English

La gobernanza de datos habla de quién tiene autoridad y control sobre los datos y a la forma en que estos pueden ser utilizados, pero el término tiene diferentes significados y dista de ser claro en su aplicación. Como explicó un participante en la conversación inicial de una serie de diálogos regionales, "la cuestión es cómo integrar la democracia" en la gobernanza de los datos.

Las oficinas de estadística de América Latina y el Caribe (ALC) cada vez más se enfrentan a la necesidad de establecer marcos de gobernanza de datos en medio del rápido cambio tecnológico, las nuevas necesidades de información, aunado a una creciente preocupación en torno a la privacidad de los datos. Representantes  de la región están buscando encontrar su lugar en el creciente ecosistema de demanda, producción y uso de datos.

Compartir conocimientos e ideas entre los proveedores y usuarios de datos es el primer paso para navegar por los cambiantes ecosistemas de datos y replantearse sus competencias, prácticas y responsabilidades. El fortalecimiento de la colaboración regional es fundamental para crear mayor claridad en torno a la gobernanza de los datos y abordar la complejidad y la urgencia de estas cuestiones. 

En noviembre, líderes de las oficinas de estadística de Colombia y República Dominicana, organizaciones de la sociedad civil e instituciones de investigación intercambiaron perspectivas sobre la gobernanza de datos durante un evento paralelo de la Undécima Conferencia Estadística de las Américas. El debate puso de manifiesto algunos de los principales retos y recomendaciones que ayudarán a orientar las futuras conversaciones entre los líderes regionales.

Los institutos de estadística se enfrentan a leyes obsoletas, bajos niveles de confianza pública y recursos limitados.

Los marcos legales obsoletos y débiles obstaculizan la capacidad de las oficinas de estadísticas para avanzar en la gobernanza de los datos. A menudo, la legislación estadística se remonta a años atrás y está desfasada respecto a los últimos avances tecnológicos, así como a la emergente economía de los datos. Este tipo de leyes limita la capacidad de los institutos nacionales de estadística para estar a la vanguardia de la gobernanza de datos y coordinarse en todo el ecosistema de datos. En los casos en los que no es posible cambiar el marco jurídico a corto plazo, los institutos de estadística están tomando medidas para alinearse con las estrategias nacionales y hacer que la gobernanza de los datos ocupe un lugar más central dentro de estas estrategias.

Además, la falta de marcos jurídicos sólidos que promuevan eficazmente la independencia técnica de los institutos de estadística repercute en la percepción de confianza sobre las estadísticas oficiales. La ausencia de un mandato definido, de un ámbito de aplicación y de limitaciones para las oficinas estadísticas como administradores de los sistemas estadísticos nacionales ha dificultado la creación de confianza entre los nuevos usuarios y miembros del ecosistema de datos. Los participantes destacaron que la promoción de una cultura de la eficiencia y la innovación dentro de las oficinas estadísticas podría contribuir a reforzar la confianza en la institución.

Los participantes de las oficinas nacionales de estadística también explicaron que se enfrentan a un creciente desequilibrio entre las demandas y los recursos. Las capacidades y los recursos institucionales de las oficinas de estadística rara vez están a la altura de las crecientes expectativas y demandas de los nuevos productores y usuarios de datos , así como de la rápida evolución de los debates en torno a la gobernanza de los datos.

El camino a seguir

Los productores de datos deben asegurarse de que los datos representen los intereses de los sujetos de los datos, es decir, las personas que los proporcionan. "Nuestro papel es basar los marcos de gobernanza de datos en la realidad con un enfoque ascendente", explicó Julia Zulver, de Ladysmith.

Los participantes señalaron desarrollos y enfoques prometedores para los desafíos compartidos, acentuando la importancia de localizar las normas y los marcos, mejorar la transparencia y fomentar la colaboración. A continuación, se presentan las ideas principales que fueron el resultado del diálogo sostenido:

1. La privacidad de datos debe adaptarse a las necesidades locales y a los contextos políticos.

La preocupación sobre la privacidad cambia entre países y las expectativas de los individuos sobre el uso de sus datos por parte de los gobiernos son variables. Por ejemplo, en los países nórdicos las declaraciones de impuestos de todos los individuos son registros de acceso público, pero esto no es la norma en los países de AL&C. El contexto específico de la privacidad y las preferencias de los individuos hacia sus datos requieren de enfoques y soluciones a la medida. 

Fabrizio Scrollini, Director Ejecutivo de La Iniciativa Latinoamericana por los Datos Abiertos (ILDA), destacó en el evento que se debe ir más allá de la consideración de la privacidad a nivel individual para reconocerla como una cuestión comunitaria y un fenómeno colectivo. Esto es especialmente importante cuando se consideran los grupos históricamente marginados y los riesgos potenciales de una mayor discriminación a través de las prácticas de datos. 

