The voices of Urban Institute's researchers and staff
December 9, 2014

How much do we know about the prevalence of sexual violence across ethnic and racial groups?

December 9, 2014

The data we have about sexual violence are lacking. Using data on sexual violence from one of the largest surveys on personal health behaviors—the Behavioral Risk Factor Surveillance System—we are only able to find estimates for 28 of 90 different minority groups. We also don’t know how rates of sexual violence vary across different racial and ethnic groups, which can be crucial information for service providers as they reach out to victims in diverse communities.

To highlight these shortcomings and suggest ways to improve data collection, the Urban Institute hosted a data dive, where civic-minded technologists, researchers, practitioners, and other experts combed through existing data on domestic violence. Several participants worked to visualize and analyze existing data. Others illustrated the lack of reliable and consistent data; the visualization we show below draws on those initial drafts.

What’s missing in the data we do have?

For the data visualization, we used data on sexual violence and intimate partner violence for adults come from the 2005 Behavior Risk Factor Surveillance System (BRFSS), conducted by the Centers for Disease Control and Prevention. Survey respondents from 18 states answered questions about sexual violence, which was defined as rape, attempted rape, being put in an unwanted sexual situation without touching, or being touched without consent or where the respondent didn’t want to be touched. Respondents from 10 states answered questions about intimate partner violence, which was defined as rape or physical violence inflicted by someone from a former or current relationship.


Because of the survey’s limited sample size and coverage area, we benchmarked the ethnic and racial composition of survey respondents in the BRFFS to the American Community Survey (ACS), a nationally representative survey conducted by the U.S. Census Bureau. In the 18 states for which we have data, the ethnic and racial distributions of sexual violence are generally close to each other (as seen by the proximity to the 45 degree line); this is less true for surveys about intimate partner violence.

Each circle in the graphic is scaled by the share of people in each state and the racial/ethnic groups who suffered from sexual violence or intimate partner violence. Groups for which we do not have reliable estimates (either because of small sample size or a high relative standard error) are represented with an “X”.

As an example, take a look at the blue circle near the middle of the graphic that shows estimates for Hawaii. Our analysis of the BRFSS shows that just above 2 percent of Asian Pacific Islanders in Hawaii were victims of sexual abuse in 2005, and nearly 15 percent suffered from intimate partner violence. But for both measures, the share of Asian Pacific Islanders surveyed was smaller than the share of Asian Pacific Islanders in Hawaii, according to the ACS. While the ACS finds that Asian Pacific Islanders make up 53 percent of the population in Hawaii, they make up only 42 percent of the BRFSS survey on sexual violence and only 22 percent of those surveyed about intimate partner violence.

Given the small sampling of states, we aren’t able to say much about the prevalence of sexual and intimate partner violence for racial and ethnic groups across the nation as a whole. More important, most of the estimates are statistically unreliable (and are thus shown as an “X”). We are unable to report more than half of the sexual violence estimates and more than two-thirds of the intimate partner violence estimates. And this doesn’t take into account the more than 30 states for which we have no information or have estimates that are more than a decade old.

The problem of sexual and intimate partner violence remains a tremendously important social issue across the United States. To better address this problem, policymakers, support groups, and, most important, survivors and their families need a steady source of useful information. There is still much we do not know about the incidence of sexual and intimate partner violence—and with such a small sample of states from which to calculate their rates—we are not able to produce national estimates. Furthermore, that small sample makes it difficult to look at each of the racial and ethnic minority groups in every state that participated. To produce better estimates in the future, data collection efforts need to be increased and improved, while remaining sensitive to the unique challenges in reaching out to diverse groups.

For more, read our feature on how better data can reduce domestic violence. Photo from Shutterstock.


As an organization, the Urban Institute does not take positions on issues. Experts are independent and empowered to share their evidence-based views and recommendations shaped by research.