The American Psychological Association has retracted a report that claimed male-on-male sexual assault in the U.S. military may be 15 times higher than previously thought.
The research, a study published Nov. 3 in a special issue of the journal Psychological Services and reported by media outlets, including the Washington Times and U.K. Daily Mail, said a special survey of 180 Iraq and Afghanistan male combat veterans found the rate of military sexual assault among men to be 17.2 percent, instead of the 1.1 percent reported by traditional anonymous surveys.
If accurate, that would mean that as many as 159,000 male service members were sexually assaulted in 2014, as opposed to the estimated 10,600 men reported in a Rand Corp. study published last December.
The APA issued a release Sunday saying it retracted the article based on flawed data analysis.
"Although the article went through our standard peer-review process, other scholars have since examined the data and raised valid concerns regarding the design and statistical analysis, which compromise the findings," said APA publisher Gary VandenBos.
The study's lead author, Sean Sheppard, is a research fellow and doctoral candidate at the University of Utah's National Center for Veterans Studies.
He could not be reached for comment Monday.
"I have been unexpectedly called out of the country," he wrote in an email to Military Times.
For the study, researchers used a survey method called an "unmatched count technique" to gain more information about the accuracy of anonymous surveys of post-9/11 combat veterans.
A sample of veterans, recruited via "online websites for military populations," were split into two random groups, with the control group getting one set of questions and the other group receiving one additional yes/no question: "I was sexually assaulted while serving in the military."
All respondents then reveal only the number of questions to which they answered "yes."
It is thought that since respondents are most likely to answer truthfully to the additional question, the method improves accuracy.
Sheppard's article theorized that the stigma associated with disclosing a military sexual assault keeps troops from reporting them and even anonymous surveys don't offer enough guaranteed privacy to admit to being assaulted.
APA officials defended the special issue on military sexual trauma, saying it includes a dozen articles that indicate sex assaults in the military "occur more often than realized and are generally under-reported."
"One article having some problems with its statistical analysis should not undo the power and facts of the other 12 articles as a collection," VandenBos said.
Other articles in the special edition found that:
- Reluctance to admit to having been sexually assaulted may increase shame, delay treatment and impair recovery for male service members;
- Male vets who were sexually assaulted on active duty report more severe symptoms of depression and post-traumatic stress and higher rates of suicidal thoughts;
- Male troops who have been sexually assaulted while on deployment may not get the support they need from family and friends when they return home;
- A majority of women who were sexually assaulted on active duty report symptoms of PTS but just 14 percent sought medical treatment immediately following the assault, and 75 percent reported needing mental health services years after the incident.
Journal edition co-editor Michi Fu said the association wanted to publish a special issue after "hearing of reports of military sexual trauma being so much more prevalent than in the general population."
"We know that there is under-reporting among men and women and hope that this special issue will help to bring awareness and treatment for those that serve and protect us," Fu said.
An APA spokeswoman said Monday the association publishes between 4,500 and 5,000 articles a year in more than 90 journals. It has retracted two other articles this year.
Patricia Kime is a senior writer covering military and veterans health care, medicine and personnel issues.