2. Promover la transparencia y los sistemas de gobernanza de datos inclusivos genera confianza y credibilidad pública.

Patricio del Boca, desarrollador de software senior de la Open Knowledge Foundation, destacó que las oficinas estadísticas deben mejorar la transparencia de datos, metodologías y procesos para fomentar un mayor uso de los datos y fortalecer la confianza de los ciudadanos en la recolección, acceso y uso de información por parte de los gobiernos. Sin transparencia, tanto en los datos, como en los procesos de recolección y análisis de estos, explicó, puede generarse una falta de credibilidad derivando en una obstaculización en el aprovechamiento de los datos.

Las oficinas nacionales de estadística también pueden aumentar la confianza adoptando modelos de gobernanza de datos recíprocos, en lugar de extractivos. Los modelos recíprocos implican la recopilación de datos de diversas fuentes (públicas y privadas) a la vez que se devuelven a esas fuentes de datos. Este enfoque ha ayudado a la ONE de la República Dominicana a obtener acceso a datos meteorológicos y sobre el agua. A cambio, la ONE refuerza las competencias de los productores de datos, y les ayuda a reorganizar y actualizar sus procesos de recolección de datos y sus bases de datos. 

3. La colaboración de múltiples actores interesados es fundamental para lograr la equidad y la inclusión a través de la gobernanza de los datos.

Al trabajar con mujeres en situaciones vulnerables, Julia Zulver, investigadora principal de Ladysmith, reconoció que la recolección de datos puede poner en peligro a las personas. Para proteger a los sujetos de datos, explicó que es importante evitar la recopilación de datos por el simple hecho de hacerlo. En su lugar, la recolección de datos debe tener una finalidad específica y estar vinculada a las preocupaciones y los cambios que las comunidades afectadas desean abordar. Este objetivo requiere que los investigadores co-diseñen programas y trabajen estrechamente con organizaciones de base para garantizar que los datos aborden las necesidades de las comunidades y se utilicen para amplificar las voces de las personas que son sujetos de datos.

Juan Daniel Oviedo, Director General del Departamento Administrativo Nacional de Estadística (DANE) de Colombia, explicó que el DANE tiene el mandato legal y constitucional de mejorar la visibilidad de las minorías en la producción estadística y ha trabajado estrechamente con los ministerios sectoriales para lograrlo. Esto requiere considerar no sólo la forma en que las oficinas de estadística responden a los nuevos mandatos, sino también la forma en que se involucran otros actores del sistema estadístico nacional.

El objetivo de la gobernanza de los datos debe responder a las necesidades específicas de las personas que están en el centro de los datos y proporcionar un marco para protegerlos y compartirlos para el bien público. Como explicó Patricio del Boca, de la Open Knowledge Foundation, trabajar conjuntamente para crear ecosistemas de datos y marcos de gobernanza ayuda a mejorar la calidad y el acceso a los datos para el bien público. 

Mantener la colaboración regional

El diálogo, así como la cooperación regional e internacional, son fundamentales a medida que los institutos nacionales de estadística avanzan hacia un nuevo ámbito de configuración de los marcos de gobernanza de los datos, manteniendo al mismo tiempo la confidencialidad, la privacidad y la calidad de los datos.

En los próximos meses, el DANE y el GPSDD continuarán explorando la gobernanza de datos y temas relacionados con actores de toda ALC a través de una serie de actividades, incluyendo:

• La creación de oportunidades para el liderazgo, la consulta y la participación de AL&C en el Data Values Project, una campaña de consulta e incidencia de política que tiene como objetivo desbloquear el valor de los datos para todos.
• Reforzar las asociaciones entre ciudades, organizaciones locales y actores a nivel nacional para evaluar y replantear conjuntamente el modo en que la gobernanza de los datos puede permitir soluciones normativas.
• Profundizar en nuevos marcos, conceptos y experiencias a través de un nuevo grupo de trabajo sobre la administración de datos (data stewardship) en el marco de la Conferencia Estadística de las Américas - Comisión Económica de las Naciones Unidas para América Latina y el Caribe, y reforzar el intercambio de conocimientos con el Grupo de Trabajo sobre la Administración de Datos de las Naciones Unidas, bajo el co-liderazgo del DANE y de Statistics Poland.
• Fomentar los intercambios de conocimiento regionales y Sur-Sur sobre prácticas de datos inclusivas, interseccionales y participativas a través de la Carta de Datos Inclusivos (Inclusive Data Charter-IDC).

Deseamos colaborar con una amplia gama de organizaciones de toda la región y agradecemos las ideas y los comentarios. Póngase en contacto con nosotros en [email protected].

Camilo Mendez y Daniel Peñaranda del Departamento Administrativo Nacional de Estadística (DANE) de Colombia con Jenna Slotin y Fredy Rodriguez del Global Partnership contribuyeron a este blog.

Notas: El Departamento Administrativo Nacional de Estadística (DANE) de Colombia y la Global Partnership for Sustainable Development Data, han reunido a estadísticos, investigadores y activistas de toda América Latina y el Caribe para explorar las normas que deberían orientar la gobernanza de datos y el quehacer de las oficinas nacionales de estadística. Este diálogo regional se integrará en un proceso global de política e incidencia: el Data Values Project.

