RADAR ALERT:
Lies, Damned Lies, and DV Statistics

Correction: Since issuing this alert, RADAR has discovered that the statement "there has not been a threefold increase" is not correct. Jane Doe Inc's confusing presentation in their document counting DV-related homicides, and the contradictory statements they made to different media outlets confused us as well. (For further details, see below.)

Two weeks ago the Boston Herald ran an inflammatory story claiming "by New Year's Eve 57 people will have died this year in Massachusetts because of domestic violence - a toll not seen in 15 years." The death toll information was compiled by the Massachusetts Coalition Against Sexual Assault and Domestic Violence, a.k.a. Jane Doe Inc. The spokeswoman for Jane Doe Inc. declared the situation a "crisis" and predictably called for more funding. This flawed article was the focus of RADAR's December 12 Alert.

This past week the Boston Globe ran a copy-cat story quoting a Jane Doe Inc. spokeswoman who made the same claim: http://www.boston.com/news/local/articles/2007/12/14/man_charged_with_killing_wife_and_her_2_daughters

But the Jane Doe spokeswoman neglected to mention one critical fact: Jane Doe Inc. started counting domestic violence homicides differently in 2006 -- see http://www.janedoe.org/involved/Narratives%202007.pdf

Jane Doe Inc's projection of 57 "domestic-violence-related" deaths by year end is based on the 42 deaths they'd counted under their new rules as of September 21, 2007. Among the deaths Jane Doe Inc. included were 3 people killed in Florida who had once lived in Massachusetts as well as 9 perpetrators who committed suicide.

So the Boston Globe article erroneously reports, "Over the past two years, the number of domestic violence homicides has almost tripled, from 15 in 2005." But obviously you can't compare numbers based on different counting methods. By using their new counting method, Jane Doe Inc. has artificially inflated the count by 40% over what it would have been using the methods in effect in 2005.

There's more to the disinformation.

Based on figures published in the Globe's annual Dec. 31 "In Memoriam" editorials, the number of DV-related deaths in Massachusetts fluctuates dramatically from year to year. These fluctuations are unrelated to changes in funding. Although there were 6 fewer DV-related deaths in Mass. in 1998 when Violence Against Women Act funding jumped by 63%, maintaining the funding at the newly increased level didn't prevent 12 additional deaths the following year. And when Federal funding dropped by 10% in 2001 due to across the board budget cuts, the number of DV-related deaths in Mass. decreased rather than increased as compared to the previous year. (See http://holt.house.gov/pdf/CRSonVAWADec2005.pdf for the Congressional Research Service summary of VAWA fund allocations.)

The following graph shows the death counts given in the Globe's "In Memoriam" editorials. In some years the number has dropped as low as 14, while the number in the year before or after is double that. The 2007 count fits that same general pattern.

Even one death is one too many. But there has not been a threefold increase, as Jane Doe Inc., is claiming. (See correction below.)

The conclusion is clear: Jane Doe Inc. is playing fast and loose with the numbers to create a crisis atmosphere and pressure legislators to enact more DV laws.

Please contact the Globe and politely inform them that the projected threefold increase in DV-related deaths is based on a flawed apples-to-oranges comparison designed to sway Massachusetts lawmakers:

• Call the City Editor at 617-929-3100 and request the Globe to run a correction.
• Send an email to the letters editor at letter@globe.com.

Please also send a copy of your email to RADAR at info@mediaradar.org.

Correction details: Just a few weeks ago Jane Doe Inc. told the Boston Herald they were projecting 57 DV-related deaths in Massachusetts by year end. Last week they told the Boston Globe they projected the year end death toll would be three times the 15 deaths counted two years ago. Obviously both statements cannot be true.

To gain a better understanding of Jane Doe Inc's data, RADAR examined their document at http://www.janedoe.org/involved/Narratives%202007.pdf which gives homicide data through September 21, 2007. That document opens with a statement that Jane Doe Inc. had changed their counting methodology in 2006, which calls into question their reasons for choosing 2005 rather than 2006 as the year to compare 2007 to.

After stating that they'd changed their methodology, the document lists four categories they counted as DV-related homicides: 1) ex's, 2) bystanders, 3) cases in which the motive was jealousy, and 4) cases in which there was a power and control relationship between the victim and the perpetrator. Are these the new categories? They certainly can't be all the categories to be counted because none of them include currently married partners. In the very next paragraph, they list an two additional categories: 1) perpetrators who committed suicide and 2) homicides in other states of people who had once lived in Massachusetts. Are these the new categories? In spite of our best efforts to make sense of Jane Doe Inc's claims, their confusing presentation made it extremely difficult.

Here's what we do know at this point:

• Jane Doe Inc. told the Boston Herald they were projecting 57 DV-related homicides by year end, yet they told the Boston Globe they were expecting the number to be three times the 15 homicides counted in 2005. Both can't be right. 57 is the number they'd have calculated if they had included perpetrator suicides and out-of-state homicides in their projection.
• Jane Doe Inc. compared this year's data to the data from two years ago rather than comparing it to last year's data. In media interviews they neglected to make it clear that the data from two years ago was collected using a different counting methodology. Furthermore, the death toll in 2005 was notably lower than either 2006 or 2004, thus inflating the rate of increase they could quote to the media.

In spite of our best efforts, RADAR was misled by Jane Doe Inc's confusing presentation, and we incorrectly stated "there has not been a threefold increase".

Date of RADAR Release: December 16, 2007

Want to improve the chance that they'll pay attention to your letter? Click here.

R.A.D.A.R. – Respecting Accuracy in Domestic Abuse Reporting – is a non-profit, non-partisan organization of men and women working to improve the effectiveness of our nation's approach to solving domestic violence. http://mediaradar.org