Navigating the Politics of Open Data

Overview

On 15 July 2019 the Individual Deprivation Measure (IDM) Program^[1] alongside Open Data Watch and UN Women, with support from the Global Partnership for Sustainable Development Data, hosted a High-Level Political Forum (HLPF) side event, Navigating the Politics of Open Data: Learning from producers, users and partners. The event was designed as a multi-stakeholder dialogue on the uncertainties that disaggregated data and new measurement approaches can create and how to manage them while supporting efforts towards better, more accessible and open development data. The intention was to broaden the conversation from technical aspects of open data to the political economy of open, disaggregated data and the challenges and opportunities that come alongside. This was seen as important to create conversations that help the development data community get to the ‘good side of politics’.

The event brought together participants from around the world who gathered in New York City for the HLPF. It included individuals playing key roles within government, multilateral organizations and civil society to share approaches, issues and opportunities based on their experience and expertise. Speakers included colleagues from United Nations agencies such as the United Nations Statistics Division (UNSD), country representatives and civil society. The event was held under the Chatham House Rule.^[2]

This outcome document highlights the main takeaways from the event and offers recommendations for navigating the challenges and maximizing the opportunities of making data open, accessible and properly disaggregated. It is organized by four themes: Disaggregating the Data, Opening the Data, Ensuring Privacy and Trust, and Navigating the Politics. It is our hope that it can further contribute to advancing the dialogue and action in support of open, development data.

Background and motivation of the event

The Sustainable Development Goals (SDGs) are driving an increased focus on measurement. Yet, what is measured and not measured reflects particular development priorities, which leads to greater visibility and attention for some issues over others. Demand for disaggregated data is growing, and it is vital to understand how progress towards the SDGs translates into better outcomes for all. However, new approaches and insights from new measurement tools, as well as growing expectations to open data, can create uncertainties for stakeholders about what these might show and how to locate in the landscape of existing data. Overcoming uncertainties and navigating politics and sensitivities, including those regarding openness and privacy, are vital to remove barriers and support the availability of more quality data to accelerate the achievement of the SDGs.

This is important because, as of May 2019, there are still at least 122 of 232 SDG indicators that are classified as Tier 2 or Tier 3.^[3] For Tier 2 indicators, this means that while the tools themselves may be established, countries will have different levels of experience using the tools and working with the insights they produce. For Tier 3 indicators, it means there is still a need to develop, test and refine measurement approaches. Even for the 104 SDG indicators that have established methodologies with regular and widespread data collection (Tier 1), the limitations of current measurement approaches mean there may still be missing insights – particularly in relation to gender, which remains a significant gap in data, worldwide. This was highlighted in the March 2019 Bridging the Gap: Mapping Gender Data Availability in Africa Report by Data2x and Open Data Watch. While the report looked at both SDG and non-SDG indicators in 15 African countries, it found that 48 percent of the gender-relevant indicators were either missing entirely or lacked sex disaggregation.^[4] This highlights the challenge of working only with established methodologies if we are to get beyond business as usual and achieve the SDGs while leaving no one behind.

The disaggregated data needed to make visible and address the overlapping and structural discrimination contributing to the vulnerability of particular groups – children, youth, persons with disabilities, people living with HIV, older persons, indigenous peoples, refugees, internally displaced persons, and migrants, as specified in the 2030 Agenda, are sparse. Understanding who is benefiting from development, in what ways and to what extent, and who is being left behind requires open and disaggregated data and achieving this requires an understanding of the accompanying politics.

The urgency is not only for data to be collected but for data to be accessible by all in a timely and comprehensive manner. According to the Open Data Charter, a collaboration between governments and organizations working to open up data based on a shared set of Principles, data should be “open by default”. This means that government data should be made open and available for the public to find, access, and use under an open and unrestricted license, unless there is a specific, pressing reason why that data cannot be made open and that reason is clearly communicated to the public.^[5] This initiative is part of a broader movement for greater transparency and accountability of governments and citizen rights to public information.