Data governance remains a difficult concept to unpack even as efforts to understand what good data governance looks like have recently multiplied across the world. The Centre for the Study of the Economies of Africa (CSEA) is leading a number of research activities aimed at clarifying what data governance means for the continent and what should be done to set up governance frameworks in African countries. These include a recent event on African data governance (recording available here), a project and subsequent report on Strengthening Data Governance in Africa, and the African Digital Preparedness dashboard.

The Data Values Project spoke with CSEA Director of Research Dr. Adedeji Adeniran to learn about the current state of thinking on this topic and research priorities and needs for the coming years:

Data governance is a broad concept. What's your working definition of governance?

In the context of CSEA’s work, we define data governance as a set of rules, laws, and strategies used by governments and private organizations to manage, share, and analyze data. This concept also encompasses the ecosystem of players and relations that are built around data, especially those generated by new technologies. Technological innovations produce vast amounts of data and generate significant aspirations in terms of digitalization across the continent. The purpose of data governance is to make sure that these aspirations are effectively channeled and to stay away from two possible extreme scenarios: one in which the private sector has the free rein to do what it wants with these new technologies and another in which governments over-constrain innovation by means of regulations. 

Why has CSEA focused on data governance as a primary area of research?

It is precisely the need to find the middle ground between the two extreme scenarios I mentioned which is the impulse for all these activities. Further discussions on data governance are needed in Africa to find our pathway. Our continent is characterized by weak governance and institutions which allow private companies and platforms to enter the market and reap the benefits of the data economy while facing few constraints. However, these institutional challenges make Africa quite vulnerable to threats linked to digitalization and also lead to violations of rights and negative externalities which would not so easily appear in other geographies. 

If we do not start talking to private players now, they will gain too much power and influence, and it will be too late to counter possible negative consequences of their operations. At the same time, Africa has a lot of potential when it comes to the expansion of the data economy, and we want our continent to benefit from the rapid digitalization currently occurring. This means that we should not fall into the other extreme and overregulate companies which will then be pushed away and not encouraged to come into our markets. Data governance must be able to establish some rules of the game, to avoid negative externalities linked to data exploitation while at the same time enabling benefits to emerge for citizens and economic players. 

In this respect, what we have learned from the European experience on data governance is that, to establish suitable rules for the game, you must be sitting at the right table. The European Union is trying to regulate companies which are mostly based in the United States and struggling to do so because decision-makers do not sit at the same table as these large platforms and Europe does not have the right software champions to join the debate. We must learn from this experience and make sure we create a seat at the right table when it comes to establishing data governance rules.

CSEA’s African Digital Preparedness dashboard shows significant differences among African countries in data governance laws and frameworks. What are some lessons from national experiences in African countries and what can the world learn from them?

There are several lessons that can be drawn from the data collection and analysis we performed for the African Digital Preparedness webpage. Maybe the most important is that establishing good data governance frameworks is not a prerogative of the biggest countries. As our analysis shows, countries such as Mauritius, which are among the smallest of the continent, have set up better data governance frameworks than many larger countries. In this domain, it seems that the size of the country really does not affect the quality of the rules adopted.

A second key takeaway concerns the explanation of differences between countries. Our analysis suggests that the main driver of country differences in terms of rules lies in their human resource capacity. To establish good data governance frameworks you need human and social capital. This is because you need rule-makers who understand technological development to be able to adopt rules that steer and don’t prevent innovation. Human capacity is distributed unequally across Africa, which leads me to a third lesson.

Besides the growing Global North-South divide, we are also witnessing increasing inequalities across African countries. Only a few of them are truly benefiting from the rapid development of the data economy although a vast majority of countries are experiencing sustained economic growth. This considerable gap and these differences across countries also reflect a disconnect between national priorities and regional priorities. To ensure that the data economy benefits us all, we need to strengthen our African approach from a regional perspective and to act as one on the international scene. In the context of the data revolution and to grow the data economy further, Africa needs the world, but the world also needs Africa. 

What are the most important emerging trade-offs in data governance (i.e. between allowing unrestricted data flows and ensuring citizens’ data are not misused by corporations) and how can we tackle them? 

I believe that there are two major trade-offs which deserve our attention at this stage. One concerns the balance of power between the public and the private sectors. There is a fine line between giving free rein to private sector players and empowering the public sector to overregulate the economy. I am not entirely sure that we want to replace private sector failures with public sector failures. Recent examples of state intervention from different African countries show that governments might be tempted to impose unnecessary and anti-democratic (although often temporary) restrictions on the data economy (i.e. the recent Twitter ban in Nigeria or the limits on Facebook and Twitter in Uganda before the elections). Taking this into account, we should be careful about what we ask for from governments, and we should not empower governments to make bad decisions on data governance.

