Research

Improving qualitative data management by agencies responding to refugee crises

Carleen Maitland, associate professor of information sciences and technology, is exploring how qualitative data could be managed more effectively by organizations responding to the Venezuelan refugee crisis in Ecuador. Credit: Adobe Stock: Devon YuAll Rights Reserved.

UNIVERSITY PARK, Pa. — As of September 2019 there were approximately 330,000 Venezuelans that had fled to neighboring Ecuador amid an economic and political crisis in their home country, according to the Regional Interagency Coordination Platform for Refugees and Migrants from Venezuela. The platform is led by the International Organization for Migration and the Office of the UN High Commissioner for Refugees (UNHCR). As humanitarian organizations respond to aid the displaced individuals and families, could helpful information in the Venezuelan refugee crisis in Ecuador be used more effectively, overcoming software, analytic or organizational limitations?

That is a question that Carleen Maitland, associate professor of information sciences and technology, is exploring as part of a new project — QualMiner — funded by the United Nations High Commissioner for Refugees (UNHCR) Innovation Program.

Maitland is working with UNHCR Information Management staff, together with staff of international non-government organization HIAS, in Quito, Ecuador, as well as Netherlands-based software firm BeDataDriven, whose platform, Activity Info, is used by nonprofit organizations to upload data about their programs. Ecuadorian relief agencies use the platform to input information about the services they provide as they help Venezuelan refugees access housing, healthcare and education.

According to Maitland, humanitarian systems in general, as well as the platform in particular, excel at managing quantitative data — such as how many refugees have been helped and the timeline to find housing — but present challenges to using, that is inputting, managing and analyzing, qualitative data.

“We find that in managing qualitative data there is a little bit more concern about what we in information sciences call ontologies, or the meanings associated with data,” said Maitland.

She explained that in quantitative terms, the platform can easily specify various dimensions of aid, such as what housing options have been secured for refugees. For example, quantitative data can be entered that housing has been found that has at least one bedroom per three people, and is available for longer than two weeks.

But how can qualitative data that involves a narrative element, such as a description of the neighborhood where the housing is located, best be captured? Currently, use of open-ended comment boxes in the software raise questions about the processes for collecting and analyzing these data. For instance, different staff members across collaborating organizations may use different terms to refer to the same concepts, generating discrepancies.

“So we’ve found organizations may handle data entry differently. Some use more centralized processes, whereby people from a field office send data to the capital via email, and someone there reads it all then standardizes and inputs it,” said Maitland. In others, a more decentralized approach is taken. “The question is, is there a best way? What do decision makers in the humanitarian response want? Can coordination bodies handle both approaches, letting individual organizations handle data management in a way that fits their established organizational processes?”

Maitland and College of IST master’s student Abdelhady Abuolba are researching processes for mining and managing qualitative data gathered in the operations for the Venezuelan refugee response. They are working together with the UNHCR Information Management team in Quito, Ecuador to test visual outputs of data — such as vignettes and word clouds — to try to understand which are most helpful for decision making, and if and how they differ from quantitative output.

“Through this project, our hope is that we will be able to make better use of qualitative data in humanitarian response and to improve response overall,” concluded Maitland. “We feel that just relying on quantitative data really misses a lot of the nuance of the benefits and challenges of humanitarian operations. Improved operations will help refugees overcome the trauma and loss experienced in Venezuela.”

The work is funded by the UNHCR Innovation Fund.

Last Updated June 28, 2021