India, Floods, and Learning Outcomes

In the past 20 years, India has experienced an increase in floods and the devastations and disruptions caused by them. But how does that affect learning outcomes for school-aged children?

This is the key question addressed in new research from Nazar Khalid, a fifth-year joint PhD student in Demography and Sociology; Emily Hannum, Stanley I. Sheerr Term Professor in the Social Sciences and Associate Dean of Social Sciences; Jere Behrman, William R. Kenan Jr. Professor of Economics; and Amrit Thapa, Senior Lecturer in the Policy, Organizations, Leadership, and Systems Division in Penn’s Graduate School of Education.

Published in RSF: The Russell Sage Foundation Journal of the Social Sciences, the paper pays particular attention to children from marginalized groups and communities, using analysis of a two-wave panel dataset called the Indian Human Development Survey (IHDS), a nationally representative sampling of more than 41,000 households in 971 urban blocks and 1,503 villages.

The research found that children from some marginalized groups such as lower-caste Hindus, Muslims, and poorer children are more likely than children in India not in those groups to live in flood-prone areas. In addition, girls show greater disruptions to math learning than boys when exposed to floods.

“India is a highly stratified society, divided along demographics like caste, religion, and socioeconomic lines,” Khalid says. “It’s important to understand how those stratifications can disproportionately create further climate change harm.”

India is a highly stratified society, divided along demographics like caste, religion, and socioeconomic lines. It’s important to understand how those stratifications can disproportionately create further climate change harm.

Khalid, who is from the part of India the paper focuses on, says he became interested in exploring these research questions after noticing that climate change had caused floods there to increase in frequency and intensity, and to become more widespread. “Increased flooding during the monsoon season [in population-dense regions] affects a lot of people and their ability to access resources like healthcare and education.”

With such questions in hand, Khalid approached Hannum and Behrman to advise and participate. They share an interest in broad research questions about the effect of weather and pollution on child development and have an ongoing National Science Foundation–funded research collaboration focused on the detrimental effects of climatic hazards and pollution on children in low- and middle-income countries. Both were excited to engage in the research. Together with Thapa, they prepared proposals for the India Research and Engagement Fund at Penn Global and for Penn’s University Research Foundation.

Through fieldwork that interviewed educators and villagers and analysis of the IHDS, the researchers determined that, overall, floods were exerting negative impacts on learning outcomes and timely grade attainment. However, “across many different dimensions of stratification, marginalized children were affected more,” Khalid says.

In some cases, the disproportionate impact played out as exposure differences. “For example, our data show that poorer children, those from some lower-caste groups, and those from religious minorities are more likely to be in flood-prone communities and are more exposed to floods,” Khalid says. In other cases, the effect of floods on outcomes differs. For instance, Khalid says, “girls showed more negative math learning impacts when they were exposed to floods, and there is suggestive evidence of the same pattern for reading.”

Though the study does not probe why this latter fact is true, the researchers have some ideas. “India is a very patriarchal society and sons get more preference,” Khalid says. “So, if you think of flood exposure as bringing different types of shocks and thus, placing a lot of opportunity costs on time and many other factors, girls will likely be asked to help at home or stop going to school more than boys.”

Prior research has confirmed that, Behrman says. “If there’s a negative shock or negative anything, girls are most likely to be taken out of school or to be invested in less in terms of health. It works the other way too, that their situation also tends to become more positive when there are positive shocks.” Hannum notes that, consistent with this idea, research in rural China also has suggested that girls’ education is more sensitive than boys’ to economic deprivations and to shocks like closure of community schools.

This research adds new insight, according to Behrman, because previous work has mostly focused on preschool-aged children and, for school-aged children, their attendance as opposed to learning outcomes.

The team recently presented its findings at the Asian Development Bank Conference. The researchers say they hope their paper can help policymakers and the field better consider the interaction between social inequalities and climate change, as well as inspire potential ways to mitigate their negative effects. A second paper about whether attributes of communities mitigate risks of flood exposure is forthcoming in an Asian Development Bank working paper series and currently is under journal review.