A second important trade-off relates to the debate between data localization and data protection. The option of relying on data localization rules as a form of data protectionism to allow the emergence of a local data economy should not be blindly followed. Africa is not Europe and formulas imported from other continents might not work well here. We need to find a good balance between enabling data flows and protecting citizens’ rights. This is particularly important as we know that the development of AI systems requires vast amounts of data which begs for cross border data sharing. One way of solving this trade-off is through a differentiation of types of data depending on their sensitivity: cross-border flows of health or financial data might need specific rules in terms of privacy which might not be needed for less sensitive data. 

Ultimately, succeeding in establishing the right data governance in Africa requires nailing down which roles the public and the private sector must play in setting up an adequate framework for the data economy.

This is the third in a series on data-sharing between public and private sectors focusing on emerging approaches and uncovering key lessons for regions and stakeholders around the world. Last time, Martina wrote about the identity crisis facing Mobile Network Operators around the world that, despite similar opportunities and challenges, are developing diverse approaches to sharing data with governments and private companies.

Mobile Network Operators (MNOs) are at a crossroads: They have data that governments can use to make better decisions but no clear models for sharing that data. This post  looks into a hypothetical crystal ball to imagine how solutions that MNOs are considering might evolve and translate into real-world scenarios for data sharing. If we specifically consider data sharing between MNOs and the public sector, a number of scenarios or trajectories appear: 

• Regulatory or legislative data sharing mandates: If MNOs in the coming years strongly resist voluntary data sharing with the public sector, governments may be tempted to create legislation to gain access. This is already happening in the European Union where proposed legislation would mandate business-to-government information exchanges through the proposed Data Act. It’s easy to envision other governments following this path. The regulatory approach gives certainty of access to the public sector but raises a number of questions in terms of proportionality and data protection. 
• Collaborative routes: If more MNOs decide to share data for public good with the public sector, the number of public-private partnerships (PPPs) will increase substantially. The proliferation of such partnerships will require more in-depth discussions about operational and business models, incentives and resourcing for these initiatives as well as long term strategies for sustainability. Along this route, many different types of partnerships will emerge depending on how challenges are addressed. 
• Profit models via the business highway: To increase efficiency of investments and economies of scale, an increasing number of MNOs could also build data platforms and pipelines aimed at selling data and related services to the private sector. In this case, the public sector will face pressure to use these same infrastructure systems and conform to the rules and business models created for private players. Public authorities could then have less freedom to negotiate specific data sharing agreements and would likely face greater pressure from MNOs to adopt commercial models. MNOs, in fact, may struggle to justify to private clients why public agencies should get the same access to data at lower costs. 

These scenarios are not mutually exclusive. In fact, most are already happening to different extents depending on where MNOs are operating in the world. While we can’t predict the future, here are likely consequences of these scenarios for the public sector:

• Public sector entities will have to strengthen data-related capacity. Whether public authorities follow regulatory or collaborative routes to sharing data, local public agencies will have to considerably strengthen their capacity to understand and put in place adequate pipelines to share data with MNOs and the private sector more broadly, both via sustainable business models and through legislation. 
• Legislators will have to carefully define what data is most needed. While the importance and usefulness of MNOs’ data for the public sector is likely to increase in the coming years, public authorities will have to walk a thin line between asking too much and asking too little from MNOs if countries pursue legislative measures to mandate data sharing. Proportionality of measures must be ensured for the public sector not to disproportionately affect MNOs in their business models and digital transformation.  This is currently reflected in the debate around the Data Act currently taking place in Europe. 
• Public officials and institutions will have to improve or rebuild public trust. Regardless of which data sharing scenarios emerge as dominant in the future, citizens’ trust in governmental use of data collected by MNOs will be a paramount concern. As governments’ data driven-responses to COVID-19 have highlighted, people do not automatically trust their governments to use MNOs’ data. The public sector needs to start building trust now by engaging with citizens in dialogue around data values and putting the right safeguards in place to protect people from government abuse and misuse of data. Ensuring that benefits of data use are felt by people and that feedback loops exist could also contribute to increasing trust. 

It’s clear from this thought experiment that the public sector should step up efforts to build capacity in this domain. No public authority will be able to take advantage of opportunities for data driven decision making without understanding the governance, legal, and operational challenges underpinning public access to privately held data. This is especially true for MNO data. These skills are crucial for the modernization of the public sector and for addressing the growing data power imbalances between public and private stakeholders. 

• Martina Barbero is Policy Manager at the Global Partnership for Sustainable Development Data.

This is the second in a series on data-sharing between public and private sectors focusing on emerging approaches and uncovering key lessons for regions and stakeholders around the world. See the first post on the European Commission's Data Act here.

Mobile Network Operators (MNOs) are experiencing an identity crisis. As the digital transformation accelerates and private companies amass increasing amounts of data sought by the public sector, MNOs are asking where they fit into this rapidly-changing data economy.

It’s widely recognized that the public sector has lost the historical data supremacy it once held. Today, private sector companies hold more and better population data than governments. As a result, there’s increasing interest from the public sector to tap into privately-held data resources. MNOs hold extensive data on their customers’ movements, personal information, and activities. Developing new data sharing services and products constitutes a means to diversify their business models and adapt to changing customer needs and expectations despite the difficulty of reconciling stakeholders’ positions on ethical data sharing. 

