Heterogeneous relations between climatic and environmental shocks and schooling and learning in Sub-Saharan Africa

• In Event: The climate crisis and learning and development: some global perspectives

Proposal

Children in Sub-Saharan Africa (SSA) are exposed to multiple climatic and environmental shocks, including water scarcity, flooding, tropical cyclones, heat waves, vector-borne diseases, and environmental pollution. These stressors may hinder opportunities to learn through direct and indirect pathways. While SSA’s vulnerability to climatic and environmental stressors and the potential implications of these stressors for education are increasingly recognized, there is a dearth of evidence on precise effects or mechanisms of impacts on schooling and learning. This study investigates the relations between climatic and environmental shocks and schooling and learning in SSA via the following research questions: (1) To what extent do socio-economic and climatic factors, and their interactions, predict child educational outcomes? (2) How do variables such as gender, ethnicity, rural-urban residency, among others, moderate the effects of socio-economic and climatic factors on child educational outcomes? We use multivariate regressions to investigate socio-economic and climatic factors that, independently and interactively, predict (a) timely initial enrollment, (b) progression through school, and (c) learning. We examine this by analyzing UNICEF Multiple Indicator Cluster Surveys (MICS) data from 7 SSA countries, namely The Gambia, Lesotho, Madagascar, Malawi, Nigeria, Sierra Leone, and Zimbabwe, with a sample size of about 117,000. The sample includes children aged 7 to 14 years. Examples of climatic shocks include extreme heat, extreme cold, droughts, and flooding. More specifically, we plan to merge MICS children's information with climatic and environmental datasets. Potential climatic datasets include ERA5 and EM-DAT. ERA5 is a temperature dataset compiled by the European Centre for Medium-Range Weather Forecasts (ECMWF). EM-DAT: the Emergency Events Database is compiled by the Centre for Research on the Epidemiology of Disaster(CRED). It contains information about disasters that happen globally, including in SSA. Using approaches piloted by our team with MICS data from several Asian countries and based on a set of assumptions about location histories, we will create age-varying exposure measures for each child for the first 1,000 days from conception, the most recent years, and the time in between. To address Question (1), we will use marginal structural models (MSMs). To address Question (2), we expand upon our first model by including potential moderating variables such as gender, ethnicity, rural-urban residency, and child age on the average exposure effects on child outcomes. For some past exposure measures, we anticipate potential challenges associated with data on migration. Given the information available in the MICS data, we will develop as accurate data as possible on the lagged exposures. MICS recorded whether they have always lived at the current place, and if not, whether they lived in a city or a town. MICS also provides information on the previous first geographical division they lived in before moving to the current location, if there was any migration at all. Findings will contribute to academic and policy discussions about climatic-change impacts on children in various contexts in SSA. Finally, as an extension to Question 2, we will investigate sources of heterogeneity across countries by comparing individual, country-specific estimates of climatic factor impacts with pooled estimates.

Authors

• Yabo Vidogbena, University of Pennsylvania

• Amrit Thapa, University of Pennsylvania

• Wendy Chan

• Emily Hannum, University of Pennsylvania