Impact of short school closures (1-5 days) on overall academic performance of schools in California

Abstract

Climate change impacts such as disasters and higher temperatures can disrupt academic learning and reduce academic performance. Here, we use two-ways linear fixed effects regression to estimate the effects of short-term school closures (1-5 days) due to wildfires, natural hazard impacts, infrastructure, and student safety on academic performance in California, focusing on mathematics and English scores from state assessments and college preparatory exams. Wildfires are responsible for the majority of school closures. Wildfires generate significant negative impacts on academic performance among younger students. We primarily find insignificant impacts on academic achievement due to school closures from other causes, including from the interaction between number of closure days and socioeconomic and racial/ethnic makeup of the school, across all causes. The effects of school closures lasting more than one week (6-10 days) are also generally insignificant, except for the negative impacts of wildfire closures on elementary school students. These results suggest that older students are resilient to most unexpected short-term school closures (1-5 days) or that teachers can make up lessons effectively after schoolwide closures.

Figure 1

School closure days by cause. (a) Total school closure days across academic years ending in 2003–2019. (b) Histograms of frequency of school closure days by length of closures.

Figure 2

Effect of school closures on standardized test scores by closure cause for (A) grades 3–11 and all grades (grade 13) for mathematics, (B) grades 3–11 and all grades (grade 13) for English, and (C) college standardized exams. Lines indicate confidence intervals two standard deviations from the mean. Note that the y-axis scale varies among plots.

Figure 3

Effect of school closures caused by the interaction of closures and FRPM rates by closure cause on standardized test scores for (a) grades 3–11 and all grades (grade 13) for mathematics, (b) grades 3–11 and all grades (grade 13) for English, and (c) college standardized exams. Lines indicate confidence intervals two standard deviations from the mean. Note that the y-axis scale varies among plots.