Abusing Statistics about Homosexuality
The topic of same-sex marriage is highly charged. The religious right are strongly opposed to it, and seem to feel so strongly about it that they think it’s okay to suspend the ninth commandment: “thou shalt not bear false witness against thy neighbour”
One lie in particular is quite blatant; the claim that heterosexual married couples have practically no problems of domestic violence and partner abuse, while for homosexual couples the problem is pervasive, even hundreds of times more pervasive.
Widely referenced to support this claim is the US Department of Justice 2000 report Extent, Nature, and Consequences of Intimate Partner Violence. Now this initially seems like an odd report to quote, as it does not support their claim, it actually says:
Then there’s a variety of figures and graphs, followed by this analysis, the importance of these figures will be made clear later:
Research on violence in same-sex relationships has been limited to studies of small, unrepresentative samples of gay and lesbian couples. Results from these studies suggest that same-sex couples are about as violent as heterosexual couples
It seems that the study the conservatives quote does not support their claim at all. So how can they possibly use it? Watch:
At first glance, these findings suggest that both male and female same-sex couples experience more intimate partner violence than do opposite sex couples. However, a comparison of intimate partner victimization rates among same-sex and opposite-sex cohabitants by perpetrator gender produced some interesting findings: 30.4 percent of same-sex cohabiting women reported being victimized by a male partner, whereas 11.4 percent reported being victimized by a female partner. Thus, same-sex cohabiting women were nearly three times more likely to report being victimized by a male partner than by a female partner. Moreover, opposite-sex cohabiting women were nearly twice as likely to report being victimized by a male partner than were same-sex cohabiting women by a female partner (20.3 percent and 11.4 percent) Somewhat different patterns were found for men. Like their female counterparts, same-sex cohabiting men were more likely to report being victimized by a male partner than by a female partner. Specifically, 15.4 percent of same-sex cohabiting men reported being raped, physically assaulted, and/or stalked by a male partner, but 10.8 percent reported such violence by a female
partner. However, same-sex cohabiting men were nearly twice as likely to report being victimized by a male partner than were opposite sex cohabiting men by a female partner (15.4 percent and 7.7 percent). These findings suggest that intimate partner violence is perpetrated primarily by men, whether against male or female partner
Note the curious mixture of truth and falsehoods. Yes, 11.4% of lesbians suffer abuse from their female partner. But no, the percentage of heterosexual women abused by a male partner is is not 0.26%, it’s 20.4%, and for men it’s not 0.05%, it’s 7.7%. It seems like James Watkins is just making up numbers. His assertion seems like a lie, and in fact the source he quotes actually suggests that lesbian couples are nearly three times LESS likely to suffer abuse than heterosexual couples.
According to the U.S. Department of Justice: Office of Justice Programs and "Intimate Partner Violence," the percentage of heterosexual wives abused is one-quarter of one percent (0.26); married heterosexual men, one-twentieth of one percent (0.05).
The same study revealed that 11.4 percent of women in lesbian relationships suffered abuse while 15.4 percent of men in gay relationships suffered abuse
The falsehood is repeated and magnified with graphics:
The falsehood was also repeated with better graphics by the Family Research Council in their 2004 handout “The Slippery Slope of Same-Sex Marriage”:
No such graph occurs in the report, and in fact the report paints a very different picture (graph created by me from the actual figures):
How does the real figure of 20.3% turn into the lie of 0.26%? Well, if you look at the legend of that graph, (and the FRC reference) they actually quote TWO studies. The second being “Intimate Partner Violence”, Bureau of Justice Statistics Special Report: 2002, page 11, which has this little info-box:
2.6 per 1,000 is 0.26%, so that’s the source of the figure. Now note that’s only the married figure, but even if we take the divorced/separated figure, it’s still only 3.19%, significantly lower than the 11.4% figure for lesbians. But then we also know that the actual comparable figure is 20.3%. So why is there a discrepancy between the two studies, surely the heterosexual figures should be similar if they are both measuring “intimate partner violence”.
The answer is quite simple. They are measuring totally different things. The DoJ study, with the high figures says:
Whereas the BJS study says:
The definition of intimate partner violence used in the NVAW Survey includes rape, physical assault, and stalking perpetrated by current and former dates, spouses, and cohabiting partners, with cohabiting meaning living together at least some of the time as a couple. Both same-sex and opposite-sex cohabitants are included in the definition. The survey’s definition of intimate partner violence resembles the one developed by CDC2 because it includes violence occurring between persons who have a current or former dating, marital, or cohabiting relationship and same-sex and opposite-sex cohabitants. However, it deviates from CDC’s definition because it includes stalking as well as rape and physical assault. “Stalking” is defined as a course of conduct directed at a specific person involving repeated visual or physical proximity; nonconsensual communication; verbal, written, or implied threats; or a combination thereof that would cause fear in a reasonable person, with “repeated” meaning on two or more occasions. The definition of stalking used in the survey does not require stalkers to make a credible threat against victims, but it does require victims to feel a high level of fear.
So, the DoJ study measures violence and stalking, and the BJS study measures actual violence. Obviously the results are going to be totally different. Vastly more people experience stalking (especially with the definition given above) than experience actual violence.
Violent acts covered in this report include murder, rape, sexual assault, robbery, and aggravated and simple assault.
Simple assault is an attack without a weapon resulting either in no injury, minor injury (such as bruises, black eyes, cuts, scratches, or swelling) or an undetermined injury requiring less than 2 days of hospitalization. Simple assaults also include attempted assaults without a weapon.
So that’s how you lie with statistics. First take a report that actually shows you are wrong, then remove the figures you disagree with, then take another report that measures something totally different (but also not supporting your case), take some figures from that, and combine then with what you left in the first report, then make a graph, and the put the official titles of both reports underneath it.
Statistics, and science in general, should not be used as a blunt weapon to bolster your case by mixing and cherry-picking points of data. It should be used to determine what actually is the case. Sadly it seems individuals and organizations like the FRC have already decided what they think the “truth” is, and they are not going to let pesky facts get in their way. So they change the facts. They lie, with statistics.