The papers in this volume sought to answer why students enroll in for-profit colleges and universities and what factors influence their enrollment. It further sought to determine the degree to which current models of choice are applicable for these students in the U.S. and the degree to which models of college choice have become universal. Using a multinomial regression, paper 1 found that low-income or first generation status made one more likely to enroll in a for-profit institution. Paper 2 found a potential explanation for the findings from Paper 1; for-profits offer significantly more help and support in admissions than their traditional counterparts. Student surveys revealed help with the process as being important in enrollment decisions. Paper 3 duplicated Paper 1 but in the Brazilian context, analyzing the effects of a policy (ProUni) meant to increase enrollment by low-income students. The results were encouraging, finding an increased likelihood of enrollment by eligible students after implementation. More generally, these papers found that current models of college choice are largely inapplicable when dealing with students in for-profits and that although we may be moving towards a universal model of college choice, significant differences between systems still exist which make that untenable.

Despite criticisms on foreign aid, empirical studies show mixed results on aid effectiveness. Since past studies have limitations, as they used an aggregate country-level approach, recent studies began to focus on aid effectiveness in the individual sector. However, there are very few studies examining aid effectiveness in the education sector. Education plays a fundamental role in poverty alleviation and economic growth. Considering that country level approach has limitations in explaining the large variation in success and failures of development projects, this study examined the factors that are associated with education project outcomes implemented by the World Bank. This study used a mixed method design. Utilizing the World Bank project performance ratings data, I employed linear probability, logit, and ordinal logit models with country and year fixed effects. Interviews with the World Bank staff were also conducted to complement the limitations of the quantitative data. The findings suggested that the role of borrower governments (government effectiveness and commitment) is significant for the success of education projects, which supported Burnside and Dollarâs (1997) theory that aid works in a sound policy environment. Administrative factors such as project cost and duration were not significant, while project design and staff quality were found to be significant for project outcomes. In addition, this study suggested evidence that challenged the validity of the World Bank evaluation ratings data. The findings of this study call for a closer examination of the role of the borrower countryâs government in project implementation, and suggests that donors should consider ways to deliver aid effectively and efficiently rather than providing more amount of aid.

The following dissertation consists of three studies, demarcated as chapters, seeking to advance the use of longitudinal data in the measurement and evaluation of the dynamics K-12 teacher and school effectiveness, each using distinct quantitative methodologies. The first chapter employs a regression discontinuity design to evaluate the effects of the North Carolina Department of Public Instructionâs school turnaround program on student academic outcomes. Separate analyses by levels of schooling (i.e. elementary, middle, and secondary) and outcome measures reveal potential heterogeneity in the response to intervention and underscores the necessity of multifaceted approaches and nuanced reporting in the evaluation of broad, multiyear education reform initiatives. The second chapter analyzes the potential effects of performance-based teacher retention reform using the statistical framework of diagnostic accuracy to evaluate such policies in their ability to correctly predict future teacher performance. This approach offers new metrics for the consideration of retention reformâs consequences, both intended and unintended, and highlights the limitations natural teacher churn imposes on such reforms, as well as the potential benefits of using multiple measures of teacher performance to more accurately predict teachersâ future performance. The third chapter considers how student achievement gains are distributed within teachersâ classrooms as a way of comparing teacher performance and understanding teacher improvement. Specifically, it examines how the mean, variance, and skewness of the distribution of individual student learning gains change with teacher experience, across grades and subjects, and between teachers at varying levels of estimated effectiveness. Notable among the studyâs findings is that the distribution of student learning gains and the dynamics of teacher improvement demonstrate patterns for teachers of English-Language Arts (ELA) distinct from that of teachers of mathematics and science, with more experienced ELA teachers reducing the variance in their studentsâ learning gains in ways not reflected by analysis of mean learning gains over time. Taken together, the chapters take three distinct methodological approaches to longitudinal education data and span a range of research goals, from the exploration of the dynamics of how the distribution of individual student learning gains are distributed within classrooms, to the development of more efficient performance-based teacher retention measures, to the evaluation of the efficacy of a state-led school turnaround initiative. Within each, contextual knowledge of an unsettled policy debate is used to advance the measures and methods used to evaluate the dynamics of K-12 teacher and school effectiveness.

