Covid County Data

Context

This data is generated by the Covid County Data project. In addition to a subset of the data being distributed below, we have a REST API and client libraries in Python/R/Julia. Please visit our website for more data or to make recommendations to us.

The COVID-19 pandemic has pushed many researchers and policy advisors to seek to understand and evaluate the costs and risks that policy decisions and proposals would incur. COVID-19 policy decisions will have lasting economic, political, and social consequences, and it is important that these decisions are rooted in reliable data.

The Covid County Data project is an initiative with the mandate to aggregate, collect, and distribute reliable data to anyone interested in evaluating these policies. The core values that drive our data collection are:

• Commitment to open sourced data: This is not proprietary data. We promote diversity of thought by freely and openly distributing the data that we collect and believe that this diversity of opinion is key to finding the answers to the questions that impact our communities.
• Collaborative research with an interdisciplinary team: While many policy questions could be answered by existing collaborations and labs, we believe the issues at hand are complex, and therefore, inherently interdisciplinary. In order to achieve this level of collaboration, data must be available to everyone, not only select institutions. Our data are already being used by interdisciplinary teams from economics, epidemiology, statistics, and the tech sector.
• Finding and implementing clean and quality data: Your research is only as strong as your data. All researchers face similar obstacles: data can be difficult to locate, difficult to aggregate, and difficult to validate. We aim to alleviate all three obstacles by digging deep into data sources, evaluating and organizing the information before adding the source to a centralized repository.

This project has been made possible through invaluable collaborations with researchers, consultants, and academics, and with the support of Schmidt Futures and Google Cloud.

Content

Our goal is to collect public health data at the county level. Currently, our data coverage is most complete for cases/deaths/tests.

Inspiration

We are including a few of our more popular datasets here, including our COVID data (cases/deaths/testing), but also mobility data and the non-pharmaceutical interventions. We think it is important (and interesting!) to explore the relationship between these datasets. In particular, it would be useful across the world to understand what types of non-pharmaceutical interventions are actually effective or how large of an effect does mobility have on spread, etc...