UC Irvine

This dissertation addresses three timely issues in early childhood education (ECE) policy, focusing on how to build effective ECE programs at scale. I use eight large-scale longitudinal datasets at the state and national levels with causally informative methods to evaluate the peer, provider, and policy environments that jointly shape the development of children during early childhood and beyond. The first study provides experimental evidence on the effects of having peers who went to preschool (“preschool peers”) in one’s junior high school classroom on student cognitive and noncognitive outcomes in 7th and 8th grade. I use nationally representative data in China and exploit the random assignment of students to classrooms upon junior high school entry in this unique setting. I find that preschool peers significantly improve students’ educational expectation in 7th grade, which translates to improvements in academic performance and cognitive ability in 8th grade. An examination of mechanisms shows that a higher proportion of preschool peers is associated with increased after-school study time in 7th grade. The positive effects of preschool peers are more pronounced when there are more than 75% of preschool peers in the class and among students from disadvantaged backgrounds. The second study provides an up-to-date evaluation of the effects of competition on child care quality and 3rd grade academic achievements. Despite the sizable literature on the effects of competition in K-12 settings, little is known about the competitive effects on quality and child outcomes in the ECE context. Using administrative data of all center-based providers and their Quality Rating and Improvement System (QRIS) ratings in a large and diverse state, North Carolina, from 2005 to 2018, I find that competition is significantly associated with higher QRIS ratings and a higher probability to achieve a 5-star rating – the highest tier in the QRIS. More competition also reduces the time providers take to improve. Private providers significantly improve their quality when facing more competitive pressure, whereas public schools are less responsive. However, I do not find any effects of competition on district-level 3rd grade reading and math outcomes. The last study shifts to the macro-level interaction between the state pre-kindergarten (pre-k) programs and the federal Head Start (HS) program, comprehensively examining the impacts of state pre-k on HS enrollment and quality. I take advantage of the consistently collected administrative data of all HS programs from 1988 to 2018 and use a differences-in-differences approach, exploiting the rich variation in states’ timing of pre-k adoption during this period. I find that the introduction of state pre-k is associated with a statistically significant decrease in HS enrollment of Black children and a large increase in HS capacity to provide full-day 5-day services. It is worth noting that state pre-k did not significantly affect HS enrollment of other children (e.g., 3- or 4-year-olds, Hispanic children, families receiving social services from HS). Nor did state pre-k significantly affect HS staff and children turnover and the proportion of CDA teachers. Dynamic event study models reveal that the effects of pre-k were most salient in the first 10 to 15 years of the pre-k introduction and gradually diminished after that, which partially explains the non-statistically significant results on most outcomes. Overall, results from this study suggest that, while state pre-k has exerted some influences on HS enrollment and quality, HS may have managed to co-exist with the state pre-k programs and expand access to full-day services.