High payments and default on undergraduate debt have consequences. These high payments and defaults are national concerns if aversion to those consequences deters students from making optimal postbaccalaureate decisions regarding postbaccalaureate educational aspirations, enrollment, and early-career occupation. This work proposes to expand our general understanding of how undergraduate loans influence and potentially constrain the postbaccalaureate decision-making for students. I utilize a mixed methods approach to investigate the extent to which undergraduate student loan debt influences postbaccalaureate educational aspirations, enrollment and early-career occupational choices. For the quantitative analysis, I use instrumental variables to estimate the causal effect of undergraduate loans using two compiled datasets of either Beginning Postsecondary Students: 04/09 or Baccalaureate and Beyond: 08/12 merged with the Integrated Postsecondary Education Data Systems and the Barronâs Admissions Competitiveness Index. For the qualitative analysis, I conduct multiple semi-structured interviews with six graduating underrepresented students at a Historically Black College or University as they enter repayment on undergraduate loans. The quantitative results support the conceptual framework as a theory of behavior for studentsâ postbaccalaureate decision-making. I find that $10,000 increase in total undergraduate debt: does not appear to induce a change in studentsâ aspirations; decreases the likelihood of students enrolling in graduate school by 3-4%; and, increases the average annual salary of students by $1,550 in 2009 and $3,410 in 2012. A $10,000 increase in federal undergraduate debt: does not appear to induce a change in studentsâ aspirations; decreased the likelihood of students enrolling in graduate school by 5%; and, increases the average annual salary of students by $2,100 in 2009 and $4,620 in 2012. The qualitative results, while preliminary in nature due to the small sample size, suggest that following students past graduation is key as the students reported confidence or emotional state may change as repayment draws closer. Practical and policy implications for these findings are discussed.

English learners (Els), defined as students who are identified as needing support learning English in school, are a rapidly growing and underperforming segment of the public school population. Schools in new immigrant destinations, that is, areas in the United States that traditionally have not had a large immigrant population, are having to create new structures and policies to support and instruct Els. This study focuses on the schooling experiences of English learners in North Carolina, one new immigrant destination. Specifically, this study documents two classroom characteristics: the extent to which Els are segregated from native English speakers and whether Els have access to teachers who have an English as a Second Language teaching credential or experience teaching Els. Additionally, it estimates the relationship between these two classroom characteristics and achievement using OLS regression and regression with school and student fixed effects. Results include that: 1) Els are largely integrated with native English speakers; that is, they are typically enrolled in classes where the majority of their peers are native English speakers; 2) Less that 5% of Els are taught by an ESL credentialed teacher but over 50% of Els are taught by an El experienced teacher 3) There is little evidence that being assigned to a segregated class or an ESL credentialed teacher impacts math or reading achievement; and 4) There is evidence that being assigned an El experienced teacher has a positive impact on math and reading achievement. Implications from this study include that policy makers in new immigrant destinations should focus on strengthening the English as a Second Language credentialing process, providing more hand-on training for teachers of Els, helping Els reclassify as English proficient before they reach middle school, and supporting schools where Els comprise a small proportion of the population.

This paper is to identify the relationship between parental socioeconomic status, social capital within the family, social capital within a school, technological capital and students' scores for college entrance in South Korea. This study builds on and extends existing work in that it examines parental socioeconomic status and the effect of social capital, dividing social capital into two forms of social capital, and their relationship to college enrollment in Korea. This paper contributes to the empirical study of social capitals and technological capital on education by examining the relationships between those capitals and their college entrance in Korea.

This paper explore the choice provisions contained within the No Child Left Behind Act (NCLB). Examining the literature around student mobility in the context of school choice, the paper discusses the possible expectations for student achievement growth. NCLB choice is intended to be compensatory, giving students in low performing schools the opportunity to move to a higher performing school. However, the examination of mobility literature shows that the negative influence of mobility on student achievement may outweigh a compensatory effect that could be achieved by NCLB choice.