Increasing demand in relation to both the range of data and its accessibility, underpinned by the requirements of the SDGs regarding disaggregated data, sit alongside continuing financing constraints. The result is a data ecosystem under pressure. Meeting changing expectations requires political and financial support. Unfortunately, despite the clear benefits of quality data, including to inform and help realize the 2030 Agenda, statistics worldwide remain underfunded. And while essential, adequate funding is not sufficient in itself to manage the political sensitivities or contention that disaggregated data can present. In some cases, such data may reveal significant inequalities among regions or particular social groups that averages tend to mask, attracting critique and demands for action. Privacy and security concerns also enter the conversation as making disaggregated data openly available becomes a focus.

It was with these challenges in mind that the co-hosts identified the HLPF as an opportune time to convene stakeholders from the data community as they met in New York City to discuss progress towards the SDGs. The event serves as a reminder to take stock of the challenges and recognize the opportunities and progress made – at both technical and political levels – in opening and disaggregating data for the SDGs.

Producing disaggregated data

The 2030 Sustainable Development Agenda, including the commitment to leave no one behind, has increased demand for disaggregated data to understand how people’s circumstances are affected by sex, age, disability, and other factors. Individual-level data is key to revealing which groups are experiencing what barriers.

Panelists working in various official roles in their country or in multilateral institutions shared their experiences in meeting the disaggregated data demands of the SDGs by navigating both the supply side and demand side of the data value chain. On the supply side, countries have begun to develop country data portals that present SDG indicators with national and subnational data sources to strengthen local ownership and relevance of the SDGs. A number of panelists took the opportunity to show these portals and their functionality with those gathered.

There were references throughout the panel discussion to the importance of improved legal frameworks as a foundation to build and sustain data disaggregation, production and use, both at the national and international level. One panelist shared how their country has worked with the UNSD to train line ministries on the importance of data disaggregation and its accompanying metadata. These trainings and engagements stemmed from internal demand to have more disaggregated data to guide policy action as aggregates were proving not to be useful. At the international level, global standards such as the Fundamental Principles of Official Statistics^[6] have proven helpful to integrate agreed data standards and guidelines into new areas of disaggregation or inclusion of new data sources, such as geospatial information.

On the demand side, the Inclusive Data Charter (IDC)^[7] serves as a strong example of overcoming the political barriers to granular data while keeping technical and methodological ones in sight. The IDC was developed by a task team of Global Partnership for Sustainable Development Data partners to mobilize political commitments and meaningful actions to deepen disaggregation. The IDC is a multi-stakeholder mechanism with a core group of governments, international organizations, and civil society organizations that signed on as the first champions. It launched publicly at the 2018 HLPF. By the end of its first year, it had mobilized public commitments from the Department for International Development (DFID) in the U.K., the Government of Kenya, the World Bank, the Internal Displacement Monitoring Centre and the Ministry of Basic and Senior Secondary Education in Sierra Leone.

Another panelist shared how the demand for disaggregated data was being addressed in their country. Ensuring the government was meeting the no one left behind principle of the SDGs required integrating the perspectives of numerous stakeholders including non-state and non-technical actors. To navigate this, the government developed an SDG Centre with nine universities across the country to build data literacy and demand and harness the knowledge of the academic sector to improve the ability to meet the disaggregation challenge.

The discussion also touched on current opportunities to maximize and improve data disaggregation for the SDGs. Such opportunities include the 2020 census, specifically the emphasis on digitalization and incorporation of geospatial data; the potential of civil registration and vital statistics as a source for additional data; the use and integration of new tools such as the Individual Deprivation Measures (see Box 1 below); and the use and integration of citizen-generated data.

