Building the Evidence Base for Open Data

“Open data” has become the new paradigm for developed and developing countries alike.  Countries including the UK, Kenya, Ghana, and the United States among many others have launched open data portals (to varying degrees of success) to increase the disclosure of federal spending data. Others are taking steps to develop and implement policy reforms that will make their government spending more transparent; this week President Obama signed what is becoming known as the first US “open data” law.

In whole, these steps to make governments more open and transparent are good. But we must better understand the strengths and limitations of data to fully realize the potential of open data. Look no further than recent backlash against another popular data trend, “big data,” to underscore this point.

As a start, is there evidence to demonstrate that open data works?  And what does it even mean for open data to be “working”? 

Proponents of open data say it improves governance, increases civic engagement, and encourages research and innovation, among other positive attributes. There is a basis for some of this in the literature; for example, increasing access to information has been found to have positive impacts on reducing corruption—including in the education sector in Uganda and the health sector in Argentina.

But there isn’t much evidence for open data’s contribution to other goals in low- and middle-income countries as of yet. The Open Data Research network and World Wide Web Foundation are making a good effort to improve this, collecting stories on the emerging impacts of open data in developing countries. At this point the repository seems to pose more questions than contain answers. Case studies will examine the effects of open data on topics including poverty reduction, human rights, improved energy resource development, and even university governance—but most of this work is still in progress. 

We explored the potential and pitfalls of open data in Africa in our Data for African Development Working Group, in partnership with the African Population & Health Research Center, and find that open data has been slower to catch on in Africa than other regions. Many national statistics offices and other government departments are either hesitant to publish their data, don’t have the capacity to publish and manage data according to international best practices, or do not understand what data users want to know and how to get information to them. 

Still, our report recommendations encourage the use of open data, suggesting that national governments release all non-confidential, publishable data, including metadata, free of charge and online in a machine-readable format. We also call for donors to increase coordination and cooperation -- particularly between UNECA, AUC, AfDB, AFRISTAT,  UIS, World Bank, and sub-regional organizations -- to prevent duplication of efforts and to better utilize limited country resources.

Efforts to improve open data can (and should) happen alongside efforts to learn about its usefulness.  We would argue that as a public good, open data should be published and supported as long as the costs of doing so are not prohibitive.   But it will be interesting to see if and how the research community will be able to quantify the effects of open data.  Whether open data will be a win-win remains to be seen. 


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.