Many education systems around the world use a centralized admission process to assign students to schools (Pathak & Sethuraman, 2011; Lucas & Mbiti, 2012). However, applicants to oversubscribed schools are not always assigned to their most-preferred school. Thus, one naturally asks whether it makes a difference to applicants’ subsequent educational outcomes if they are offered a place in, and if they enroll in, their top-choice school. In Mexico City, each year about 300,000 teenagers apply for a seat at one of the nearly 650 public upper-secondary school options. The centralized admission process in Mexico City, where applicants are assigned based on the interplay between their score on the admission examination and their ranked list of school choices, provides a natural experiment that permits these questions to be addressed in a case study of one school system. Thus, in my thesis,capitalize on the natural experiment created at each oversubscribed upper-secondary school in Mexico City, by the imposition of exogenous admission cut-off scores. Using rich administrative data provided by the school system for the 2005-2009 application cohorts, I use a sharp regression-discontinuity design (RDD) to estimate the causal effect of the offer of admission to a first-choice option on the probability that an applicant, at the cut-off, graduates on-time subsequently from upper-secondary education. Since some admitted applicants never register at the schools where they were offered admission, I then use a follow-up instrumental-variables estimation strategy (fuzzy RDD) to estimate the causal effect of attendance at a first-choice school on the same outcome.