Evidence-based Research to Combat Global Poverty
June 10, 2014 by Jennifer Cobb
Innovations for Poverty Action (IPA) designs and runs economic development research projects and evaluations in more then 15 countries around the world. 500 IPA staff, based in-country, implement evidence-based projects designed by academics and NGOS to illuminate policy and philanthropic responses to combating global poverty, much of which is gathered in the form of paper. The organization works on issue areas ranging from global financial inclusion to post conflict recovery, often in very challenging settings with marginal infrastructure and poor resource availability.
As Niall Keleher, Director of Research Methods and Knowledge Management for IPA, commented, “We partner with other implementing organizations—mostly local and international nonprofits, but also governments and private firms—and design and run program evaluations to find out what works and what doesn't, and then we let the world know.”
The vast majority of IPA’s staff, working in countries of disproportionately poor populations, manage data collection or some other component of a randomized trial study. Most of these studies involve personal interviews with upwards of 1,000 participants. A recent study in Ghana interviewed 40,000 students. As Keleher said, “We are the boots on the ground that oversees the implementation of the research.”
Niall’s team based in New Haven works with the in-country staff to help them use the best tools and processes to effectively do their jobs. Historically, IPA has been very paper intensive, relying on printing out thousands or millions of pages of surveys every year. While the organization has been seeking to transition some processes to direct data collection using computers, this is not practical in many countries, such as Ghana, Malawi and Kenya. “While about 60% of our processes are moving to a more digital environment, we are far from paperless,” Keleher commented.
“I was country director in Malawi for three years and we ran a lot of paper surveys,” he continued. “It could take three to six months before we had a fully reconciled, double entry data set for the researchers. That was very taxing for our research team.”
In those areas where paper remains the most efficient approach, Captricity offers a highly effective tool for getting paper data into the digital form they need, quickly and accurately. For example, for most of the surveys the organization runs, there are thousands of respondents participating in 30 minute to 4 hour interviews. These interviews are very thorough and it is imperative that IPA check on the quality and reliability of this data. To do this, the organization sends auditors back out to do a second, short interview with between 10% and 20% of the respondents. These audits check on measurement errors and validate the location and participant information and other variables. These audits are always done with paper forms, which are then processed using Captricity.
Using the Captricity QA tool, IPA validates the results of the audits against the larger data set very quickly, enabling staff to spend more time reviewing and analyzing results and less time doing data entry.
Another area where Captricity adds significant value is in digitizing administrative information from partner organizations, such as attendance sheets from a partner in Liberia that is part of a larger research project. Administrative information often forms an important backdrop to survey-based research projects.
IPA works with a significant amount of Personally Identifiable Information (PII). The way Captricity is set up, IPA is able to ensure that only the teams working on a given project have access to the data sets in question. This is critical given the sensitive nature of the work and the IRB code of ethics that governs all of their research.
Keleher concludes, “Captricity has helped transform the way Innovations for Poverty Action approaches data entry. With quick and accurate delivery, Captricity allows us to obtain reliable data for analysis and decision-making.”