Harvard Collects Data in the World's Most Challenging Environments

The Women Environmental Programme assessing the state of infrastructure development in Nigeria. The Fishery Commission monitoring illegal fishing practices in Ghana. International organizations and agencies assessing humanitarian needs of Syrian refugees and across virtually all disasters worldwide. An aid agency monitoring food security in Tanzania. Researchers collecting data on control over natural resources and, separately,  on perception of transitional justice in South Sudan.

At first glance, these organizations and their work have little in common. But along with thousands of non-governmental organizations and agencies in Africa, they share a common platform and a connection to Harvard.

The platform is KoBoToolbox – a free and open source digital platform for data collection which was created ten years ago and now continues to be developed at the Harvard Humanitarian Initiative(HHI). KoBoToolbox is a simple way to create questionnaires and other forms and deploy them for data collection through in-person surveys on a tablet or on the web for online surveys. As data is collected, a dashboard offers a quick and simple way to visualize results and map the data being collected.

What makes KoBoToolbox unique is that it was developed by users and for users, with challenging contexts like war zones and humanitarian disasters in mind: it needed to be robust and work off-line. For example, HHI had to ensure that they could collect data in remote areas of the World with no electricity or internet, and during emergencies. Features are developed in close consultation with leading humanitarian agencies, including the United Nations Office for the Coordination of Humanitarian Affairs, the United Nations High Commission for Refugees, and humanitarian NGOs like Médecins Sans Frontières (MSF)/Doctors Without Borders, Save the Children and the International Rescue Committee. The result is that KoBoToolbox is now used across every continent to gather critical data to inform humanitarian action. Every month, HHI receives over one million assessments and surveys, which can cover everything from needs assessments to evaluation surveys.

KoBoToolbox also caters to the needs of researchers like the ones at the Harvard Humanitarian Initiative – that is why the platform was initially created. HHI’s goal was to improve how quickly they were able to move from data collection in the field, to obtaining a usable dataset for analysis.  The defining moment was a piece of research conducted in Iraq in 2003. It took close to two years to publish because of data entry, translation, analysis and other factors. It was valuable academic research, but it came too late to be directly relevant for those working to rebuild the country.  From that point forward, HHI set to develop a user-friendly way to ‘digitze’ information directly at the source -  for example during interviews or focus groups.

The first prototypes were built using money that would normally be budgeted for data entry and photocopies of questionnaires. Over time HHI adopted emerging standards and tools, like the Open Data Kit, and built an ecosystem of tools and services to improve data collection.  In the process, they were also able to improve the quality of the data collected, including building constraints and validation checks that reduce errors.

This work is far from finished – HHI is constantly improving KoBoToolbox. A new major area of development is the creation of templates and libraries of questions and scales. HHI has also launched an online course and is developing more resources to help build capacities of researchers around the World, encourage standardization, and facilitate the process of data sharing. 

Today, HHI continues to use KoBotoolbox to support our research on the process of rebuilding societies affected by violence and the evaluation of humanitarian action among other objectives. HHI travels to remote areas of conflict-affected countries to understand the effects of violence and what people see as the way forward to build more stable and peaceful societies.

“Kobo” means “transfer” in Acholi, a language spoken in northern Uganda, where HHI first tested and implemented a survey using the platform. The name was chosen because the goal was to ease the transfer of knowledge from communities to researchers and ultimately decision-makers, hopefully helping to create a dialogue on how to best take local perspectives into account in research. KoBoToolbox originally was a true toolbox with tablets, batteries, shovels, rain gears –all the equipment needed to navigate data collection in remote areas.

Building KoBoToolbox has opened new areas of research on diffusion of innovation and the role of technologies in the humanitarian and peacebuilding fields. In a world of expanding capabilities fueled by innovation, we must understand how to leverage technologies to reduce violence and increase resilience of communities at risk, and do so in manner that is ethical and protects those most vulnerable. This is a major aspect of HHI’s current research agenda, which led them to create Data-Pop Alliance in collaboration with MIT. 

###

KoBoToolbox was founded in 2005 by Phuong Pham and Patrick Vinck. They joined the Harvard Humanitarian Initiative in 2011. KoBoToolbox is funded through the generous support of numerous partners, including the John D. and Catherine T. MacArthur Foundation, the United Nations Office for the Coordination of Humanitarian Affairs, the United Nations High Commission for Refugees, USAID’s Office of US Foreign Disaster Assistance, Doctors Without Borders, Save the Children, the World Resource Institute, the Hilton Foundation, Cisco foundation, Fidelity foundation and other organizations and philanthropists.

Patrick Vinck is Research Director of the Harvard Humanitarian Initiative. He is assistant professor at the Harvard Medical School and Harvard T.H. Chan School of Public Health, and lead investigator at the Brigham and Women's Hospital.

Phuong Pham is the Director of the Implementation and Evaluation Science Program at the Harvard Humanitarian Initiative. She is an assistant professor at the Harvard Medical School and Harvard T.H. Chan School of Public Health, and lead investigator at the Brigham and Women's Hospital.