A new, gender-sensitive and multidimensional measure of poverty: The Individual Deprivation Measure

New measurement approaches are needed to address data gaps or gender-blindness in available data. Making the case for better data that measures what matters for gender equality and can be disaggregated is key to improving the availability of quality data. One example is the Individual Deprivation Measure (IDM), a new individual-level, gender-sensitive, measure of multidimensional poverty. The IDM has been developed to assess deprivation at the individual level and overcome the limitations of current approaches that measure poverty at the household level.

The IDM collects primary data from all adult members of a household on 15 dimensions of life, plus financial deprivation, including ones particularly relevant to the circumstances of women such as voice, time use and family planning. The resulting data provides insight into the uneven experience of deprivation both inside and outside the household – a feature unique to IDM data. The data can also be used to reveal the relationship between poverty and factors such as gender, age and disability, to inform priorities, support targeted action, and assess impact. The IDM makes it possible to see in data where deprivation and inequality are concentrated (which groups, what dimensions, where) and how factors overlap and intersect to deepen marginalization. As a new measure generating new insights and disaggregated data, it speaks to both the technical and political aspects of the disaggregated data challenge.

For more information visit: www.individualdeprivationmeasure.org

Opening up disaggregated datasets

Disaggregated datasets are only valuable to advocates and decision makers if they can access them. The openness and accessibility of these datasets is crucial to realizing the full social and economic benefits of the data and to leave no one behind.

Disaggregated datasets that are open and available to the public for use produce more societal value than closed datasets as they can be freely used to analyze circumstances (of particular groups, in particular locations), advocate for action, improve government programs, and inform citizens. The 2018-19 Open Data Inventory (ODIN)^[8], released in March 2019, provides the latest assessment of coverage and openness of official data and statistics in 178 countries. It shows an improving trend overall, with 11 countries increasing data openness by over 35 percent in two years. But as more disaggregated datasets become available, consistent political support and financing are needed to continue this progress. Changes to legal frameworks and access to information policies may also be needed to support open data policies. Incorporating open data in countries’ national strategies for the development of statistics (NSDS), and in the planning and implementation of SDG national reporting platforms, can signal the value of open data and provide a foundation for shifting the ‘default’ towards making data open. In a context where financing for statistics is constrained, making data open helps to maximize its value.

At the global level, UNSD is promoting open data through several initiatives such as the Working Group on Open Data as part of the Friends of the Chair of the Statistical Commission, to review countries’ uptake of the Fundamental Principles of Official Statistics (FPOS). The Working Group provides tools, standards and frameworks for countries to open their data to the public and engage with users, while the Friends of the Chair group aims to expand implementation of FPOS, with open data as an important part of this. Such political and technical support for open data at the UN-level signals the importance of open data and movement towards it becoming the standard.

Countries must also have the legal, technical and political frameworks in place to successfully open data and provide the types of disaggregation that the SDGs require. This may require legislative action such as revising the mandate of the national statistics office (NSO). One of the expert panelists noted that a key challenge in navigating the complexities of data disaggregation and opening data stems from the absence of legal frameworks and strong mandates for NSOs to engage with all actors in the statistical system. As the data ecosystem expands, NSOs are the leading connector between producers and users. A stronger commitment from governments is needed to enable NSOs to have a clear role in SDG and data discussions from the start. NSOs in turn need to improve their communication with policymakers, specifically ministries of planning.

Such examples show the interactions between the supply and demand for data and governance arrangements. The SDGs are increasing demand for data. This is affecting the scope of data collection, increasing the drive to ensure development progress is country-owned, and spurring further data demand. Unlike the Millennium Development Goals (MDGs), which were more concerned with global indicators and data at the international statistical systems level, the SDGs put national statistical systems in the driver’s seat. National-level responsibility for monitoring and measuring the indicator framework has both advantages and disadvantages. It ensures that the agenda is country-owned but puts additional stress on systems that are under-resourced and under-funded. Ensuring national statistical systems have the capacity to produce the data necessary to measure and monitor progress on the SDGs – and make the data open and accessible to the public – will take time and resources. This investment in strengthening both data and national statistical systems is essential if countries – and the people who live in them – are to know how commitments to the SDGs are translating into changed lives.

