Search
Browse By Day
Browse By Time
Browse By Person
Browse By Mini-Conference
Browse By Division
Browse By Session or Event Type
Browse Sessions by Fields of Interest
Browse Papers by Fields of Interest
Search Tips
Virtual Exhibit Hall
Change Preferences / Time Zone
Sign In
X (Twitter)
During the COVID-19 pandemic, many countries have issued stay-at-home orders (SAHOs) to reduce viral transmission. SAHOs are health policies that tightly restrict social interaction. In many cases during the current pandemic, this has meant that countries prohibit individuals from leaving their residences unless they are going to an “essential” job or seeking medical care or essential supplies like groceries. Although SAHOs are effective tools for stopping infectious disease transmission, they are a politically risky decision for governments because they also result in significant social and economic costs. For instance, in the case of the coronavirus pandemic, school closures have impacted over 60 percent of all students globally, with localized closures affecting millions more. Business lockdowns have caused major economic ripples across a wide range of countries, leaving millions of workers furloughed and leading to predictions that an additional 70-100 million people will slip into extreme poverty for now and into the foreseeable future. The health policy literature identifies five broad and theoretically consequential factors that should be significant influences on the issuance of SAHOs: economic, external, medical, political, and social. However, research exploring the relative importance of these factors regarding coronavirus-related SAHOs is limited. To test the effects of these factors, if any, we apply a random forest classifier to predict the issuance or not of SAHOs in 54 African countries during the first wave of COVID-19 in early 2020. Africa makes for a useful testing context because the continent is understudied, and the countries are highly heterogeneous. The analyses employ a novel dataset of more than 200 measures that includes variables motivated by theoretical expectations (i.e., economic, external, medical, political, and social factors) as well as a large, cross-domain selection of variables related to public health capacity and disease, political institutions, state fragility, population and social dynamics, the economy and development, poverty and inequality, energy, the environment, communications, transportation, military/security issues, and geography. The variables were collected from sources such as the World Bank, World Health Organization, Oxford COVID-19 Government Response Tracker, European Centre for Disease Prevention and Control, and Database of Political Institutions. A preliminary model generated using 10,000 simulations predicts the issuance of SAHOs in the sample with 93% accuracy using six variables. While theoretically motivated variables do appear
in this model, the model also identifies variables that are not clearly related to conventional theory as important for SAHO classification. This offers an opportunity to formulate and test new hypotheses related to public health policy and demonstrates the value of machine learning as a means to advance theory.