In Chapter 1, I study the effect of school absenteeism on secondary school students academic outcomes using the Chilean student strikes in 2011 as a source of exogenous variation. The strikes, led by university students but promptly joined by hundreds of thousands of secondary school students, triggered a significant drop in public secondary school attendance (a decline of about 15 percentage points in all four grades). Attendance returned to normal levels in 2012. Using the type of school that students attended in 2011 as an instrument for school absenteeism, I show that school absenteeism has negative effects on secondary school students’ results in a postsecondary high-stakes math exam and university enrollment rates. Instrumental variables estimations suggest that a 10 percentage point decrease in attendance during secondary school is related to a 9.5 percent of a standard deviation decline in the math exam score, and a 3.2 percentage point reduction in the associated probability of university enrollment. I do not find any significant effect on the high-stakes language exam at the 5 percent level. A key finding is the persistent negative effect of school absenteeism on students’ academic performance: this negative effect is present even for those students who sat the high-stakes exams three years after the strikes had ended, that is, after three years of regular schooling following the negative shock to their attendance. These results are not driven by inputs to the education production function that might have been affected by the student strikes, such as disruptiveness at the time of the high-stakes exams, school environment, teachers, class instruction, or class size. Chapter 2 presents the first value-added (VA) estimates for doctoral teaching assistants (DTAs). We focus on the undergraduate program of the Economics Department at a UK university, where the match between students and DTAs is random. We find that a one standard deviation change in DTA quality increases students’ test scores by around 8.5 percent of a standard deviation. A novel feature of our data allows us to examine within-course dynamics in the VA estimates: These are larger for assessments taken during term-time, drop for end-of-term tests and are not statistically different from zero for final exams. The analysis suggests that the lack of persistence of the VA measures might be connected with: (i) Students’ endogenous investment responses and (ii) temporal decay in teacher-related human capital. We discuss how our results can inform the broader debate on the measurement of teachers quality via the VA approach. In Chapter 3, we study the effects of a penalty points system (PPS) introduced in Spain in 2006. We find a 20% decrease in cumulative road fatalities in the five years after the reform, compared to a synthetic control group constructed using a weighted average of other European countries. Evidence suggests that the persistent reduction in road fatalities might not only be driven by deterring risky-driving behavior, but also by taking reckless drivers out of the roads. Using estimates of the value of a statistical life, we calculate that the PPS yielded a net economic benefit of €4.6 billion ($6 billion) over this period, equivalent to 0.43% of Spain’s GDP.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:752468 |
| Date | January 2017 |
| Creators | Gaete Romeo, Gonzalo |
| Publisher | University of Warwick |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://wrap.warwick.ac.uk/106602/ |
Page generated in 0.0133 seconds