CoronaNet: COVID-19 Government Response Event Dataset

Governments worldwide have implemented countless policies in response to the COVID-19 pandemic. We present an initial public release of a large hand-coded dataset of over 12,000 such policy announcements across more than 190 countries. The dataset is updated daily, with a 5-day lag for validity checking. We document policies across numerous dimensions, including the type of policy; national vs. sub-national enforcement; the specific human group and geographic region targeted by the policy; and the time frame within which each policy is implemented. We further analyze the dataset using a Bayesian measurement model which shows the quick acceleration of the adoption of costly policies across countries beginning in mid-March and continuing to the present. We believe that the data will be instrumental for helping policy makers and researchers assess, among other objectives, how effective different policies are in addressing the spread and health outcomes of COVID-19.

Citation: Cheng, Cindy, Joan Barceló, Allison Hartnett, Robert Kubinec, and Luca Messerschmidt. 2020. COVID-19 Government Response Event Dataset (CoronaNet v1.0). SocArXiv. April 12. doi:10.31235/

Download the working paper here.

Observed COVID-19 Test and Case Counts

As the COVID-19 outbreak increases, increasing numbers of researchers are examining how an array of factors either hurt or help the spread of the disease. Unfortunately, the majority of available data, primarily confirmed cases of COVID-19, are widely known to be biased indicators of the spread of the disease. In this paper I present a retrospective Bayesian model that is much simpler than epidemiological models of disease progression but is still able to identify the effect of covariates on the historical infection rate. I show that while observed cases and test counts cannot be used to measure the true infection rate due to confounding between the infection rate and the observed data, it is possible to use informative priors derived from susceptible-infected-recovered (SIR) models to estimate associations between background factors and the disease. The model is validated by comparing estimation of the count of infected to projections from expert surveys and extant disease forecasts. To apply the model, I show that as of April 7th, higher infection rates are increasingly concentrated in U.S. states with smaller economies, larger youth populations, less public health funding and fewer Trump voters. On the other hand, the percentage of foreign born residents, air quality, and the number of smokers and cardiovascular deaths are not clear predictors of COVID-19 trends.

Citation: Kubinec, Robert. 2020. “A Retrospective Bayesian Model for Measuring Covariate Effects on Observed COVID-19 Test and Case Counts.” SocArXiv. April 1. doi:10.31235/

Download the working paper here.