UC Berkeley

This dissertation contributes timely evidence to the debate surrounding which policies may be most effective at raising college- and career- readiness in the United States. While over 75% of people in the U.S. graduate from high school, less than 25% of the population possesses a college degree. Over 60% of college students are required to take remedial coursework, which does not count for college credit. Using the National Longitudinal Survey of Youth 1997 Geocode, we provide a portrait of college readiness in the United States and discuss how it has been affected by popular high school accountability policies. We then examine the returns to education for students who have been impacted by these policies. Part I discusses different measures of college readiness, comparing their predictive power for success in college. Unlike previous research, our data allow us to examine curricular measures of high school and college readiness in addition to binary measures of educational attainment. After an exploratory analysis of these measures, we examine trends in college readiness over time in the U.S., detailed further within various socioeconomic groups and geographic regions. These college readiness measures are shown to provide information that is distinct from high school graduation rates when predicting college remediation and college graduation. Part II focuses on two types of high school accountability policies that emphasize basic proficiency: high school exit exams and consequential accountability. We develop a simple framework of high school behavior to predict how high schools will target their resources after these policies are implemented. We test the predictions of this model for college readiness. Our measures of college readiness are: the difficulty of students' hardest high school math course, high school completion, attending college without taking remedial math (college preparation), and college graduation by age 23. Identification is based on variation in the timing of state policies, controlling for state-invariant factors, time-varying national shocks, and regional time trends. We conclude that both policies decrease high school coursework difficulty and college preparation, but consequential accountability also decreases rates of high school completion and college graduation by age 23. These trends are not due to higher rates of GED attainment or lower college attendance, which suggests that these accountability policies lowered college readiness of college-goers. The last part of this dissertation addresses longer-term outcomes for students by examining how educational attainment and earnings have been impacted by these school accountability policies. After documenting an overall decrease in educational attainment, we find evidence of heterogeneous impacts by student ability prior to high school and school resources that imply increasing inequality in educational attainment across schools. Then, a two-stage least squares approach is used to measure the returns to schooling for students whose educational attainment was impacted by these policies. We find an earnings return to education of about 10% per completed grade, which is in line with previous literature using education policies as instruments for schooling.