The emergence of open data, that is data sets that public and private entities make publicly available, has created a groundswell of excitement within governance, academia, and multi-lateral sectors because of its potential to make public services more efficient, hold elected officials accountable, generate economic value, and shorten the distance between governments and citizens. The United Nations has been opening up statistical and programmatic data for years now, and as experts learn more about the field of open data, it is important to reflect on the opportunities to maximize open data adoption within the UN.
On April 2, the Governance Lab (the GovLab) and the United States Mission to the United Nations (USUN) convened a meeting to reflect upon a wide array of issues pertaining to the value of open data – with a particular focus on the United Nations. The meeting brought together technology experts, academics, United Nations officials, and officials from the U.S. Mission to the UN (see below for a full list of participants) to discuss how the UN currently uses open data; reflect on open data initiatives both inside and outside the UN to draw comparative analyses; discuss the value proposition for using open data within the UN; and determine the gaps and additional areas of consideration.
Some highlights from the meeting:
Open Data within the UN: In the past decade, there has been a push to open up data at the United Nations, primarily to the end of increasing transparency and accountability. The International Aid Transparency Initiative is one example of the voluntary, multi-stakeholder organizations that seek to increase transparency and accountability within international development. Many UN agencies, among them the World Food Programme, UNICEF, OCHA, and the UNDP, have signed up to the International Aid Transparency Initiative standard.
Other open data initiatives also exist within the UN, including the centralized, all-UN open data portal, un.data.org, as well as agency-specific data portals where agencies release both statistical and programmatic data. The Statistics Division of Department of Economic and Social Affairs (DESA) conceived of the UN.data.org portal in 2008 as a way of streamlining the open data at the UN. Since 2008, the un.data.org portal has made available countless datasets from its 18 partner agencies and regularly releases new data sets through its portal. Data sets address such disparate topics as daily newspaper circulation, public expenditure on education, food and agricultural crop data, a greenhouse gas inventory, mobile telephone subscription rates, and employment to population ratio, to name a few. The goal of the portal is to streamline all UN open data so that it is readily accessible to whoever might need it, be they policy makers, students or journalists.
Challenges identified: As noted by the working lunch participants, challenges persist within the UN as it seeks to make its data more open. Some of the core issues identified by the meeting participants include the following:
- Establishing an Open Data Culture Before data is released. Compelling value propositions must be considered so that agencies are incentivized to open up their data. This could entail creating economic incentives (i.e., creating innovative business schemes such as that which the International Statistical Programs Center created where they charge global north member states to use their data while making it freely available to global south member states). Some member states won’t need to be incentivized to open up their data, and merely knowing that other member states are opening their data will be enough to convince them to do the same. However, this will not be the case for all member states. Transparency is valued by some UN member states, but not all prioritize this ideal; so other value propositions must be outlined that will make opening data attractive.
- Ensuring Quality and Governance of Open Data. As greater volumes of both statistical and programmatic data become open, how can the UN ensure the integrity and quality of its data? Additionally, high quality data is necessary to reach meaningful conclusions about the different development challenges the UN tackles. Implementing high standards for data collection will preserve data quality downstream, and minimize data cleaning efforts prior to data release. This requires creating agreed-upon standards both when data is collected and when it is released. Creating standards will also prepare agencies to address the privacy and security concerns that will inevitably arise as data is opened, particularly in sectors where privacy issues and opening data often intersect (i.e. the health sector). Establishing these standards will also minimize privacy and security concern when releasing data in fragile member states where opening data can have unintended consequences. For example, releasing procurement data in volatile states can endanger organizations currently engaged in contracts with UN agencies. Likewise, releasing the names of journalists could also have grave consequences; so establishing protocols will help outline ways to approach these delicate issues.
- Making Data Computable. Releasing data in a useable and machine-readable format is a vital step in ensuring data is of high value. If data is released in a format that is not immediately useful, it diminishes the benefit of having opened the data. Additionally, when data is released in a machine-readable format, it increases the value of its associated metadata. Metadata, which is essentially data about the data, can shed light on important information such as the frequency in which data sets are accessed, which can then inform how UN agencies prioritize which data sets they release.
- Measuring Impact. How do we begin to measure and document the benefits of open data? Measuring impact in terms of its economic value and how it helps UN agencies meet their mandates would be particularly valuable in outlining the concrete benefits of open data. Measuring impact will be a critical component of spurring UN-wide open data adoption.
Many of the challenges that the working lunch participants highlighted lead to actionable next steps that will facilitate the continual release of both statistical and programmatic data by UN agencies. Below are the near-term action items that arose from the working lunch discussion:
- Develop the value proposition of Open Data for the UN: Compelling value propositions and incentives must be outlined to convince member states to open their data. These value propositions should take into account the security and privacy issues outlined above, particularly when dealing with inherently sensitive data (such as health data) or when opening data in member states that are fragile.
- Start releasing evidence that documents the real impact of open data: Robust evidence that supports the benefits of open data will be critical in convincing member states of the economic benefits of open data. Evidence must be gathered that supports value propositions for opening up data. To aid this effort, agencies must begin opening up their data, even if it is imperfect. Knowing where there are gaps will help to identify the priority areas for creating standards for data collection and dissemination. The goal is to release data that is of highest value to the public, and pressure from citizens and civil society will be a main driver in getting UN agencies to open their data. It will also be important for UN agencies to clarify and communicate the security and privacy concerns when opening data and establish best practices around this issue (and opening data in general).
- Experimentation: Begin delivering on open data through small-scale experiments rather than agency-wide initiatives that will likely move slowly and will require buy-in from the highest levels. Starting with smaller experiments will be a way of creating catalytic change.
- Collect Stories: In addition to building a evidence base that highlights the benefit of open data, stories and anecdotal evidence will illustrate how different UN agencies are using open data to help meet their mandates.
- Standardization and Governance: Establish standards for collecting and releasing data so that it has the highest possible utility for users, while ensuring the safety of entities that interact with the UN, particularly those within fragile member states.
- Richard Boly, Intrapreneur at the World Bank
- Matt Burton, Technology Advisor
- Mark Cardwell, Chief, Online & Multimedia at United Nations Development Programme
- Jonathan Cousins, Designer and Programmer at Cousins & Sears, LLC
- Case Dorkey, Director, Business Development at Panjiva, Inc.
- Joel Gurin, Senior Advisor to The GovLab and author of Open Data Now
- Ken Herman, Senior Advisor on Information Management Policy Coordination with the Chief Executives Board for Coordination at the UN
- David Kanja, Assistant Secretary-General for the Office of Internal Oversight Services at the UN
- Beth Noveck, Co-Founder and Director of the GovLab
- Anders Pedersen, Community Coordinator at OpenSpending at Open Knowledge Foundation
- Matthias Reister, Senior Statistician, Office of the Director at the United Nations Statistics Division/ Department of Economic and Social Affairs
- James Nick Sears, Co-founder and Interactive Technologist at Cousins & Sears, LLC
- Amit Upadhyay, Adviser at the United States Mission to the UN
- Stefaan Verhulst, Co-Founder and Chief Research and Development Officer
- Asiya Wadud, Research Fellow and the Governance Lab at NYU
- Alexis Wichowski, United States Mission to the UN