Over the years the number of topics that require attention, expertise, and funding that the official statistics community must contend with has expanded tremendously.
We’ve seen a well-deserved focus on disaggregated data beyond sex and age through initiatives like the Inclusive Data Charter. We’re navigating the rise of new digital technologies and artificial intelligence—ever heard of ChatGPT? We’re responding to a rise in global challenges and drop in official development assistance for data. And we’re working to rebalance our approach, so data are not just about people but for people.
As we continue to grapple with these new issues and forge solutions, let us not forget the basics. Open data being one of them.
Open data provides us with a foundation to ensure that data are available for all to use and reuse. But open data goes beyond just the technical definitions that outline how data should be produced and published. It cuts across the whole data value chain—from ensuring methodologies are transparent and available to tracking the uptake and use of data. The principles of open data may look simple on paper but for many national statistical offices around the world it is difficult in practice but crucial to keep at it.
The 2022/23 Open Data Inventory produced by Open Data Watch shows that many countries have struggled to maintain the availability and openness of their data since the COVID-19 pandemic began. Country scores are falling on coverage for the first time in six years, while openness is barely increasing compared to previous years. The latest survey of the Cape Town Global Action Plan shares a similar message of countries struggling to produce open data since the start of the pandemic. However, when countries do have the resources and capacity to apply the principles of open data, progress is possible.
Open data alone won’t solve all of today’s challenges, but it can help. Take each of the new topics above.
Inclusive data systems rely on open data principles. As the Open Data Charter rightfully notes the more quality datasets you have access to, and the easier it is for them to talk to each other, the more potential value you can get from them. Commonly agreed data standards play a crucial role in making this happen. As we move towards inclusive and intersectional data systems, open data can help paint a more comprehensive picture by merging different disaggregations. Additionally, as our community tackles new territories of measurements—such as those on women’s economic empowerment, data use, or even the amount ODA for statistics—ensuring methodologies are open, not only for critique but for collaboration, is key.
Equitable AI systems rest on open data principles. Open data is one of the tools we can tap into to make machine learning fair, accountable, and transparent. More work is needed to track the use of open data in AI systems and to develop approaches that address potential biases engrained in algorithms through careful creation of inclusive open data infrastructures. Without open data, automated bias goes unchecked. As Jennifer Oldfield from GPSDD writes, “we also need global advocacy campaigns to pressure politicians and businesses to create cultures of transparency to share data and decision-making that underpin algorithms.”
Data equity needs open data. Data equity and data feminism are growing movements that calls for more responsible data practices from collection to access to data-driven decision-making. As the authors of Data Feminism say, “equity is both an outcome and a process.” It’s more than an end goal; it is a framework to guide data work from start to finish. In a data equitable world, people are at the heart of data collection and analysis. They shape how their data are collected and for what purpose. This people-centric approach is not possible without open, transparent, and inclusive data practices to build smarter and more transparent data ecosystem.
As our world changes so will our areas of focus. And as it does, I encourage us all to not forget our basics and apply what we’ve already learned works, like open data—from production to use and impact—to each of the new areas that may emerge. And AI experts and feminist data scholars need not become open data experts, but they need to include open data in their planning and advocacy endeavors. Partnerships between open data experts and other specific data communities are where progress and potential could thrive. Through partnership aimed at embedding open data principles into new emerging initiatives, we can create a multiplier effect together.
Look at the case of the merging of the open data and official statistics communities. Two quite different groups came together to discuss, debate, and forge paths forward. After years of collective work, the 54th session UNSC adopted open data as an official agenda item. And as demonstrated by country representatives from Poland, Chile, and Malaysia during our United Nations Statistics Commission side event, the result of official statistics and open data communities coming together has increased use for real impact.
The UN World Data Forum is an opportunity to connect, break down silos, and forge partnerships that only make us stronger together. Let’s not miss the chance to take it.
Looking for more updates on why open data matters?