One country representative on the panel spoke of the contrasting country experience in moving from the MDGs to the SDGs. The MDGs were not aligned with national priorities or plans; they were imported from outside. The SDGs, however, were used as a framework for national planning from the start, allowing for a greater alignment between priorities outlined in the SDGs and national priorities. The same sentiment was expressed by another panelist, noting that the SDGs were the global indicator framework that guides their own use and production of national proxy indicators.

As evident from the remarks of the panelists, disaggregating and opening the data required to measure and monitor the SDGs requires not only technical capacity but political conviction and leadership. Technical capacity is supported through the development of standards and principles to help bring coherence within and across systems to maximize the value of data and enable more producers to fill gaps with quality data. Political conviction and leadership are demonstrated through the renegotiation of the role of NSOs as both data producers and data curators and supporting this shift with facilitative legislation and, importantly, funding.

Open Data Watch’s latest efforts to uncover the availability of gender disaggregated data

Open Data Watch (ODW) is an international, non-profit organization of data experts working to bring change to organizations that produce and manage official statistical data. ODW supports the efforts of national statistical offices (NSOs), particularly those in low- and middle-income, to improve their data systems and harness the advancements of the data revolution. In the recent Bridging the Gap: Mapping Gender Data Availability in Africa report, prepared by ODW with support from Data2X, ODW examined 15 countries and the availability of gender-disaggregated SDG indicators from international and national databases. A list of 104 indicators of relevance for identifying the status and welfare of women was selected from the gender indicators proposed by the United Nations Inter-Agency and Expert Group on Gender Statistics (IAEG-GS) or by UN Women or included in the SDGs. A team of assessors evaluated the availability of all 104 indicators in international and national databases and found that 48 percent of gender-relevant indicators are missing or lack sex-disaggregated data in the study countries at both international and national levels. This persistence of large gaps in international and national databases points to the need for a coordinated effort to improve data collection and adopt common standards for the compilation of indicators.

For more information visit: https://opendatawatch.com/publications/bridging-gender-data-gaps-in-africa/

Provide access to data while protecting privacy

Improving disaggregation of data may involve changes to current survey practices. For example, surveys that collect information at the individual level rather than the household level and avoid questions that ask respondents to speak on the circumstances of others may yield more in-depth and accurate insights. Additionally, collecting a broader range of demographic information will further enable social group analysis. For multitopic surveys that collect a range of personal information about individuals, usual privacy considerations are amplified. Risks of re-identification of participants may be a particular concern in sparsely populated areas (where combinations of personal identifiers are more likely to be unique) and in close knit communities (where people know a lot about each other and personal knowledge may be more readily combined with survey data). However, there will always be some risk of re-identification if the data are to be useful. A balance needs to be found between privacy and data utility. Both are important, and both have technical and political implications that need to managed.

Many panelists and attendees stressed the importance of preserving privacy while opening data. The continued expansion in computing power has demonstrated the possibility of re-identifying individuals in published data and custodians must therefore carefully consider where managed disclosure of data may be necessary. This also underscores the trade-off between technical solutions and political management.

To improve accountability and encourage innovation at the local, national, and global level, data must be accessible and analyzable. Processes such as those championed by the Open Algorithms projects (OPAL) could help drive research and accountability while allowing for privacy and ownership concerns, by enabling outsiders to query data and conduct analysis without owning the entire dataset. Many of the problems surrounding data localization, which is a topic for governments across the income spectrum, could be addressed in this way.

Another protective approach is for custodians of data to distinguish between policy and operational purposes. One panelist gave the example of how data on tax evasion is used to improve policies around closing loopholes and ensuring systemic fairness regardless of income. Although this data could be used to prosecute individuals for tax evasion, this course of action is not pursued as it was not collected for this purpose and would compromise the NSO’s independence.

