Facebook Launched its Covid-19 Symptoms map today
Mark Zuckerberg Said;
Today we’re releasing the first maps of the Covid-19 symptoms people have reported experiencing county-by-county across the US, based on aggregate data from the opt-in survey we’re running with researchers at Carnegie Mellon. We’ll update these maps daily through this outbreak.
Understanding how Covid-19 is spreading is critical for local governments and public health officials as they allocate scarce resources like ventilators and PPE, and eventually to decide when it is safe to start re-opening different places. Researchers believe these symptom survey maps can be an important tool in making these decisions.
Facebook is uniquely suited to run these surveys because we serve a global community of billions of people and can do statistically accurate sampling. We do this in a privacy protective way where only the researchers at Carnegie Mellon see individual survey responses — and Facebook only sees aggregated data.
The next step is to start running these surveys globally this week. This will allow us to expand the symptom maps to provide county-by-county data across almost every country in the world where Facebook operates.
We’re also looking for more research partners around the world to help generate insights from this aggregate data to help with the Covid-19 response. We’ll share more details on how to get involved soon.
Supporting COVID-19 Research
With over 2 billion people on Facebook, we are in a unique position to support COVID-19 research. We’re inviting people to participate in a survey conducted by Carnegie Mellon University Delphi Research Center, and over a million people responded to this survey within the first two weeks.
We designed this effort with privacy in mind from the start while providing new ways to help researchers and experts understand this crisis.
About this survey
The survey from CMU Delphi Research Center asks people to self-report symptoms associated with COVID-19 or the flu that they or anyone in their household has experienced in the last 24 hours. Surveys like this have been used globally for public health research. Even with as few as several thousand responses, survey data like these have been successfully used to forecast the spread of the flu and other illnesses.
Who’s taking this survey
Facebook reaches large segments of the population allowing for a significant representation of age, gender and state of residence. Every day, a new sample of Facebook users over 18 years old within the United States are invited to participate in the survey. Facebook doesn’t receive, collect or store individual survey responses, and CMU doesn’t learn who took the survey.
Adjusting for survey bias
To help CMU measure results, Facebook shares a single statistic known as a weight value that doesn’t identify a person but helps correct for any sample bias, adjusting for who responds to survey invitations. Making adjustments using weights ensures that the sample more accurately reflects the characteristics of the population represented in the data. The weight for a survey respondent in the sample can be thought of as the number of people in the population that they represent based on their age, gender and state of residence.
Using the data
The estimates can be helpful for policymakers and health researchers to forecast potential COVID-19 outbreaks. These estimates don’t represent confirmed COVID-19 cases and shouldn’t be used for diagnostic or treatment purposes, or guidance on personal or business travel. Facebook’s research partners are committed to only using survey results to study and help contain COVID-19.
INNOVATION FOR SOCIAL GOOD
Creating New Tools to Help
Facebook is supporting other research projects that help experts and policymakers more effectively plan for COVID-19 outbreaks.
As part of our Data for Good program, we’re offering maps on population movement that researchers and nonprofits are already using to understand the coronavirus crisis, using aggregated data to protect people’s privacy. In addition to popultation maps, we’re offering three types of Disease Prevention Maps to help inform disease forecasting efforts and protective measures.
We’re also using artificial intelligence to help experts more effectively plan. In collaboration with academic partners, we’re helping forecast COVID-19 spread at a county level in New Jersey and New York so local authorities can better anticipate demand for personal protective equipment, ventilators and other resources.
Check Map here: Facebook Map