Three Essays On Social Determinants Of Students Skills PDF Download

This thesis comprises three essays on how social context moderates the impact of different shocks on kids' skills. In the first essay, I study the impact of one of the largest earthquakes and tsunami on students' primary and secondary school skills. After this event, the government reacted effectively to recover economic activity but overlooked the interventions to treat the trauma-related effects on children and their families. Using rich admin data, I find that the earthquake and the tsunami accounted for fewer days of formal education - because of damages in school infrastructure- and an increase of post-traumatic stress disorder prevalence among adults. Surprisingly, there is no impact on educational outcomes in primary school due to the earthquake or the tsunami. Children exposed in primary school do, however, show a negative effect on skills when they are in secondary school, and this effect is stronger for children exposed early in primary school. These results show that these events have a larger impact on children when exposed younger and that these negative effects surge during adolescence. This evidence is consistent with the prominent surge of mental health pathologies during adolescence. In the second essay, I study how migration-induced neighborhood changes can affect native test scores. Migration waves can change the composition of neighborhoods through immigrant arrivals and native relocations. There is literature analyzing immigrant effects on native students by studying peer effects in schools. By identifying the variation in the composition of classroom peers, such approaches can capture the impacts of the neighborhood composition only in part. In this paper, I compare the results of two different methods to analyze the impact of immigration on children's test scores and show broader changes in neighborhood effects indeed can be important. My paper exploits the recent migratory phenomenon in Chile, where from 2012 to 2019 immigrant population increased from near 1 % to 8 %. I estimate the neighborhood influence on native test scores following Chetty and Hendren's (2018) methodology. On average, I find a negative impact of foreign students on municipality effects. Then, I estimate the immigrant peer effect on native test scores. I find a precise null effect using two methods: a comparison across school cohorts and classes. These results show that immigration did not affect natives directly but rather through changes in the neighborhood. Exploring native composition changes, I find that immigration induces native flight and increases socio-economic segregation across schools. These results are consistent with migration changing neighborhoods by influencing a change in the composition of natives. The third essay, joint with Nicolas Navarrete and Claudio Allende, studies how students respond when off-platform universities participate in the centralized admission system in Chile. In 2011, eight new universities (G8) were incorporated into the platform used already by 25 universities (G25). In a difference-in-difference setting, we exploit G8 location and compare students who graduated from high school in a G8 city with those in non-G8 cities. Using administrative data on university, platform application and high school enrolment, we find that the inclusion of G8 universities increases student sorting. Women and students from lower backgrounds benefit the most, implying gains in equity and efficiency. These gains, however, do not remain in long-term outcomes such as graduation and enrolment after five years. Moreover, preferences for G25 universities decrease after the second year of the reform. We hypothesize that platform releasing cutoffs one year after the reform induces the change in listing preference, but heterogeneity results do not support this mechanism.