UN Women’s efforts to ensure every woman and girl is counted

UN Women’s flagship program, Making Every Woman and Girl Count, aims to bring about a radical shift in how gender statistics are used, created and promoted. The program seeks to address the urgent need to increase the availability of accurate information on gender equality and women’s rights in order to inform policy and decision-making.

Making Every Woman and Girl Count will be implemented between 2016 and 2020. At the global, regional and national levels, the program will provide technical and financial support to countries to improve the production and use of gender statistics in order to monitor the implementation of gender equality commitments in the 2030 Agenda. Adequate monitoring frameworks will, in turn, inform policies and programs that can spur meaningful and lasting changes in the lives of women and girls everywhere.

For more information visit https://www.unwomen.org/en/how-we-work/flagship-programmes/making-every-woman-and-girl-count

Navigating the politics and maintaining trust in data

Through their agreement to the SDGs, 198 governments around the world committed to leave no one behind in realizing the 2030 Agenda. As the world enters the “Decade of Delivery”, cooperation between all stakeholders is key. Panelists spoke of the importance of coordination between NSOs, ministries of planning and all levels of government, while remaining engaged with civil society and the private sector to take advantage of advancements linked to technical innovation and collaboration. Within an increasingly collaborative environment, however, the independence of statistics agencies is key in managing the challenges and risks that government data can be politicized, which lowers public trust. Supporting data providers and custodians with legislation that enshrines the independence of the statistics office is therefore paramount. So too is the ability of NSOs to collaborate and operate with diverse actors across the data ecosystem, including civil society.

Building norms, communication and coordination between statistical agencies and other ministries, and with civil society, are important processes in order to navigate the politics of data. Publishing data results, opening datasets, and further disaggregating data will likely show variation and relative lack of progress for some groups compared to aggregate results, which policymakers might find challenging given potentially negative political implications. Bringing line ministries into partnerships with statistical offices and independent civil society organizations can help form alliances that support constructive reception of statistical releases, particularly if those partnerships are designed to weather administrative and political changes.

Building political support for legislation that guarantees independence, coordinating partnerships, and aligning work programs with national development priorities takes time, money and effort. In all countries, but especially in low-income and fragile contexts, NSOs face resource challenges, fragmented programs, a lack of coordination and a lot of politics to be managed. These interrelated challenges need to be tackled in an integrated way if the promise of the data revolution is to reach every citizen. This requires a supportive global system for statistical capacity, while ensuring country ownership and strong NSO leadership.

While NSOs are required to support data production and manage complex relationships across the data ecosystem, there is a growing opportunity for NSOs to work alongside other entities to fill gaps and improve the capacity of the statistical system overall. One panelist shared how they used the support of civil society organizations and politicians to navigate the challenges. In this case, a technical committee was formed to address gender data gaps and included a range of civil society organizations as well as national statisticians. The result was increased understanding on both sides of the table regarding official statistics, citizen-generated data, and general data priorities. This reinforced the importance of focusing not only on disaggregation but also on what is being measured, why, and how data will be used. In another example, civil society organizations have teamed up with the NSO to develop guidelines for citizen-generated data. The plans are to incorporate this into the NSDS. The increased collaboration as a result of stakeholder engagement is leading to greater openness, fewer gender data gaps, and increased trust in official and non-official data systems. Data that helps citizens and local leaders navigate the issues most pertinent to them can help secure political support and funding for NSOs, while building the trust.

Conclusion

Panelists spoke to a range of challenges they face in navigating the politics of open data. These included:

- Incorporating new sources of data such as individual-level poverty data by non-official producers, citizen-generated or geospatial data;
- Recognizing the shifting roles of NSOs in the context of the SDGs, the range of data they require, and growing numbers of data producers;
- Securing political support;
- Ensuring appropriate legislation and frameworks are in place to enable the work of NSOs;
- Coordinating across government and with new producers; and
- Building and maintaining public trust in data.

