Citizen-Generated Data and Governments: Towards a Collaborative Model

"...global discussions about evidence, monitoring and the 'data revolution' - particularly in the context of the Sustainable Development Goals (SDGs) - continue to gain speed..."

Citizen-generated data (CGD) is data that people or their organisations produce to directly monitor, demand, or drive change on issues that affect them, including issues outlined in the SDGs. Datashift is supporting civil society organisations (CSOs) that produce and use CGD in initial pilot locations: Argentina, Nepal, Kenya, and Tanzania. This Datashift piece explores the idea that governments could host and publish CGD themselves, and whether this could mean that data is applied more widely and in a more sustainable way. It was inspired by a 2015 meeting in Buenos Aires with Argentine CSOs and government representatives, where participants discussed the potential for government open data portals, such as that managed by the Buenos Aires Innovation Lab, to host and publish CGD.

Following an introduction, the paper explores the potential benefits and obstacles both government representatives and CSOs may face when collaborating on government hosting of CGD. In terms of advantages, incorporating CGD into government-managed data portals can "significantly broaden the scope and coverage of those portals, adding depth and context in sectors where government data exists, and filling gaps in sectors where it doesn't." It also "gives the impression of collaboration with civil society, which can also be important at home for deflecting criticism that governments are cherry-picking or manipulating official data to support policy aims." Yet, for governments, the most significant substantive obstacle to hosting CGD may be the quality of CGD sets. A potential workaround would be to add caveats and contextual data to explain what the data does and does not represent, or by standardizing or restructuring data sets. For CSOs, "including CGD on government data portals can improve the profile, accessibility, use and quality of that data." However, "ensuring methodological rigour can prove a significant obstacle to hosting data on government portals....Concerns about losing control or ownership over data or about how data will be presented and used can also obstruct progress."

When feasible, however, it is noted that government hosting of CGD could "dramatically increase countries' capacity to use data for monitoring and improving development initiatives." For example, while data on national access to primary education may not be disaggregated by gender or ethnicity, CGD (though not representative and with limited coverage) may cover these issues, which "can feed directly into strategies for data collection or policy action." To cite another component of this approach, collaboration on data hosting can help strengthen relations across civil society and government. A case study offers an illustration of how this can work in practice. In the United Kingdom (UK), citizen scientists contribute to the data gathered and used by the Met Office, which is the UK's national weather service. The project uses the power of the crowd to transcribe weather observations written in historical shipping logs recovered from archives around the world. Amateur weather observers also contribute regularly to weather data used by the Met Office and by climate scientists, using sensors to measure variables such as rainfall and air pressure. The Climatological Observers Link (COL) provides guidance on data formats and standards to follow, and observers are encouraged to follow these protocols so that the data they upload can be integrated with other data sets. The Met Office also has its own site where it gathers observations from weather observers across the UK, called the Weather Observations Website (WOW). The data gathered by amateur observers is, once uploaded to the site, kept indefinitely, which provides sustainability to the data created through this initiative. For the Met Office, being able to easily compare their "official data" with data sets coming in from serious amateur observers allows them to verify that what they are receiving and reading is accurate, and gives them alternative measurements for forecasts (perhaps from areas where less coverage is available).

The third section suggests components for a model that civil society and government representatives could adopt to support the successful implementation of such an initiative. "Generally, a holistic and collaborative approach to hosting, in which investments of time and money are shared by government and civil society, is likely to best meet the needs and incentives of all parties. This will inevitably begin with conversations between CGD producers and the managers of online data portals, in which each party's incentives and opportunities will dictate the scope for collaboration." A first step is mapping what data is available both in and outside of that portal. Workshops are a can help establish the criteria for data set inclusion in government portals. "This can be done on the basis of, or in parallel with a mapping exercise, and should be structured to ensure that civil society perspectives on data standards are well accounted for." Another suggestion: "Where feasible and resources can be secured (also through third parties), options should be explored for establishing fellowships or secondments for staff from civil society organizations producing data, to work within open data portal institutions, specifically on the process of incorporating CGD sets."

The closing section notes the dramatic lack of international experience with such solutions, and suggests some additional steps that could be taken to further collective understanding of government and civil society data collaboration in general, and hosting schemes in particular.