Legal implications persist around open data for development


Written by

The U.S. Agency for International Development has served as a positive force for advancing the Obama administration’s open data initiatives, but some staff in the agency are still debating the legal implications of opening agency data and wonder how that work actually advances their mission.

Panelists Wednesday at an event hosted by the Center for Data Innovation discussed a wide range of issues related to data use, including the need to consider legal implications of opening data and what their priorities should be going forward to enable more data use.

With limited resources available, agencies have to make hard decisions about where to focus their efforts to help lawmakers make data-driven development decisions.

“I think that there is really good progress and there’s potential in open data, but there’s, even on our own team within the Global Development Lab, a raging debate over: What is our stance on open data?” said Vivian Ranson, senior program manager of the Development Informatics Team in the Global Development Lab at USAID.

Ranson said the legal implications of opening data should be questioned: “What are our responsibilities to protect our partner governments and citizens? And who owns the data that we collect and build in platforms?”

“It’s not that we don’t believe in it. But open data on its own does not create better development outcomes,” Ranson said. “So… Should we be focusing on building the APIs, and opening up the data, making it available? Or should we be focusing on the capacity of governments to actually use that data to drive better service to citizens?”

Going forward, USAID is trying to learn, she said, “What does open data do for us?”

Enabling people to use the collected data to make decisions, more broadly, is a focus for the agency, Ranson said.

“We’ve reached this point where we have technology, we’re pulling in data. We’re not using it,” Ranson said.

So her team is focusing on adjusting programming so that it can incorporate the data to make decisions.

Panelist Siobhan Green, CEO of Sonjara, Inc., said the focus needs to be on making data that’s already collected more accessible and interoperable.

She gave an example from her recent work in Uganda with the Ministry of Health. A data set already existed of all the country’s districts with GPS coordinates, she said. It was ready to go and could have been released as an API  — but it was in a Microsoft Excel spreadsheet and trapped on a hard drive.

“There’s a lack of capacity around using the data that’s already being collected and making it more publicly available,” Green said.

USAID has made good first steps, she said, including creating their own open data policy and the Development Data Library, a central data repository.

Building communities of best practice around the data and promoting information sharing can lead to better data use, she said. For projects, the technology-driven people need to involve a community of practice in the conversation.

“USAID, the open data team, has actually started this, but as we’ve known from technology you can’t just throw up something on a website and say ‘Go!’” she said.

But she noted that the mSTAR project, funded by USAID, is a good example of an investment that is building communities to share information and best practices.

The project, short for Mobile Solutions Technical Assistance and Research or mSTAR, works to provide technical assistance and research in developing countries in three areas: digital finance, digital inclusion, and mobile data collection for learning and decision-making.

When thinking about collaboration to advance goals on data, including implementing best practices, Green also mentioned a related issue: data reuse.

In the development field, data is often duplicated due to many issues, from contracts to a simple lack of information sharing.

“That data reuse honestly needs to be, I believe, something that’s required,” Green said.

She added that data reuse would also improve data quality.

“If you know your data is going to be looked at and reused, the chances that it’s going to be higher quality are much much greater,” Green said. “And you’re going to think about how you’re actually capturing it.”

-In this Story-