With roughly 10 years left to deliver the 2030 Agenda, there were also a number of opportunities noted by the expert panelists to help the data community contribute including getting the right data into the right hands to leave no one behind. These opportunities include:

- Utilizing technology to gather (geospatial) and present (web portals) data in accessible ways for multiple audiences to build and meet demand while increasing trust;
- Maximizing the role of political support and champions within national and international contexts;
- Harnessing the increasingly important role of the national statistical system as part of the solution for meeting the data demands of the SDGs;
- Raising awareness of gaps in disaggregated data;
- Testing and refining new measurement tools and approaches to reveal new insights; and
- Growing the support for coordination, engagement and collaborative action by multi-stakeholder groups at national and international levels.

To produce quality, open, inclusive, and accessible data, the data community must address technical, coordination and quality issues on the supply side and use collaboration, trust, technology and innovation to respond to the demand side. It is our hope that through a better understanding of both, a virtuous data cycle can be supported, where increased demand, trust and capacity help leverage investment, production and quality to create a stronger data ecosystem.

Transformation from a vicious cycle to a virtuous cycle

Source: Open Data Watch

List of Speakers

Shaida Badiee, Managing Director, Open Data Watch

Rachael Beaven, Data Revolution Lead, Data for Development Team, Department for International Development, United Kingdom

Lorant Czaran, Senior Technical Advisor & Lead, GRID3 Project Secretariat, United Nations Population Fund (UNFPA)

Francesca Perucci, Chief of the Statistical Services Branch, UN Statistics Division

Joanna Pradela, Director, Individual Deprivation Measure team at International Women’s Development Agency

Dr. Diani Sadia Wati, Special Adviser to the Minister on Institutional Relations, Ministry of National Development Planning, Indonesia

Papa Seck, Chief Statistician, UN Women

Omar Seidu, Principal Statistician, Ghana Statistical Service

Jenna Slotin, Senior Director for Policy and Strategy, Global Partnership for Sustainable Development Data

Brigida Soares, Planning, Monitoring and Evaluation Coordinator, Office of the Prime Minister of Timor-Leste

About UN Women

UN Women is the United Nations entity dedicated to gender equality and the empowerment of women. A global champion for women and girls, UN Women was established to accelerate progress on meeting their needs worldwide. UN Women supports UN Member States as they set global standards for achieving gender equality, and works with governments and civil society to design laws, policies, programmes and services needed to ensure that the standards are effectively implemented and truly benefit women and girls worldwide.

About Open Data Watch

Open Data Watch is an international, non-profit organization of data experts working to bring change to organizations that produce and manage official statistical data. We support the efforts of national statistical offices (NSOs), particularly those in low- and middle-income countries, to improve their data systems and harness the advancements of the data revolution. Through our policy advice, data support, and monitoring work, we influence and help both NSOs and other organizations meet the goals of their national statistical plans and the SDGs.

About Individual Deprivation Measure

The Individual Deprivation Measure (IDM) is a new, gender-sensitive measure of multidimensional poverty, developed to assess deprivation at the individual level, generating a more granular understanding of poverty and overcoming limitations of approaches that measure poverty at the household level. The current four-year Program to ready the IDM for global use by 2020 is a partnership between the Australian National University (ANU) and the International Women’s Development Agency (IWDA) with strategic support from the Australian Government. The original research that developed the IDM was a four-year, international, interdisciplinary research collaboration, led by ANU, in partnership with the IWDA and the Philippine Health and Social Science Association, University of Colorado at Boulder, and Oxfam Great Britain (Southern Africa), with additional support from Oxfam America and Oslo University. It was funded by the Australian Research Council and partner organisations. Subsequent IDM research undertaken in Fiji was led by IWDA in partnership with the Fiji Bureau of Statistics with funding from the Australian Government’s Pacific Women Shaping Pacific Development program.