High temperatures have been demonstrated to have adverse effects on learning outcomes, but much of the existing evidence comes from high-income countries, primarily the US. While a few studies have explored the relationship between temperature and learning in Low- and Middle-Income Countries (LMICs) such as India and China, there is scant evidence from Sub-Saharan Africa, where the use of adaptive technologies is limited. This is a critical oversight. The impact of climate change on human capital may be one of its most consequential effects. Sub-Saharan Africa is one of the region’s most vulnerable to the damages of climate change and yet it is the region in which our understanding of the human capital dimensions of these damages is least understood.
The dearth of evidence largely stems from limited data availability. In this blog we discuss a new study that uses novel data from a high stakes exam in Ethiopia to examine the impact of high temperatures on student learning and performance.
The Ethiopian context
We analyze the test scores of 12th-grade students taking a national, standardized university entrance exam, known as the Ethiopian Higher Education Entrance Certificate Examination (EHEECE). This exam assesses knowledge of material learned during high school and determines high school graduation. It is also used to determine admission to university. In Ethiopia, individuals with a college degree have a significantly higher likelihood of securing paid employment, along with an 80% wage premium. Yet, until 2019, only about half of the students passed this exam, making it a “high-stakes” assessment. Therefore, any intervention or natural experiment which affects these test scores plays a pivotal role in determining the future outcomes of these students.
Primary Findings
To evaluate our research question, we combine administrative data of 2.47 million test takers of this exam from 2003-2019 and match it with the daily maximum temperatures that they were exposed to during the school year leading up to the exam.
Our results (Figure 1) suggest that an additional hot day (defined as a day when the temperature exceeds 33°C) during the school year is associated with a reduction in test scores by 0.009 standard deviation. This translates to a decline in performance by 2.28%.
Figure 1: Impact of Heat Exposure on Test Performance
Surprising Findings
In contrast to existing literature, which has not found strong evidence for differences by gender in the effects of temperature, our study reveals that female students' scores are less impacted by higher temperatures. Although the reasons for this disparity are not definitively established, we observe that female students tend to have fewer extended absences from school than their male counterparts, suggesting a higher level of academic commitment.
Consistent with our hypothesis of female students offsetting the negative impact of heat with greater effort, we find that the negative effects of temperature on all students are larger for easier subjects (Civics and Aptitude) as compared to more challenging subjects (English and Math). We believe this is because students exert higher levels of effort into more difficult subjects compared to easier ones and that this additional effort offsets some of the adverse effects of heat.
Climate Adaptation
Exploring the effect of regional variation in temperature, we found that students in schools located in hotter regions experience smaller declines in test scores compared to their counterparts who attend school in colder regions. This suggests that students in hotter climates are either acclimatizing to their climate or they are benefiting from some form of climate adaptation technologies at school. Due to the limited use of adaptive technologies in Ethiopia (a meager 8% of Ethiopian schools have fans), we were not able to directly test for heterogeneity in effects by schools which have cooling technologies such as fans or air conditioning.
Directions for Future Research
Our study confirms the negative correlation between high temperatures and learning in a new geographical context, yet it also highlights the need for further research on climate adaptation. As climate change leads to more unpredictable temperatures, it is crucial for future researchers to explore multiple climate adaptation interventions to determine the most effective strategies for creating optimal environments. The exact nature (e.g., installation of fans, adjustments to instructional schedules, changes to the way schools are constructed, etc.) of these adaptive interventions is a critical area for future research to address.
Join the Conversation