Lies, Damn Lies, and Military Sexual Assault Statistics

News analysis: The fuzzy math behind claims of rampant male sexual assault in the military

September 19, 2014

In discussions of the serious issue of sexual assault in the military, one frequently hears some version of what Sen. Barbara Boxer (D., Calif.) told Chris Matthews last year during a segment devoted to the matter on Matthews' show, Hardball:

Well, Chris, half of the victims are men … We're talking about violence. This isn't about, you know, somebody looking at someone and winking at them. This is about serious violence, men on men, men on women.

There is no doubt that there exist cases of sexual assault in the military in which the victims are men. It must be no less horrible than those cases in which the victims are female, and the efforts to prevent male-on-male sexual assault, and to punish the perpetrators, must be no less severe.

But I have a confession. Despite the substantial amount of news coverage and political discussion the issue of male military sexual assault has garnered in the last year or so, it was not something that came up, even once, during my seven years as an active duty Marine.

Perhaps that is because I was blind to the epidemic all around me. For an epidemic surely seems to be what is going on, according to Sen. Boxer and others. If there is a lot of sexual assault in the military, which everyone says there is, and if half of it victimizes men, which people like Sen. Boxer say is the case, ergo there must be a lot of sexual violence in the military where the victims are men.

We need not take Senator Boxer’s word for it. GQ has an emotionally moving cover story this month on the issue: ‘Son, Men Don’t Get Raped.’ Its opening sentences set the tone: "A warship is like a city—sprawling, vital, crowded with purposeful men and women. But on a warship, as in a city, there are people who will see you not as their friend or their neighbor but rather as their prey." As the story goes on, things get ever more grim:

Military culture is built upon a tenuous balance of aggression and obedience. The potential for sexual violence exists whenever there is too much of either. New recruits, stripped of their free will, cannot question authority. A certain kind of officer demands sex from underlings in the same way he demands they pick up his laundry.

The piece, for which GQ’s correspondent, Nathaniel Penn, interviewed 23 victims of male-on-male sexual assault, paints a disturbing picture of predators and their prey. Their stories are terrible and they deserve sympathy. And 23 interviewees is an improvement on earlier journalistic efforts to discuss the issue—in June of 2013, the New York Times found only six people for its own story on the matter.

Of course, just as my absence of anecdotes does not prove—nor do I intend it to prove—the absence of the problem in the military, nor do GQ’s 23 anecdotes prove the existence of a broader trend. And Penn doesn’t intend them to: He has numbers! Disturbing ones:

The moment a man enlists in the United States armed forces, his chances of being sexually assaulted increase by a factor of ten. Women, of course, are much more likely to be victims of military sexual trauma (MST), but far fewer of them enlist. In fact, more military men are assaulted than women—nearly 14,000 in 2012 alone.

14,000 is a lot of sexual assaults, even in a force with well over a million active duty members, and slightly over a million male active duty members. Taking the DOD’s 2012 figure for the number of male active duty military members, 14,000 assaults on men would constitute a rate of 1.18%—that is, slightly better than one-in-a-hundred odds of being sexually assaulted as a male in the armed forces.

What is the source of this number, and of the equally alarming larger number—26,000—that Penn quotes Sen. Kirsten Gillibrand (D., N.Y.) stating is the total number of annual military sexual assaults? Penn doesn’t say. Perhaps he is relying on the New York Times, which in its 2013 piece quoted similar statistics:

In its latest report on sexual assault, the Pentagon estimated that 26,000 service members experienced unwanted sexual contact in 2012, up from 19,000 in 2010. Of those cases, the Pentagon says, 53 percent involved attacks on men, mostly by other men.

"The Pentagon estimated…" And indeed, these numbers do come from the Pentagon, specifically from its 2012 ‘Workplace and Gender Relations Survey of Active Duty Members.’ The survey is conducted every two years, and a new one will be released in April of next year. So, the military surveyed its members and 26,000 service-members, among them 14,000 males, said they had been victims of sexual assault?

Well, no. As the New York Times (but not Penn, or Gillibrand, or Boxer) points out, those figures are estimates. How were they arrived at? The survey in question was administered anonymously and online. Responses were received from 22,792 service members. So about 1.6 percent of the active-duty force got back to the surveyors, answering their questions about sexual assault?

Again, not exactly. As the New York Times (but not Penn, or Gillibrand, or Boxer) makes reasonably clear, the question the survey posed did not ask about "sexual assault." Instead, the survey asked about "unwanted sexual contact."

