We primarily focus on the effect of the maximum temperature on student performance. While mean temperatures provide a general idea about the climate, focusing on maximum temperatures allows for a more detailed understanding of the challenges posed by extreme weather conditions. Understanding the impact of these extremes, here particularly heat, offers insights into how individuals respond to these extremes. 5.3 Empirical strategy Our identification strategy relies on variation in students’ exposure to temperature at different municipalities over time. This variation is exogenous, as students’ assessments are scheduled in advance and deviations from this schedule are implausible.5 In addition, temperature exposure at school at the time of the test is unavoidable for students. Therefore, this offers a context to estimate the causal impact of temperature exposure on student performance. We exploit the variation in students’ temperature exposure with the following empirical model: Yipds =β1TEMPds +W′dsβ+αi +γy(d) +σps +ωy(m) +εipds (5.1) where the performance (Yipds) of student i in grade-test periodpatdate 5We see in our data that schools always test in the same weeks of the school year. The time between two mid-term tests is on average around 52 weeks. 123
RkJQdWJsaXNoZXIy MjY0ODMw