This summer, the Global Partnership organized discussions in the context of the Data Values Project with officials from a diverse group of MNOs from around the world. Through these conversations, our team sought to understand how MNOs envision sharing data with the public sector in the context of broader digital transformations. We learned that MNOs, despite their many commonalities, are embarking on very different journeys to data sharing. 

Why everyone wants a piece of the MNO data pie

The common starting point for all MNOs is the type of data companies hold. All the MNOs we interviewed reported considering means to safely and efficiently share call detail records, or CDRs, which are created through calls and other telecommunication activities. CDRs include event-driven data (information on calls, messages, online traffic by a mobile phone user) and network-driven data (generated at a desired frequency by the network, i.e. by cell towers). 

This data is increasingly in demand from governments as a consequence of the COVID-19 pandemic, and potential uses of this data continue to increase. From a public sector perspective, for instance, such data can guide policy related to public health and energy and in humanitarian and disaster response efforts. New potential applications of CDR data appear every day. Within the private sector, companies in finance, tourism, and retail are already eager consumers of CDR data. New customers (for example, from the energy sector) are also starting to seek data-sharing partnerships with MNOs. 

But models for MNOs to share CDR data with public and private users vary considerably. MNOs that are farther along in their digital transformations have created platforms to consistently supply this data long-term regardless of the reason for sharing or the client seeking to access CDR. Others have responded on a case-by-case basis, negotiating access and creating data sharing pipelines based on individual requests and clients. 

Mediating risk in sharing mobile network data

Regardless of an MNO’s approach to sharing data or the specific challenges of regional markets, all MNOs confront similar concerns related to data protection and privacy. These remain the biggest obstacles to data sharing—especially with the public sector. 

Data protection regulations and telecommunication rules are often specific to countries or regions. The  specific fear among MNOs that data sharing will result in breaking these laws or incurring liability toward regulators and clients is geographically agnostic. 

All MNOs are grappling with questions concerning their business models and revenue approaches to data sharing. Despite the plethora of technical options for data pipelines and infrastructure, MNOs at this stage still struggle to develop profitable and sustainable models for sharing data. There’s a general consensus that demand for CDRs will continue to increase, but questions remain regarding users’ willingness to pay for this data in the long-run.

MNO leaders today are asking very similar questions about future possibilities to share data as assets or services. Responses among companies vary considerably, leading to a range of  potential future outcomes. 

Keep an eye out for Part III in this series on public-private data sharing, which looks into a “crystal ball” of possible future trajectories for MNOs’ data sharing to help public sector organizations understand what to expect and how to gain sustainable and safe access to CDR data. Reach out to the Data Values Project via email at [email protected]. 

• Martina Barbero is Policy Manager at the Global Partnership for Sustainable Development Data.

Since 2014, when then-UN Secretary-General Ban Ki-moon called for a global effort to bring about a data revolution in sustainable development, it’s become increasingly clear that our personal data is not some abstract entity stored on a computer somewhere in space. To the contrary, current events have shown now more than ever the extent to which our data reflect ourselves and the ways data can be used both for and against us. Data holds power to improve lives. But people’s personal data can also be used against them. This is especially true for activists, members of minority communities and persecuted people from diverse backgrounds.

Today, while data for development remains high on the international agenda, concerns about governance and power dynamics are rising related to a range of issues—from tech monopolization, vendor lock-in and exploitative business models to regulatory gaps and the unequal distribution of the benefits of the data revolution. Yet connections between these issues and sustainable development remain vague, confusing and poorly understood. In the face of increasing evidence of the results of malicious data practices, the Data Values Project and DataReady hosted a much-needed conversation about the role of development practitioners in empowering people and reducing inequities. 

This online conversation—Dissecting Digital Power Inequities: reflections from digital rights experts for development practitioners—sought to highlight current issues related to digital rights, data governance and power inequity in the context of the data revolution for sustainable development. 

Here are five recommendations that emerged from the discussion for development practitioners to consider and use in shaping conversations and action around digital inequities and data governance:

