The problem came to our attention after a plea from Mark J. Perry, an economics professor at the University of Michigan-Flint and a scholar at the American Enterprise Institute. He said that this fact was proven wrong, and yet at least two of Holder’s speeches referring to this “fact” remained uncorrected on DOJ’s Web sites, year after year. As a result, he said, the statistic has been repeated on Web sites, in news articles and even in a book. He first raised this issue a year ago, to no avail. So this time he contacted The Fact Checker directly.
“I think Eric Holder’s statement deserves a ‘four Pinocchio’ rating (a ‘whopper’) for ‘undermining our democracy’ with false information, and I hope Glenn Kessler will respond to my e-mail request and put Eric Holder’s statement to the Washington Post ‘truth squad’ test,” he wrote.
Message received. This is certainly an important issue, and it is crucial to get the facts right.
Logically, Holder’s statement does not make much sense. Intimate-partner homicide is the leading cause of death? At the very least, intimate-partner homicide is a subset of all homicides, so one can easily see that a broader category of murder would be even higher. And, then, what about diseases?
As Perry demonstrated, Centers for Disease Control and Prevention data show that, for the year 2008 (the year before Holder’s speeches), cancer, heart disease, unintentional injury and HIV/AIDS all topped homicide. Then if you break out intimate-partner homicide, that ends up being seventh or eighth on the list (depending on whether you also include all homicides.)
Okay, so it’s pretty mysterious why the attorney general would make this claim. It turns out he was relying on a source he might have thought was impeccable: a report from his own Justice Department.
As best we and the Justice Department can determine, this all started with a 1998 study by the Justice Department’s Bureau of Justice Statistics, titled “Violence Against Intimates,” that examined the data concerning crimes committed by current and former spouses, boyfriends and girlfriends. The most recent data in the study were collected in 1996.
The study does not say that intimate-partner homicide is the leading cause of death among African American women; in fact, it makes a key point that the rate of such murders had dropped dramatically. “From 1976 to 1996, the number of murders of black spouses, ex-spouses, boyfriends and girlfriends decreased from 14 per 100,000 blacks age 20-44 to just under 4 per 100,000,” the report says. “The murder rate decreased an average of 6% a year.”
Fast forward to 2003. A study published in American Journal of Public Health, concerning 220 intimate-partner homicide victims in 11 cities, includes this sentence: “Femicide, the homicide of women, is the leading cause of death in the United States among young African American women aged 15 to 45 years and the seventh leading cause of premature death among women overall.” The source is listed as the BJS study from 1998.
Note that the sentence merely asserts that murder is the leading cause of death, though as far as we can tell such a figure does not appear in the BJS study.
Later in 2003, another study, with similar co-authors, makes the link to intimate-partner killings even stronger: “Women are killed by intimate partners — husbands, lovers, ex-husbands, or ex-lovers — more often than by any other category of killer. It is the leading cause of death for African-American women aged 15 to 45 and the seventh leading cause of premature death for U.S. women overall.”
The BJS study is again cited as the source, but this time it is not murder that is listed as the leading cause for African-American women, but intimate-partner murder. Somehow, while the source remained the same, the wording mysteriously morphed from all murders to just intimate-partner murders. But these facts cannot be found in the original 1998 BJS report.
Who published the study making the explicit link? The Justice Department’s National Institute of Justice Journal. (Jacquelyn C. Campbell, lead author of both studies, did not respond to a request for comment.)
Eventually, the factoid ended up in the bloodstream of information about intimate partner homicide. For instance, a month before Holder gave his speech, the University of Minnesota Institute on Domestic Violence in the African American community displayed the fact on its Web site: “Intimate partner homicide is the leading cause of death for African-American women ages 15 to 45.” The 2003 NIJ Journal article is listed as the source.
In light of this paper trail, the Justice Department admits Holder should have not used the fact in 2009, but officials note that he has not repeated it in recent speeches. Officials apparently became aware of a problem after a columnist in USA Today in 2011 called attention to the errant fact.
“To our knowledge, the statistic has not been cited by the Department in years,” spokesman Brian Fallon said. “It originated with — and was widely circulated by — multiple, trusted authorities, including the Justice Department’s independent research arm under then-Attorney General [John] Ashcroft. Those who invoked it over the years did so in good faith in the context of discussing the scourge of domestic violence. At this point, all sides would agree that the figure is, at best, out of date, and should no longer be used.”
Department officials said that in the coming days they planned to append a note to the Web pages in question making clear that the claim is not valid.
The Pinocchio Test
Perry called for Four Pinocchios, and if this request had been made shortly after Holder gave the speech, we might have been so inclined. (Note: The Fact Checker did not exist in 2009.) But regular readers know that we generally do not go far back in time to award Pinocchios, especially if the offending sentence has not been recently repeated.
The Justice Department should have been quicker to fix its Web site. But, given that officials have conceded the factoid is not valid, and plan to make that clear on Justice’s Web site, we see no reason to award Pinocchios at this time. There’s certainly a lesson here: When a fact sounds too startling to be true, always go back to the original source. Secondary sources can sometimes lead you astray.
Send us facts to check by filling out this form