How Little Data We Have About Domestic Violence

December 22, 2014 1:25 pm PDT | Open Data

“What gets measured gets done.” That’s the foundation of data-driven decision making. But what happens when information isn’t recorded? When there’s no opportunity to set benchmarks, let alone measure progress?

Incomplete or uncollected data is a pressing problem for researchers and activists working to combat domestic and intimate partner violence. The lack of information is even more prevalent in marginalized communities, such as LGBTQ, immigrants, and Asian-Pacific Islanders.

On November 1st, the Urban Institute and the Asian/Pacific Islander Domestic Violence Resource Project, held a Data Dive to better understand what data are available on domestic violence and how better to reach underreported populations.

The event brought together researchers, data visualization experts, and policy advocates and started with an introduction to the issue of domestic violence. Quickly the group got to the crux of the issue said Jonathan Schwabish, Senior Researcher and Data Visualization Expert at Urban Institute, in an interview with Socrata.

“We had a really good, productive day,” Schwabish says. “Although a lot of what was highlighted was the unfortunate lack of reliable, or even collected, data.”

The day produced several different data visualizations, ranging from a map of DC sexual assaults layered with domestic violence resources to pie charts of sexual violence by ethnicity, age, race, and other characteristics. Perhaps the most revealing of the data visualizations is an interactive scatter plot, which illuminates all the data that doesn’t match up and which states do not provide data on domestic violence.

“Looking at the data that is available reveals a real problem,” claims Schwabish. “We can see that the non-responses aren’t random. When we have surveys where non-responses are random, that’s fine. But, when we can see that there is one ethnic group, for instance, that doesn’t answer, we have to look at why.”

Ultimately, what came out of the Data Dive is the groundwork for a new branch of research about how to construct culturally and language sensitive surveys on domestic violence.

“We started a dialogue between the data folks and the policy people. At the beginning of the Data Dive, they didn’t understand what the other was talking about. But we organically clarified that lack of understanding, so policy specialists know what coders like to see and vice versa. From there, we can start to create better standards for collecting domestic violence data,” Schwabish reports.

Since the event, Urban Institute has called for new guidelines to improve domestic violence data collection, especially among marginalized communities. More research is required, Urban says, to determine how best to collect data, overcome concordance barriers, and ensure cultural and language sensitivity.

This Data Dive is a first step toward creating a real-world, broad-based understanding of who experiences domestic violence and where. Without this critical data, it will not be possible to truly prevent domestic violence and protect its victims.


Previous Article
Data Apps & Visualization, Data into Dollars, Open Data
Buildingeye: Mapping A Clearer City Plan

December 24, 2014

Next Article
Open Data
Civic Awesome: Open Data In The News For December 19, 2014

December 19, 2014