Marble is a science and art collective that tackles complex problems affecting children around the world. Marble brings together world-class scientists, strategists, and designers to leverage corporate data, scientific breakthroughs, and visual arts to accelerate impact for child rights.
The world’s most complex socio-economic problems affecting children - from migration, the spread of epidemics, to the needs of the new Sustainable Development Goals (SDGs) - cannot be tackled by a single organization alone. Collaborative solutions and shared value partnerships are critical to understand and unpack these problems, diversify risks, and make them actionable. At the heart of Marble’s approach is the merging of art and science for social impact.
Marble uses an ethical match-making methodology to repurpose a company’s privately-held data to decode real-life socio-economic problems affecting children. We then leverage expertise across academia to design and test scientific tools and solutions using that data. And we don’t stop there. Working with design thought leaders, we create powerful, scalable products and frameworks that turn evidence into action. Together, data and design can make the complex simple and impactful.
How do you tackle the world’s most challenging problems affecting children using data, science, and art?
How can we better understand the drivers of suicides and inform prevention through the lens of web data?
According to the World Health Organization (WHO), close to 800,000 people die by suicide every year, with 78% of global suicides occurring in low- and middle-income countries. Teenagers and young adolescents are particularly at risk, as suicide represents the second leading cause of death among 15-29-year-olds worldwide.
WHO estimates that good quality data on suicide exist for only 60 countries. So we partnered with Microsoft to see if web searches for specific keywords related to suicide could be used as a proxy of suicidal behavior. We focused on India, home to 20% of the world’s suicide deaths and the second largest number of internet users. We created robust models on suicides rates by extracting web queries on targeted keywords and combining it with official suicide rates and different kinds of demographic data. All queries were aggregated and anonymized. The published paper can be found here. We’re now expanding this work to all countries and many languages.
How can mobile phone data help make cities more gender-sensitive and inclusive?
As cities continue to expand, the poor have to travel greater distances to work, study, and live. Investigating the role of gender - and other socio-economic dynamics - in urban mobility is key to better understand whether women and young girls can fully benefit from the opportunities offered by cities, and in the process realize their human rights.
With funding from Data2X, we created a data collaborative to establish the first ever baseline of urban mobility experiences of women and girls within cities. Focusing on Santiago, Chile, we used anonymized and aggregated mobile records to successfully quantify gender disparities and other socio-demographics factors affecting the mobility of Santiago residents. We also identified the availability of transport options that are associated with mobility inequalities. We’re expanding this work to include other vulnerabilities in urban living (e.g. migration) and deepening the analysis by using data from internet phone usage (XDR).
Universidad del Desarrollo
How can social media data be used to better understand migration patterns?
Venezuela is going through the worst economic, political, and social crisis in its history. This situation is creating an unprecedented refugee and migrant crisis in the region. Governments and international agencies have not been able to consistently leverage reliable information using traditional methods. So we established two data collaboratives using anonymous and aggregate social media data to monitor the ongoing crisis.
In one data collaborative, we repurposed Facebook’s Advertising Data to measure and validate national and sub-national numbers of refugees and migrants and break down their socio-economic profiles to further understand the complexity of the phenomenon. In another case study, we are also exploring the value of using aggregated geolocated tweets to estimate migration flows from Venezuela, such as preferred routes, new settlement areas, integration in cities, etc. This can help authorities forecast and monitor migration flows and define when and where to intervene.
Qatar Computing Research Institute
New England Complex Systems Institute
MIT Media Lab
Institute for Cross-Disciplinary Physics and Complex Systems (IFISC)
How can we predict and prevent the next pandemic in low-resource settings?
More than 3 million children die each year due to preventable infectious diseases. And while infectious diseases see no borders, low-income countries bear most of the brunt in terms of socioeconomic impact. Here, we propose the creation of a framework for local healthcare infrastructures to monitor and predict the outbreak of emerging and re-emerging diseases in developing countries.
Through participatory surveillance, people volunteer to regularly report on disease symptoms using cell phones. This helps local governments capture early epidemiological signals of existing and emerging diseases. Working in partnerships, we would then create a forecasting framework to help predict how diseases will spread. This modeling framework will be based on the use of mobility data inferred from the use of cell phones through partnerships with mobile phone operators. The final goal is to generate an ecosystem of novel and cost-efficient digital surveillance infrastructures and a data-sharing platform for the effective surveillance, monitoring and control of circulating diseases, as well as preparedness when facing emerging disease outbreak.
How can remote sensing and agent-based model be used to forecast population movement in fragile contexts?
Approximately 60% of the Somali population engages in pastoralism, primarily relying on water and vegetation to support livestock and in turn, livelihoods. Conflict, recurrent droughts, and changing environmental factors are several of the forces shaping pastoral livelihoods and are frequent contributors to internal displacement. As of June 2018, approximately 2.6 million Somalis were internally displaced, and the numbers continue to grow.
As one of 12 grantees of the World Bank’s Collaborative Data Innovations for Sustainable Development, we’re developing an Agent-Based Model (ABM). An ABM captures the interactions between individual agents (pastoralists, in this case) and their surrounding environment across space and time, including data on vegetation, conflict events, ethnic divisions, and water availability as variables influencing migration. The integration of ABM and high resolution satellite imagery provides a simulated overview of population movement, which can be further refined with the integration of insight from the field. The model will contribute to our overall understanding of what is driving population movement in Somalia and can inform emergency response interventions to displaced population, identification of camps for targeted responses, and the protection of migrants, especially children on the move.
Harvard Humanitarian Initiative Signal Program
Data, Research, Policy Manager, UNICEF
Art Director, UNICEF
Design Strategist, UNICEF
Contact us if you want to collaborate and help us tackle the world’s most complex problems affecting children - or if you simply want to learn more about us.