Three Essays on the Economics of Education in Upper Secondary Schooling

Abstract

The thesis is a collection of three papers on the economics of education. All three use administrative data from Norwegian upper secondary education. The papers all take advantage of quasi-experimental variation to assess how school outcomes are affected by various factors.

The first paper is co-authored with Prof. Bjarne Strøm, who is also the main supervisor for the thesis. The paper addresses the question of how high school students’ schooling decision responds to improved labor market opportunities. We take advantage of a national reform of closing hours in the retail sector that differentially affected opening hours across municipalities depending on local regulations prior to the reform. The paper shows that the deregulation in the retail sector had a direct negative effect on the probability that students graduated high school.

While the first essay is devoted to exploring how the quantity of education is chosen by individuals, the two remaining essays are devoted to questions regarding how exam scores affected by disturbances and exam scheduling. Both the second and third essay are single authored.

The second paper provides evidence of the effects of pollen exposure during high-stakes examinations in Norwegian upper-secondary schools. I take advantage of within-student variation in pollen exposure across exams. I find that students experience significant and strong negative effects of pollen exposure ceteris paribus. The results are robust to a series of robustness tests, and I am also able to provide some evidence of the negative effects on placement in tertiary education. The paper demonstrates that natural occurring factors can cause serious noise in exam scores.

The third essay contributes with new evidence on the effect of exam scheduling. Taking advantage of the unique announcement system and allocation of students to exam-subjects in Norwegian high schools I am able to use random variation in the time students have to prepare for exams. Using this random variation, I show that the return to preparation time is concave and the marginal return drops to zero after a relatively short time period. I also show how relatively small variations in preparation time affects the outcomes of students in tertiary education.