Three applications of matching estimation in applied microeconomics

In this dissertation, I examine two decisions that adults make that affect their children's future educational and labor market outcomes. The educational outcomes examined are highest grade completed, highest degree earned, and grade point average of the student. The future labor market outcomes examined are wages, time spent out of the labor force, and time spent unemployed. I also examine how a performance-based coaching change made by a university that participates in National Collegiate Athletic Association (NCAA) Division I basketball affects the future success of the program. Over the course of three essays, I utilize the theoretical background and empirical methodology that is found in contemporary labor and sports economics in order to find links between these parental and employment decisions and possible outcomes. The first chapter draws on data from the 1979 cohort of the National Longitudinal Survey of Youth, while the second chapter draws on data from the 1997 cohort of the same survey. The third chapter relies on data that I collected from various sources.;The first chapter of this dissertation examines the short-term and long-term effects of parental problem-drinking on children's future educational and labor market outcomes. Results indicate that having a problem-drinking parent has negative consequences for children's education in the form of lower grades and less schooling. It is also associated with lower wages, longer periods of being out of the labor force, and longer spells of unemployment. The second chapter examines the relationship between varying, large sibling age gaps on the future educational and labor market outcomes of youngest children. Results differ by gender. A positive academic effect is seen for males who only have one sibling, when that sibling is between three and five years older. The opposite is true for females; positive effects are seen in the larger age gaps when the respondents have two older siblings. The third and final chapter examines whether a team in NCAA Division I basketball that replaces their coach for performance-based reasons experiences an increase in win percentage, future NCAA tournament appearances, and a performance measure called the Simple Rating System (SRS) in the first three years of the coaching tenure. Similar teams are divided into groups based on how likely they are to replace their coach. Teams that were most likely to replace their coach saw no signicant increase or decrease in future performance following the coaching replacement. Teams that were moderately likely to replace their coach saw modest gains in future performance resulting from the coaching changes, while teams that were least likely to replace their coach saw a fall in team performance after the change. Each essay uses a matching technique that is not commonly found in previous literature, and helps provide a better understanding of the analysis done in this dissertation. This particular matching technique is also detailed in the Appendix.