Why publishing more open data isn’t enough to empower citizens


Why publishing more open data isn’t enough to empower citizens

Adi Eyal | January 11, 2017

When I attended the International Open Data Conference last October in Madrid, there was plenty of talk about whether open data has failed to live up to its promise. Has open data really increased transparency, improved government efficiency, brought about world peace, ended world hunger? What are we really talking about when we talk about open data’s “impact?” Whatever impact there might be is restricted to a couple of interesting case studies, but there is not yet a larger body of work describing how open data has brought about systemic, long-term change to societies around the world.

At the conference, I heard the same tired arguments about the need for data to be “open by default.” I heard countless examples of subversive hackers liberating data locked up in PDFs, in order to “uncover corruption.” Never mind that the mere act of making certain datasets public has rarely resulted in tangible policy changes. It seemed to me that the open data purists’ mantra is that we can’t predict how data will be used, so the release of data is important in and of itself, without concern over its value for society.  

But simply “liberating” data is not enough. Even last year’s UN high-level conference on Africa’s data revolution recognized that private citizens are unlikely to use open data, and hence intermediaries — or “infomediaries” — must play an important role. These groups (data wranglers, academics, data-proficient civil society organizations, etc.) turn data into actionable information, which can then be used to lobby for tangible change.

Increasing the impact of the open data movement isn’t just a matter of emphasizing the role of these “infomediaries” — it means shifting focus from supply to demand. As many have argued, increasing the supply of data sets without focusing on what data is actually needed to solve specific problems is unlikely to lead to satisfying impacts.

I won’t rehash the same points made by others who’ve explained the importance of releasing the datasets that are most in demand. I’m interested in what I think is the next frontier in the open data movement — data literacy.

Of course, this is nothing new under the sun. The School of Data has been a leader in this area, aiming to teach journalists and others the skills they need to use data effectively. There are many other data journalism initiatives around the world doing the same. What’s lacking is a better definition of what “data literacy” actually means. A “data literate” citizen isn’t someone who knows how to handle a spreadsheet — it’s someone who inherently understands the value of data in decision making.

As open data advocates, we cannot ignore the importance of this. Assuming that more citizens aren’t advocating for better government due to lack of data is naive. While Louis Brandeis sounded clever when he wrote, “sunlight is said to be the best of disinfectants,” this doesn’t reflect reality. Information is power… but only sometimes. If citizens aren’t empowered to take action based on information, releasing one dataset after another can quickly become demoralizing. We need to do more work in this area. Ultimately, I think Aaron Schwartz was more accurate when he said, “Sunlight is not in fact the best disinfectant; actual disinfectant is. Sunlight just makes it easier for people to look at the pus.”

So how do we increase data literacy and push the open data movement towards producing more tangible impacts? I wouldn’t recommend hackathons, where the focus tends to be on technologists to the exclusion of others. Data literacy should extend to ordinary citizens, even the innumerate.

One way forward is the example set by Black Sash, a South African social justice organization that helps communities collect data from their neighbors about local government services. Black Sash helps them collate that data and produce simple infographics, which can be used to start a conversation between communities and service providers. An individual complaining about waiting in line for an entire day in order to receive her pension can easily be dismissed. But 400 pensioners complaining about waiting six hours on average is a much stronger argument. In their work, data doesn’t replace dialogue but merely facilitates it. While many in these communities may be innumerate, simple graphics are a powerful mechanism to support discussions.

Data literacy means more than simply understanding graphs. It is an appreciation for how data can be used to strengthen advocacy. When Code for South Africa developed a map of Cape Town’s informal settlements with activist organization Ndifuna Ukwazi, the goal was to start a conversation with the city government over why so many neighborhoods lack sanitation systems.

This is just one example of how “infomediaries” leveraged data in order to address a specific, solvable problem that local communities said was important to them. If the open data movement is to move forward, it would do well to adopt “more data literacy, not more open data sets” as a guiding mantra.

Adi Eyal is the founder of Code for South Africa. Learn more about his work as an ICFJ Knight Fellow here. 

Main image CC-licensed by Flickr via CyberHades.



Parallel efforts required...

Great article! - and I agree with the importance of data literacy to realising impact from open data. I think though that we can't let go completely of the drive to release more datasets otherwise we'll end up with a lot of frustrated data literate people without the data they need to solve the problems they are now capable of solving. Both streams of work are important and need to be persued in parallel for maximum impact


Plain text

  • No HTML tags allowed.
  • Twitter message links are opened in new windows and rel="nofollow" is added.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Please log in or register in order to comment this post.