| dc.description.abstract | This dissertation investigates, in a causal way, how interactions of student with teachers and peers affect his or her educational outcome.
First, we use random assignment of students to Korean middle school classrooms and show that female students perform substantially better on standardized tests when assigned to female teachers; there is little effect on male students. We find evidence that teacher behavior drives the increase in female student achievement.
Also, we shed light on the importance of teacher student gender matches in closing the gender gap, especially in STEM fields in the long run. We exploit data from middle schools in Seoul, South Korea, where students are randomly assigned to a middle school and where students and teachers are randomly assigned to a physical classroom. Our finding is that female students taught by a female versus a male teacher keep achieving higher scores in standardized tests compared to male students even four years after the exposure to the teacher. We also find that if female students learn math from female teacher in seventh grade, then the likelihood increases that they take higher-level math courses and aspire to a STEM degree in their 11th grade. We show the evidence that the long lasting gender gap effects are driven by student’s behavioral change.
Lastly, we examine classroom peer effects on BMI. In response to increasing child obesity, many researchers have studied the sources of obesity, with social scientists focusing on peer effects. However, three well-known challenges make it difficult to find peer effects. We avoid self-selection using random assignment of classroom peers. To address common environmental factors and reflection problem, we instrument for peer BMI with number of peer siblings. We find that if peer BMI increases by one unit, student’s own BMI increases by 0.83 units and that the reduced social outdoor activities drive the effect. | en |