The majority of college students in the United States are ânontraditionalââonly 26% of currently enrolled students fit the criteria to be considered a traditional college student. Developmental education is another aspect of American higher education that is considered nontraditional, while being fairly prevalent. At two-year institutions, 52% of students take at least one developmental course, in addition to 20% at four-year institutions. Presented here are three papers that address these central issues, currently under-examined and frequently misunderstood. The first paper explores the college choice process for nontraditional students and identifies a new potential conceptual framework for examining that process among this particular group of students. The second paper looks at the impact of multiple life roles on postsecondary completion among nontraditional students, finding significant negative effects of additional life roles, and indicating that the negative impact is more severe for women than for men. The third and final paper is a meta-analysis of existing research on the impact of developmental education. It finds significant disagreement within the field, but very little indication as to why; it suggests new directions for research on developmental education that might help the field build consensus.

A substantial literature has emerged to describe teachersâ systematic labor patterns. A fundamental assumption of this literature has been that teacher turnover occurs between school years. I examine the tenability of this assumption using rich administrative data from North Carolina that enables me to measure teacher turnover not only as an annual event but as occurring at any month throughout the school year. Documenting the teacher turnover that occurs within school year allows for a more complete and accurate picture of how this instability occurs not just between school years, but during the school year as well. Each of the studies of this dissertation, demarcated as chapters, address a different element of within-year teacher turnover. In the first chapter, I describe the frequency with which within-year teacher turnover occurs and the types of teacher and school characteristics associated with higher levels of this type of turnover that has not been addressed in prior research. I also seek to identify the ways in which within-year turnover patterns resemble or differ from end-of -year turnover. In the second chapter, I conduct survival analysis of novice teachersâ monthly turnover patterns to understand differences in the risk of turnover for different teacher entry pathways and school characteristics. In the third paper, I leverage the within-year teacher turnover measure to better understand how three mechanisms linked to teacher turnoverâ(1) teacher instability; (2) classroom disruption; (3) replacement teacher qualityâharm student achievement. I find that an average of four percent of teachers turn over during each school year, suggesting that conventional measures of end-of-year turnover underestimate the actual frequency of turnover by as much as twenty-five percent. In line with the extensive research on end-of-year teacher turnover, within-year turnover is not evenly distributed across schools, with teachers who turnover midyear more likely to exit lower-performing schools with higher concentrations of economically disadvantaged students and underserved racial/ethnic minorities. This finding is troubling given evidence from the third chapter that finds within-year teacher turnover to have a negative effect on student achievement in English Language Arts and mathematics.

Over the past number of years, postsecondary students have increasingly enrolled in online courses. In this dissertation, I use human capital theory and Bayesian statistical analyses to investigate how students choose these courses, whether students perform as well in them as they do in face-to-face courses, and what the long-term degree outcomes among online course-takers may be. I first examine a potential mechanism of a studentâs choice to take online courses: access to high speed broadband. In preferred specifications, I find that every tier increase in download speed is associated with a 41% to 56% average increase in the number of students who enroll in at least one online course. I next estimate the effect of online delivery format on course outcomes among students enrolled in the University System of Georgia, finding that those in online courses were 2.8 percentage points more likely to withdraw and, conditional on completing the course, 5.4 percentage points less likely to pass with a C- or better. Finally, I produce state-level estimates of differences in aggregate degree outcomes between sometimes-online and never-online students using survey data that are only nationally representative with a technique from political science, Bayesian multilevel regression with poststratification (BMRP). On average, I find that never-online students were slightly more likely to earn a bachelorâs degree in six years, while students who took any online courses were slightly more likely to earn an associate degree in three years. Results vary considerably across the states, however, justifying the use of the method. My dissertation contributes to the education policy literature both through its empirical findings and its use of Bayesian methodologies. With each question, I offer new insights into postsecondary online courses that support a better understanding of how these courses may fit into studentsâ human capital enrollment decisions.