Show your love for Data Science Studies this Valentine’s Day!
Please join us in the Data Science Studio (6th Floor of the Physics and Astronomy Tower) on February 14th from 2:30-3:30 pm for a discussion seminar on the theme of “Data Science from the Bottom Up.”
A hallmark of data science is the promise of improving the aggregation, reuse, and repurposing of data at scale. But this means data scientists themselves are often several degrees removed from the context of data production, missing key insights into the motivations, complications, and negotiations of actors at the local level where data is produced and used. In this session, we turn our attention to those local contexts.
We’ll start the meeting with a presentation by Grégoire Lurton of Global Health and IHME, whose work interrogates the tension between the needs of local healthcare constituencies and their insertion in a Global Health infrastructure, often characterized by the imposition of global data standards from the top down. He develops statistical and data science methods that recenter the locus of control within local constituencies while still allowing for aggregation and comparison across contexts.
Below you’ll find a more detailed preview of his talk and some brief biographical information.
As usual, we’ll reserve a decent chunk of the hour for a group discussion following the presentation. Hope to see you there … you may even get some chocolate treats out of it!
Preview of Grégoire Lurton’s talk:
Global Health as a field relies on the articulation of a global and a local levels of knowledge production. The global level is the seat of international institutions and NGOs, academic structures, and funding agencies that work on addressing health questions framed as unified fields around the globe. The local level is the seat of national and sub-national governments, administrations, and civil society organizations that focus on implementing health interventions to tackle specific health issues. While the data needs and data cultures are different at each of these level, they utilize the same information infrastructure, and share data, tools and methods. The definition and development of these elements have long term impacts on the organization of national information systems and on health systems. In this regard, the diffusion of normed tools and methods defined at a global level for local utilization should be questioned for the ways in which they may disempower local institutions, communities, and cultures. Meanwhile, if the perverse effects of the use of quantified indicators in governance is often presented as inherent to the characteristics of quantitative work, there should be a place for statisticians and data scientists to offer an inner critique of the ethical and political dimensions of measurement, aimed at improving current practices.
Describing the organization of computational work in low resource health systems, the presentation will discuss the political and ethical implications of this organization, and will present work aimed at helping rebalancing the distribution of power in this technical system. A first example will discuss how using agent based models to simulate cohorts and data collection processes can help adapt measurement methods to local situations. A second example will show how data hybridization strategies can improve the usability of high resolution population maps.
Grégoire Lurton is a PhD Student in the Department of Global Health and a Research Associate at the Institute for Health Metrics and Evaluation. He graduated in Macroeconomics and Forecasting at the French National School for Statistics and Economic Administration (ENSAE ParisTech) and in Development Economics at Paris Institute of Political Studies (Sciences-Po). He has worked with NGO Solthis, working on the strengthening of country Health Information Systems for HIV programs in Guinea, Mali, Niger, Sierra Leone and Burundi and has short term work experiences in DRC, Burkina Faso and Tunisia, working for diverse organizations such as the World Bank, the French Development Agency or information systems start-up Bluesquare.