1. Acknowledge harmful business models to rethink data practices in development. Current data business models and practices—even in the context of development—exacerbate the digital divide,  trigger profound inequities and produce new challenges to the goals of sustainable development. The development community must rethink its data practices and approaches to avoid reproducing harmful models and further disempowering marginalized communities. 
2. Center conversations about data justice. The principles of data justice should guide discussions around individual and community privacy rights, data sovereignty and access to key infrastructure. Dialogue is important but not sufficient: Regulations—especially at the national level—are important to protect people and lay the groundwork for a fairer data economy. Practitioners should incorporate principles of data justice into the context of data for sustainable development, including issues of inclusiveness, data disaggregation and more.
3. Support fair data economies and responsible governance. Government actions impact public trust in the data economy. Governments strengthen trust by establishing rules and digital rights protections that give citizens control of their personal data. Governments erode trust when they take negative or repressive stances toward digital spaces, reduce individuals’ digital rights or limit freedom of expression online, as is currently happening in a number of countries around the world. The development community should monitor and highlight government abuses while sustaining country-level efforts to establish fair, rules-based data economies. 
4. Put people at the heart of data design. Designing data systems for people and empowering citizens to control their personal data are two essential steps to addressing data inequities. People need access to the design-stage process of data collecting, analysis, dissemination and use. And, they need the skills, knowledge and agency to determine how their personal data is used. This requires investing in research, alternative models and creative experimentation to empower historically excluded communities to develop and implement local solutions for their digital future.
5. Level up technical capacity and data practices in development. The development and humanitarian sectors must step up their game when it comes to data practices. It’s time for data awareness to be mainstreamed within these sectors. Donors also play an important role: They should take a hard look at their own policies and practices around data and digitalization to assess the roles they play in structuring incentives. Donors should take the lead in discussing best practices, sharing lessons learned and rewarding good data approaches.

We need more conversations that break down the sectoral and ideological divisions between digital and human rights activists and development practitioners. We are two sides of the same coin and as such, must work together to ensure equitable outcomes and create accountable and transparent systems. It’s our responsibility to put power back into the hands of the people whose data is being used. 

• A full description of this online conversation is available here. The online conversation, Dissecting Digital Power Inequalities, was recorded live on June 22. Tom Orrell, founder and managing director of DataReady, organized and moderated the discussion. His blog post on the personal experiences that led to him organizing this event is available here. The session’s participants included human rights lawyer Renata Avila, digital rights advocate and consultant Linda Raftree, and surveillance and development scholar Linda Raftree, and Paradigm Initiative founder ‘Gbenga Sesan. For a detailed description of themes emerging from the event, please click here. Tom Orrell of Data Ready, Beverly Hatcher-Mbu of Development Gateway, and Martina Barbero and Janet McLaren of the Global Partnership for Sustainable Development Data contributed to this post.

We’re emphatic in data for development about increasing visibility through data. So many people are missing from data that it’s almost an automatic response to seek to fill these gaps. But data deficits are a narrow slice of a much larger issue, and data itself is more than just a means of preventing people falling through the cracks. 

Who should decide what we measure?

The push for greater data visibility can be a double-edged sword: Not only are people often harmed by their inclusion in data, but there are many people who dispute the way they are defined and represented in existing data. 

As an Indigenous data advocate and member of the Ktunaxa Nation, Gwen Phillips explained that her community’s power of self-determination directly relates to data. When the Canadian government collects data on the Ktunaxa people, they “report on how sick and ugly and dysfunctional Indigenous people are” instead of on the community’s assets, strengths, and abilities. The Ktunaxa have their own set of metrics to measure community wellbeing that are different from those of mainstream Canadians. 

“As long as others are controlling the agenda, data, and investments, we’re always going to be subject to being beggars in our homeland,” Gwen explained. Having control of defining concepts results in different measures and outcomes. “What you measure is what matters,” Gwen says, “and if it's not what matters, then why measure it?”

In the context of the Data Values Project, defining agency—what it means, who should have it, and how it should be promoted and expanded—has become a key focus within each of the three thematic tracks.

Questioning the good news narrative in data for development

This idea of agency has emerged as fundamental to the Data Values Project in part because it’s largely missing from current development discourse. In the Global North, and predominantly in Europe, the concept of data agency is focused on controlling your personal data and being able to choose when and with whom to share it. 

While important, this concept doesn’t address the idea that shaping whether and how you are counted matters or that being involved in that process can lead to greater equity in society.

In the modern history of counting people, the field of statistics has its origins in the basic functions of the state. Measuring people, land, and possessions became a key source of power for governments, especially for European nations seeking to subjugate and control colonized people from afar. 

These days data for development is entrenched in the good news data narrative. We must know the number and health of people in rural areas in order to deliver life-saving medicine just as we can’t solve hunger without quantifying the causes of food insecurity. In this context, collecting more and better data becomes key to solving the world’s most pressing problems—as it should. But conversations in the Data Values Project have prompted us to expand this narrative.

The pursuit of equity is not just about counting people, making them more visible through data, and listening to diverse perspectives. It’s also about giving people control to define how they are measured and how their data is used.

By acknowledging and overturning the top-down origins of counting people, is it possible to make data a tool of self-determination and a means of restructuring current power imbalances, especially among historically marginalized places and people around the world? 

Looking ahead

There are plenty of models of inclusion that expand people’s agency and create better outcomes. A UK research agency this week released a roadmap for participatory data stewardship that lays out ways for people to gain control over their data, “from being informed about what is happening to data about themselves, through to being empowered to take responsibility for exercising and actively managing decisions about data governance.” The CARE Principles for Indigenous Data Governance are another example.

