By Tom Walker
Our last blog post explained what we produced and shared in the Learning Zone over the last year. This post will focus on what we found.
Benefits to using a broad term to connect knowledge across sectors
Why use the ‘umbrella term’ of citizen-generated data? It’s useful in that it helps us to recognise and learn from similar initiatives across different sectors – but we shouldn’t get too caught up on the labels. Everywhere, information provided directly by citizens is being shared with more ‘official’ institutions. All of these initiatives can learn a lot from each other.
Citizen-generated data is not new, but technology offers new opportunities
Civil society organisations have long collected information from citizens as part of their advocacy, as with the Concerned Citizens of Abra for Good Governance’s work over the last 20 years (see this case study). Still, digital technologies do give civil society organisations an opportunity to collect information from many more people, analyse it in new ways (as with this community land-mapping project in Indonesia), and reach more people with what they’ve found. DataShift aims to identify ways that civil society organisations can do this more often, and more effectively.
Citizen-generated data initiatives are making a real impact
Overall, we’ve learned that citizen-generated data initiatives are making a real impact in a huge range of areas. To take just a few from the studies by DataShift’s in-country research teams:
- a public authority in Argentina now includes and responds to data citizens who submit information about river clean-up operations;
- Nepal’s cabinet used citizen-generated data to identify community needs following the April 2015 earthquake; and
- data collected by the Kenyan organisation CARD was used to get more teachers allocated to schools in Turkana.
Citizen-generated data is not just an exciting prospect: it’s already getting results.
Recognising its value and limitations
For citizen-generated data to be used to its full potential, we need to understand what it can (and can’t) do. As outlined in our Changing what Counts report, and our white paper on government-civil society collaboration, citizen-generated data is not a replacement for quantitative statistics. Rather, it complements existing data collection methods, often providing essential qualitative data about citizens’ opinions and perspectives on what’s needed.
Citizen engagement is key
Our in-country research teams found many cases where official institutions were reluctant to use data from citizen-generated data initiatives. Sometimes this was because the authorities did not see how or were unable to use data effectively. In other cases it was actively rejected after the data highlighted problems that official institutions didn’t want to acknowledge. For example, CARD was asked to stop monitoring teacher absenteeism after it highlighted inefficiencies and corruption in Kenya. This relationship doesn’t have to be adversarial, as Buenos Aires’ open data portal’s hosting of citizen-generated data shows. However, as the Argentine research team concluded, using data in this way may be just one element of a successful advocacy strategy. Citizen-generated data initiatives may help to initiate dialogue, but to get results, civil society organisations will probably need to accompany them with broader efforts to mobilise people.
Important to provide capacity and support for citizen-generated data initiatives
The research has also highlighted practical challenges. Few of the initiatives had systematic ways of checking the information that they collected. Those that did try to validate data mainly relied on time-consuming manual verification procedures like telephoning people who had submitted reports and visiting projects on the ground. We still need to know more about these challenges, as well as finding and sharing practical ways of dealing with them. The amount of data collected by different initiatives also varied dramatically – from small-scale projects focused on land rights in a particular market in Uganda, to combining data from existing government complaints systems with Twitter data. Capturing all of this diversity in one unifying framework may prove impossible.
As the East African research team point out, many of citizen-generated data initiatives are small and localised, and the data they collect can’t be generalised across the whole country or compared with other countries. In almost all cases, initiatives didn’t explicitly connect their work to larger narratives like the Sustainable Development Goals (SDGs) – even though their work was often directly connected to them. On this evidence, there’s still a lot to be done before this data can be used to monitor the SDGs in a systematic, cross-country way.
Finally, it’s important to note that most of the initiatives that we’ve come across depended upon consistent, long-term grant funding. In a number of cases, initiatives that were producing useful, high-quality data simply stopped when their funding ran out.
All the research conducted through DataShift suggests that citizen-generated data initiatives need support so that they can continue their work, develop their methodologies and learn from other, similar organisations. In many cases, these initiatives were generating data that couldn’t be found anywhere else. Whether it was information about land-grabs, police violence or pollution data, these initiatives are making real, valuable contributions in a range of areas.
Citizen-generated data is more than just a ‘type’ of data. Our in-country research teams often found that citizens felt empowered simply by participating in citizen-generated initiatives. Citizen-generated data initiatives allowed communities to articulate their needs from their own perspective and portray themselves in a more positive light, countering their feelings that their communities were misunderstood. They’re a tool for citizen engagement; for proving that institutions are listening to the public beyond just soundbites or promises; and potentially for collaboration too. There’s still more to do, however, and DataShift is ready to help support this work now and farther into the future.