples. I have done a lot of work in this area, including with local and national LGBTQ groups and this seems to be the party line without much substantive data. And I don't MIND if it is true, it just is not my gut feeling that it is true. More importantly, I always say that until we are totally rid of societal heterosexism and LGBTQ hatred, we can and will never know the actual incidence of same-sex IPV because both victims and those who harm them will have to come out (to themselves and to others) as queer before they report being in IPV relationships. 

It seems to me that the problem with the NISVS study is that this finding implies that LGB respondents have been victimized by same-sex partners, but my reading of the survey results is that it is more likely to be by male perpetrators for lesbians, gay men and bisexual men or women. Do I read this correctly? My second question is related to the finding that of the 43.8% of women who identified as lesbian, 67.4% reporting having only female perpetrators (p 2 summary). I know you cannot exactly extrapolate a percent of a percent, but can you help me understand what that means? If less than half of females identified as lesbian, and over two-thirds of those lesbians were abused by lesbian partners, does that mean it is equivalent to the victimization rates for straight women?

Thanks much for your help and for all you do,

Happy summer,


Dear Kalei,

This is a great question to ponder as we celebrate marriage equality for all people.  There are dark sides to relationships and it is important to understand those. 

The National Intimate and Sexual Violence Survey (NISVS) conducted by the CDC is an important study on domestic violence and the CDC report you mention is an important study on violence in the LGBTQ community. 

Still, as you note, there are a lot of complexities and this report is best understood as an early effort to understand the problem.  The CDC report is better than many because it does ask a question on sexual identity:  "Do you consider yourself to be heterosexual or straight, gay or lesbian, or bisexual?" 

Of course it is widely known that sexual identity emerges in a developmental trajectory that can vary from one person to the next.  There are many paths to forming a sexual identity.  The discrimination you mention plays a role in that because it can be one reason that some people explore relationships with opposite-sex partners even if they eventually adopt a lesbian or gay sexual identity.  Sexual identity is not sexual behavior and people who identify as gay, lesbian or straight can have sexual partners that do not "match" these identities. 

An example is in the movie The Kids Are All Right.  Julianne Moore's character, who identifies as lesbian, has an affair with Mark Ruffalo and then breaks it off by telling him she's gay (which he already knew).

You can see that in these data because many respondents who identified as lesbian or gay reported victimization by opposite-sex intimate partners.  Additionally, a few people who identified as straight reported victimization by same-sex intimate partners. 

You ask 2 questions.  The first is about overall prevalence rates, the second looks only at the group of victims and who their perpetrators were.  This involves a lot of percentages that are unfortunately scattered throughout the document and are hard to follow.

One important fact to know in understanding the CDC data are the very low rates of people who reported anything but "straight" in answer to the sexual identity question:

"The sexual orientation of the sample included 96.5% females identified as heterosexual, 2.2% bisexual, and 1.3% lesbian. For males, 96.8% identified as heterosexual, 1.2% bisexual, and 2.0% gay." (p 6.).

The CDC sample is HUGE by the standards of most psychology research—16,507 people!  However, these low rates still means that only 130 women identified as lesbian, for example. 

That's already getting kind of low even for comparing victimization rates for all lesbians to victimization rates for other women.  It's one reason why the rates are not statistically different. 

To get at differences in the gender of perpetrators means slicing and dicing those 130 women into even smaller categories.  First, as you said, 43.8% of them reported a victimization.  That's 43.8% of 130 so now we are down to 57 self-identified lesbians who have experienced domestic violence. 

Then, slicing and dicing again, 67.4% of those 57 victimized lesbians said they had only had female perpetrators—so 38 women.  The rest of that group of 57, 19 women, reported at least one male IPV perpetrator. 

Personally, I think one striking thing about these data is the point that you made—that even among women who identify as lesbian a lot of their domestic violence victimization comes at the hands of men. 

Maybe even more strikingly, both bisexual women and bisexual men report mostly male perpetrators.  As you note, so do gay males and straight women. These are interesting findings, consistent with men being more likely to perpetrate physical assault against other targets.

However, from a research standpoint, one of the other big takeaways from this study is that it is very hard to get good nationally representative data on violence in the LGBTQ community. 

Another way to think about those numbers is 57 lesbians who reported domestic violence is barely more than one participant per state.  You've probably seen more in your own work. 

Bottom line:  This is not the kind of data to overrule years of clinical and community experience.  Your gut feeling about it matches mine.  Yes, domestic violence happens in lesbian, gay and bisexual relationships.  The data are good enough to know that there is no segment of society that is immune from domestic violence—not the rich, not the famous, and not people in same-sex relationships.  That's what I suggest you and your colleague say in trainings.

I would not use this as strong evidence about whether lesbian or bisexual women have higher or lower rates, because changing the answers of just a few women would have changed the results.

We have not yet figured out a way to get good comparative data on LGBTQ versus straight relationships.  And to add one final wrinkle, there are forms of abuse—such as threatening to "out" someone who does not want to be outed—that are unique to LGBTQ relationships and are not included in most research. 

Congrats on the WNBA gig!  That does sound fun.  And I'd like to return the thanks for all you do and all those you inspire (including me!).

--The Data Doctor

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