Within the Data Values Project, we’re starting to see agency as more than an issue of individual ownership and as a goal of effective data governance that can lead to more equitable outcomes. Instead of answering philosophical questions of what agency looks like in diverse contexts, members of the Data Values Project are struggling with the more difficult task of defining agency, what it means, where it sits, and why it matters—in addition to identifying approaches that foster and expand agency over data. 

• Jenna Slotin is Senior Director of Policy at the Global Partnership for Sustainable Development Data. She’s on Twitter @jslotin. Janet McLaren is Consultant for Data Policy at the Global Partnership and on Twitter @janetlmclaren. 

This article is also published here on August 31, 2021 by Data-Pop Alliance. It's the second edition of the Gender Data Series, which comprises interviews, videos, and opinion articles from experts to bring light to issues at the intersection of gender and data including health, migration, livelihoods, and gender-based violence (GBV). A related webinar, organized by UNU-IIGH, the University of Cape Town, and BBC Media Action, is available here.

Lockdowns, quarantine, and travel restrictions have been crucial tools in limiting the spread of COVID-19 globally. However, these policies have unintentionally created a ‘shadow pandemic’ of gender-based violence (GBV)—including increases in violence against women and girls. Existing efforts to tackle these forms of violence may even have regressed during COVID-19, requiring countries to explore new and innovative solutions to this important issue.

In countries around the world, healthcare systems and societies have witnessed a ‘quarantine paradox’ whereby policies to keep people safe have also led to serious harm for some. The pandemic has led to economic uncertainty and increases in unemployment, as well as to the reinforcement of existing harmful gender norms and government restrictions that have prevented women from leaving their homes. These, combined with the deprioritization of life-saving support—including clinical management of rape and intimate partner violence, reproductive health and shelter services—have cost lives. Around the world, COVID-19 has restricted women’s access to healthcare even as the need has increased. In some countries, calls to GBV hotlines, mobile-apps, and use of low-tech alert and support services have risen significantly.

Combating the shadow pandemic requires a whole-of-government and whole-of-society response. Technology and innovation also have a key role to play. With COVID-19 driving increases in digital transformation and digital service delivery, these tools have also been used to assist in the prevention of and response to GBV, including reporting and supporting services for survivors. These tools are likely to continue to be an important resource for women, policymakers, and support agencies after the pandemic. They also play an important role when coupled with face-to-face and low-tech services to reach the 390 million unconnected women in low- and middle-income countries. However, we need to ensure that these tools are used effectively and inclusively. To do that, it’s important to first understand more about governmental mitigation approaches to GBV during the pandemic.

How pandemic policies have tried to tackle gender-based violence (GBV)

Governments around the world have had to quickly identify, expand, and implement policies and approaches in the context of COVID-19. The COVID-19 Global Gender Response Tracker compiled policy measures that have been planned and/or implemented by governments worldwide in response to the COVID-19 pandemic. Globally, the top three interventions to tackle violence against women during the pandemic—beyond financial support—have been the expansion of helplines (37 percent of countries) and running campaigns to promote support service availability (29 percent) and to raise awareness about GBV (26 percent). 

                    Chart data from UNDP and UN Women's COVID-19 Global Gender Response Tracker.

Some of these interventions rely on digital technologies, such as the use of social media and other platforms to highlight services, raise awareness, and enable survivors to report instances of violence. The most common digital channel is mobile phones, which have been used in nearly 70 percent of digital responses to GBV during the pandemic, globally.

                    Chart source: O’Donnell, 2021 Gender and Digital Health webinar.

However, technology solutions do not work for all survivors, even those with digital access and high digital literacy. The conditions required to access digital services preclude many women, particularly those whose devices are shared with or monitored by abusers, and others who do not have privacy in their homes, to reach out to report and support services. Women and girls living with disabilities and in refugee camps are overrepresented in this group. 

How data can drive responses to GBV

Increases in usage of digital services have also resulted in data that can be used to drive responses and decision making. A study from Data-Pop Alliance analyzed administrative and non-traditional sources of data from México City, Bogotá, and São Paulo to better inform public policy around violence against women at home. Acknowledging the rise in cases since the start of the pandemic, the study explored human “capabilities” (as defined by Amartya Sen) to achieve well-being, in this  case, as expressed through actions of seeking help and reporting experiences of violence. Through analyzing data from the “Línea Mujeres” helpline, police reports, and 911 emergency calls, the team developed a model that explored the impact of demographic and contextual risk factors to report violence against women and girls. 

Analyses of these results show that the data sources considered reflected biases related to capabilities and opportunities to report GBV. The following factors were associated with greater reporting and registering among GBV survivors: age (30-49 years old in Mexico City and São Paulo, 20-40 in Bogotá); marital status (being single in Mexico and Bogotá); access (proximity) to support services for victims (Mexico City, São Paulo); people cohabitating in a living space (less than three people sharing a room in São Paulo), and greater mobility (less restrictions).

Figure shows Data Pop-Alliance study authors' application of Amartya Sen’s Capabilities Model to the concept of reporting capabilities. It suggests that women depend on “functionings” and the “context” or “environment” to report incidents of violence.

