The COVID-19 pandemic suddenly and quickly forced changes to virtually everyone on the planet. Understanding these changes is of the utmost importance so that governments, policy makers, and other stakeholders can help those affected to cope with the negative consequences of the pandemic. However, in many parts of the world, including in Sub-Saharan Africa, the spread of COVID-19 and the related lockdown and social distancing measures have severely constrained the ability of national statistical authorities to conduct face-to-face surveys of households, individuals, and firms. But, in spite of these challenges, it is imperative to know:
How are people coping with the pandemic and how they coping with other difficulties in their lives? How do we answer these questions when we cannot go and ask people such questions face-to-face?
While our research is motivated by the desire to evaluate and understand the outcomes and impacts of the pandemic on individuals and households in Sub-Saharan Africa, its origin is rooted in the unique data analysis. As the pandemic hit and researchers around the world realized that face-to-face interviews were not – and likely would not be for some time – a possibility, an opportunity was recognized to continue data collection, in spite of the circumstances of the pandemic. The Living Standards Measurement Study (LSMS) from the World Bank is a household survey program which focuses on strengthening household survey systems around the world, with the objective of improving the quality of data to better inform policy. With an existing survey structure in place, including telephone numbers of individuals interviewed in the most recent round of face-to-face surveys, the LSMS team acted on this opportunity. The team identified that households could be contacted via phone, allowing for interviews of households during the pandemic without face-to-face contact. Further, by interviewing these particular households, there would be an opportunity to connect the pandemic survey responses to information from before the pandemic began, as households had been interviewed previously.
Our paper leverages the early rounds of the phone surveys in Ethiopia, Malawi, Nigeria and Uganda. These were the four countries that were the first to start data collection in the LSMS phone survey program. But, the program continues and has expanded: the World Bank LSMS supports national monthly phone surveys in Burkina Faso, Mali and Tanzania, and since April 2020, nearly 90,000 interviews have been completed across these seven countries. The anonymized unit-record survey data are made publicly available on the World Bank Microdata Library, and we have made available processed data and syntax files that can be used to replicate our findings and to inform downstream research by others.
By working to track responses to and socio-economic impacts of COVID-19, these surveys provide unique evidence on how households in Sub-Saharan Africa are coping with the pandemic and other associated events. It is these data that underlie and motivate our paper and that are feeding into monitoring of COVID-19 impacts globally. Although evidence suggests that the pandemic has not hit much of Africa as in other parts of the world, the effects of the pandemic are still felt by millions. The LSMS high frequency phone surveys give insight into the evolving impacts – and will to provide updated information in the months to come, as households continue to be surveyed and the data are made publicly available.
Beyond examining the impacts of the pandemic, the high-frequency phone surveys on COVID-19 and what we learn from the process of their distribution provide insights on how phone surveys, more generally, can transform to be a mainstay in the toolkit of national statistical authorities in low-income countries. There are active research efforts, including by the LSMS and partners, that are working to provide cross-country syntheses of the practical experience with the phone surveys on COVID-19. These will be instructive for designing and implementing phone surveys in these settings in the post-pandemic era. Phone surveys that can be rolled out rapidly in response to large-scale shocks or those that can be part of longitudinal household surveys that can implement high-frequency phone interviews in between face-to-face encounters are likely to have a greater role in post-pandemic research. The latter design could help reduce recall errors that are common in face-to-face surveys and allow us to better capture seasonal changes in development outcomes.