What is that? According to the report that presents the results of the survey:

Although this term does not appear in the Uniform Code of Military Justice (UCMJ), it is used as an umbrella term intended to include certain acts prohibited by the UCMJ. For the purposes of the 2012 WGRA survey, the term "unwanted sexual contact" means intentional sexual contact that was against a person’s will or which occurred when the person did not or could not consent, and includes completed or attempted sexual intercourse, sodomy (oral or anal sex), penetration by an object, and the unwanted touching of genitalia and other sexually-related areas of the body.

So it covers roughly the same territory as the broad legal definitions of the various degrees of sexual assault. But would those answering the online survey respond differently to a question about "unwanted sexual contact" than they would to a question about "sexual assault"? Who knows. Perhaps the survey currently underway will correct this obvious methodological problem.

In any event, of those who responded, 6.1 percent of women and 1.2 percent of men said they had experienced unwanted sexual contact in the year leading up to the survey. Then, in its 2013 ‘Annual Report on Sexual Assault in the Military,’ the DOD does a bit of math and extrapolates what the total figures would be on the assumption that the respondents to the online, anonymous survey accurately represent the military as a whole.

And there you go: the military estimates that 26,000 military members have been victims of unwanted sexual contact. And even though a significantly higher proportion of women in the survey said they had experienced unwanted sexual contact than men, because there are so many more men than women in the military (in 2012, about 200,000 women, compared to almost 1.2 million men) it turns out the numbers work out such that 14,000 men are estimated to have experienced unwanted sexual contact.

That’s how Senator Boxer is able to say that sexual assault in the military isn’t just a women’s issue, but a violence issue, as "half the victims" are men—note how, as the issue leaves the dry pages of DOD statistical reports and enters the political and journalistic conversation, qualification and careful language steadily drop away.

There is an epidemic of sexual assault, says the New York Times—with an attempt at precise language—and say GQ, Senator Boxer and Senator Gillibrand—with no such attempt. And half the victims are men.

How do the results of the survey compare with a substantially more concrete set of figures—the actual number of reported sexual assaults in the military? In terms of the raw numbers of reported sexual assault, the reported figures are substantially lower than what the survey predicts they should be.

In 2012, 3,374 sexual assaults involving a military member (as either the alleged perpetrator, victim, or both) were reported. In 2013, the most recent year for which data was available, the number was higher—5,061—which most attribute to the military’s efforts to increase the rate of reporting.

The effort to encourage more victims to come forward and report that they have been assaulted is a good one, and it is surely the case that sexual assault is an underreported crime. Nonetheless, there is a big gap between the concrete figure of reported assaults in 2012—3,374—and the 26,000 number generated by the survey question regarding unwanted sexual contact. The gap is only slightly narrowed by the 2013 numbers.

Is the 26,000 figure the "true" number of sexual assaults in the military? Again, who knows? If you are ideologically inclined to believe that there is an epidemic of military rape, then surely you believe that it is, and GQ, Senator Boxer, and Senator Gillibrand report the number without qualification, as do many others.

But what about that claim made by Senator Boxer and repeated by GQ's Penn—that "half the victims are men"? Not according to the DOD’s figures of reported assaults. For the 2013 figures, the report explains the gender breakdown as follows: For "unrestricted reports," where a criminal process is initiated, "the vast majority of victims in investigations tend to be female, under the age of 25, and of junior enlisted grades, respectively."

Source: DOD Annual Report on Sexual Assault in the Military, FY13

For "restricted reports," where the victim seeks assistance but does not want to initiate a criminal process, the "victims…were primarily female, under the age of 25, and of a junior enlisted grade."

Source: DOD Annual Report on Sexual Assault in the Military, FY13

What is striking about this demographic breakdown is not only that women are more likely to be the victims of an assault—as the survey also indicates would be the case—but that they are victims of this crime in numbers vastly disproportionate to their representation in the military as a whole.

So, GQ and Senator Boxer notwithstanding, there is a problem with the argument that this an issue of plain violence without any relationship to gender relations: the most reliable numbers available don’t support the claim at all. Of course, it is possible that men report sexual assaults at a rate even lower than the already low rates that women do. Yet again, who knows? One might be inclined to believe this, but even if this is so, the gap between what is actually reported and the 14,000 figure from the survey is actually off by an order of magnitude.

But why let statistics get in the way of good activism. This lets politicians get passionate over an issue about which, in reality, no one actually disagrees—that any sexual assault at all in the military is deplorable—and allows journalists to write sensationalistic and emotionally provocative copy about untold stories which further the cause of social justice.

Then again, for the journalists at least, one occasionally hears about an institutional preference for honesty over agitprop. The numbers show that the case of sexual assault in the military is, at the very least, complicated. They rather clearly suggest that male-on-male sexual assault constitutes a fraction of a larger problem that typically involves men and women in combination.

Victims of sexual assault, male or female, deserve sympathy and compassion. Lazy and tendentious journalists deserve none.

Published under: Barbara Boxer , Military