Additionally, this study revealed important gaps in GBV reporting (such as lack of availability to data sources, different and unclear conceptualizations of GBV, and data standardization issues that limit comparability of datasets, among others) and other crisis data, particularly during disasters and emergencies. So, how do we improve collection, management, and analysis of GBV data, especially to drive effective policy making, service delivery, and response? The Data-Pop Alliance study highlights four recommendations. In particular, we need to improve:

1. Access to data: improve transparency of data for responsible sharing.
2. Coverage and data disaggregation: increase spatial and temporal granularity of data to allow time and geographical analysis.
3. The quality of data: include more demographic information including gender and several indicators to improve our understanding and conceptualization of GBV.
4. Data governance: adopt further ethical guidelines and promote the use of data to inform policies.

The role of Big Tech 

In order to best support women experiencing GBV, efforts have focused on platforms familiar to many women (reaffirming several of the Principles for Digital Development). Twitter was one social media platform at the forefront of information dissemination during the pandemic. The organisation launched a dedicated gender-based violence search prompt as part of its #ThereIsHelp campaign to connect users with local hotlines and support. This has now been deployed in 27 countries and 30 languages in collaboration with local NGOs, governmental agencies, and UN Women. This approach has been particularly effective in Thailand where the initiative has connected GBV survivors with local law enforcement, government agencies, and NGOs who can provide longer-term services and assistance. 

These solutions are not a panacea. They rely on women with suitable devices and  access to them as well as the digital skills and confidence to use them effectively. In the context of emerging digital tools and rising GBV cases, it is important to focus on those affected by the gender digital divide. More digital efforts aimed at GBV prevention are also needed, in addition to further knowledge about the impact of policies and digital interventions on GBV during the pandemic.  

Work conducted in South Africa, for example, highlights the complexity of using data generated in crisis to drive sustainable response. In this case, the use of different digital platforms seemed to have contributed to misestimating the cases of GBV. While police data in South Africa showed a reduction in GBV-related calls, this may have been due to reallocated police attention away from domestic cases to other GBV initiatives available (thereby spreading data across a number of different services) and movement restrictions. This nuance reaffirms the importance of exploring the detail and context, through approaches such as those used by the Data-Pop Alliance.

That said, the global efforts of Twitter and other Big Tech organizations play an increasingly central role in an ever-growing toolkit to tackle GBV and other forms of violence against women. This role is likely to increase as these platforms continue to grow.

Learning from the past to design future innovations 

As policymakers continue to design digital interventions for GBV, it is crucial that lessons and better practices from previous crises inform future response. The  pathways to violence, for example, against women in prior emergency situations provides a means to streamline the process of creating effective solutions in the COVID-19 context. 

Similarly, this historical context can also be used to inform priorities in policies addressing GBV, and yet, the importance of this work is not always recognized. For example, an analysis of the UNDP and UN Women Gender Response Tracker shows less than 40 percent of included policies were gender sensitive. This is crucial in addressing the structural factors contributing to GBV and other forms of violence against women. The analysis highlights a particularly stark need for inclusion of gender-sensitive policies developed in lower-income countries.

However, according to a systematic review of literature on violence against women and children during the COVID-19 pandemic, challenges with data and reporting also exist:

• Most studies find at least some increase in violence against women and/or children. Findings apply across different demographics and contexts.  
• Where mixed or decreasing trends appear, evidence suggests underreporting may, at least in part, account for these results.
• Some studies examine risk factors and other dynamics around violence against women and children in the COVID-19 context, but few studies examine the impact of policies and interventions (e.g., helpline campaign in Mexico and youth empowerment programme in Bolivia).

Designing better systems to protect women from GBV

Globally, innovative measures to combat GBV and other forms of violence against women were sparked or accelerated during the pandemic. Nonetheless, gaps in understanding of the magnitude of this problem and best practices to protect women and girls have remained both during and beyond this ‘shadow pandemic’.

One fundamental theme emerging from this period is the value of digital platforms in connecting victims of GBV to supportive services through messaging and outreach, making sure policymakers recognize the existence of these events, and assuring that access to GBV-related support services increases. Nonetheless, it is crucial that responses to GBV critically evaluate digital platforms, resulting data, and even the conceptualizations of GBV in a pandemic- and digital-first setting. 

Such evaluation must also focus on identifying solutions to the gender digital divide and other factors that constrain individuals in being able to report and access support for survivors. The nexus of digital technology and GBV support has great potential, but it is imperative that implementation is driven by sound data and theory, particularly regarding closing existing gaps.

• Cláudia Abreu Lopes (@cabreulopes) is a Research Fellow and Calum Handforth (@calumAH) is a Digital Health Consultant at United Nations University ​​International Institute for Global Health (UNU-IIGH), a think tank that incorporates a gender lens in its policy-relevant analysis to inform the development, implementation, and evaluation of health programs.The views expressed in this post are those of the authors and author and may not reflect those of UNU-IIGH.