Is Domestic Violence Declining in America?
Why Federally-reported DV data may only represent the tip of the iceberg
By: Andrew M. Campbell
Statistical analyses and published reports from the U.S. Department of Justice have long reported a decline in domestic violence victimization in the United States...Yes, you read that right....a DECLINE (https://www.bjs.gov/content/pub/pdf/ndv0312.pdf). When I make this statement at the many conferences and trainings for law enforcement officers, medical professionals, attorneys, and social service providers that I speak at across the country, I receive the same response – confused looks and shaking heads. Why would the federal government continue to report a decline in domestic violence victimization when those on the front lines continually tell me they feel the opposite to be true?
Federally reported numbers, and those reported by many other agencies and entities across the country, are often dependent on retrospective victim self-reports obtained through over-the-phone or in-person interviews. Though cost and ease of execution often make these methods the go-to resource for researchers, they often fail to capture data from large portions of the population. Studies have shown that these types of surveys often miss families on military bases; individuals who do not speak English; individuals who are hospitalized or institutionalized; families with “chaotic lives”; and individuals with low incomes and/or limited education – all characteristics that have been well documented in academic literature as risk factors for domestic violence victimization. Additionally, the survey often used to compute federal statistics – the National Crime Victimization Survey, and others like it, such as the National Violence against Women Survey, often report little participation from African American women (10%), female adults under the age of 30 years (19%), and households with one or more children (30%).
My recently published studies in the Journal of Interpersonal Violence sought to determine if utilizing a different data source – information collected by first responders at the scene of domestic violence incidents – might shed new light on the true frequency and severity of domestic violence victimization in America (http://journals.sagepub.com/doi/abs/10.1177/0886260517704230). After an analysis of nearly 10,000 domestic violence incident police reports in Marion County, Indiana (largest county in the state of Indiana; population of 903,393) we found a disproportionate representation of African American victims (51%; county population = 26%), female adult victims under the age of 30 years (44%; county population = 20%), and households with one or more children (59%) – all groups with minimal representation in federally reported statistics. In addition, we found that over a three-year period, domestic violence incidents occurred in Marion County, Indiana at a rate that was nearly double the national average according to federally reported statistics (441 victims per 100,000 population aged 12 years or older).
Further research is needed to clarify the generalizability of these findings (our study included multiple jurisdictions but was confined to a single geographic area) but at the very least, they warrant real concern. Are federal reports of declining domestic violence victimization flawed in that their data is not inclusive of populations who are in fact disproportionately represented as victims? Victim-serving agencies are often very dependent on data to secure adequate funding for services – if estimates of violence victimization prevalence across the United States are inaccurate, these critical agencies may continue to experience cuts in funding and be forced to turn “at-risk” victims away (https://www.indystar.com/story/news/2014/08/31/domestic-violence-shelters-find-greater-need-help/14916077/). These victims may then return to their home environment and be at great risk of suffering additional harm or even death – particularly if the abuser knows they sought outside assistance or attempted to flee. We owe it to these agencies and the vulnerable victims that they serve, to re-evaluate federal reporting measures to ensure accurate and truly representative data collection and reporting of this nationwide epidemic.
Furthermore, even though my study found domestic violence to be occurring at nearly double the Federally-reported rate, this may still only represent the tip of the iceberg. Prior studies have estimated that only 1 in 4 domestic violence victims ever contact law enforcement – meaning the actual occurrence rate could be 4 times the rate cited in my study and 8 times the nationally reported rate (1,764 victims per 100,000 population aged 12 years and older). Additionally, victims involved in Marion County incidents reported, on average, 10 prior incidents of domestic violence (many involving strangulation, weapon-use, or death threats) with the suspect before they contacted law enforcement. All men, women, children, and pets residing in these homes should be considered at great risk of suffering severe emotional and/or physical harm or death. We can, and must, do better as a nation to accurately count and identify victims of domestic violence (men, women, children, and animals), ensure appropriate funding for victim-serving agencies, and guarantee the continued availability of essential community resources and services for ALL victims of domestic violence.
Your comment will be posted after it is approved.
Leave a Reply.
Andrew Campbell is an expert on domestic violence and the associated risks of harm for adults, children, and animals residing in homes where this violence occurs. Utilizing cutting-edge data analysis and a unique perspective on violence in the home, he provides education for agencies across the country and assists in developing more efficient and effective community responses to physical and emotional violence. Andrew’s several recent publications in major academic journals include his analysis of nearly 10,000 first responder reports and observations from the scene of domestic violence incidents and are the first in academic literature to use law enforcement data to quantify risk of harm/injury in this manner. Contact Andrew at: email@example.com