Homeless elementary school students are a vulnerable population. My mixed-methods study looks closely at one large urban district to determine how its homeless students fare academically and how their schools strive to meet their multiplicity of needs. My quantitative analysis shows that homeless students do not perform notably worse than low SES students who have housing. No school-level factors are found to predict homeless student achievement, and differences in average homeless students scores across schools are very small. In qualitative interviews and surveys with principals, teachers, and guidance counselors, I find that most are primarily concerned with logistical issues and providing emotional support, rather than designing specific academic interventions for homeless students. Although the strategies interviewees described mirror the literature closely, I did not find that schools whose homeless students performed above district averages employed different strategies than those below. Most strategies described as supporting homeless students are used for struggling housed students as well. I offer implications of my findings for research, policy, and practice.

This dissertation focuses on teacher retirement systems. The defined benefit pension plans that cover K-12 public school teachers in most states face two major problems: they may not be financially sustainable as large numbers of teachers retire in the coming years, and they may not be serving to recruit and retain a high-quality teaching force. I propose that in order for pensions to serve as a policy lever to attract and retain high-quality teachers, three conditions must be met. First, teachers must understand their retirement plans and the incentives imbedded therein. Second, they must value these incentives. Finally, the incentives must be aligned with the desired pattern(s) of retirement behavior. This dissertation investigates the extent to which each of these conditions is being met using data from the Schools and Staffing Survey and Teacher Follow-Up Survey as well as original data collected via a survey and embedded focus groups. It also outlines how pensions could be reformed to be better aligned with desired retirement behavior to recruit and retain high-quality teachers.

This research project examines school superintendents advice-seeking behavior as it relates to strategic decision making. Using an exploratory design intended to launch follow-on studies to pioneer this field of research in education administration, the study combined field observations with open-ended interviews to find out if mandated state and federal school reforms had increased superintendents focus on strategic matters, and, if so, how advice-seeking tendencies influenced related decision-making processes. Time-on-activity data showed a large increase in strategic behavior, and triangulation of interviews revealed four advice-seeking behaviors (Isolation, Broad Search, Confer with Consultants and Other) that basically described two tendencies (limited search or broad search for alternatives). Although similar to CEOs in several areas, superintendents were found to differ in a unique way: school administrators who tended to limit input leaned toward isolation, whereas business executives who limit input prefer to confide in friends or CEOs like them (seeking people who would affirm rather than challenge their ideas).

Public colleges and universities exist under constant pressure to reduce budgets and diminish spending. One way to do this is to utilize contingent faculty; but does this come at the expense of student success? The purpose of this study is to evaluate the impact of institutional faculty characteristics on the likelihood of degree completion for remedial education students. This study utilizes a multilevel analysis of students within institutions. Students are first assigned to a curriculum based on their remedial education status. Within each institution and by subject, the percentage of remedial education and general education courses taught by the different faculty types (rank, full-time status, education level) was determined. Thus, each student was assigned a treatment based on the institution they attended, their remedial status in each subject and the characteristics of professors. The results of this analysis found little to no support that the interaction between remediation curriculum and select faculty characteristics had a positive or negative effect on the likelihood of degree completion.

Far too many students who begin at a community college are unsuccessful in transferring to a four-year institution and completing a baccalaureate degree. How can we explain this phenomenon and how could policy interventions be implemented to improve student success? To better understand the role of the community college, I examine three specific pathways. First, I investigate the transfer process between two- and four-year institutions with a focus on credit hours earned in the first semester. Second, I examine the degree attainment patterns for those students who successfully transfer to a four-year school in comparison to their peers who initially began in the four-year sector. Third, I explore the time-varying factors associated with overall undergraduate degree attainment, with a focus on wages earned while currently enrolled, for those students beginning at the community college.

Race to the Top is a new policy lever in the struggle for authority in educational governance among various levels of government, and it provides an opportunity to examine the state-level policymaking context. Tennessee passed comprehensive education reform legislation in an effort to win the competition. This effort was successful: the Volunteer State was one of two winning state in the first round. I conducted a two-part case study of the events surrounding the states application and the development of an educator evaluation policy as required by the legislation. Interviews with key actors in the policymaking process reveal that the legislation was not an isolated attempt to improve education, but part of a series of steps led by the governor. His proactive leadership and inclusive style were crucial factors in the legislations successful passage. The legislation established a 15 member committee of diverse backgrounds to develop policy recommendations for an annual educator evaluation system. Interviews with these committee members and an analysis of pertinent documents show the institutional challenges the committee faced and the ways in which the process was open to public input. This study presents an examination of rapid, fully supported policy change in one state, including a description of how non-traditional actors developed policy. It is a story of how democracy worked cooperation, coercion, or backroom deals in one Southern state as it tries to improve education for its youngest citizens. Implications for policy and practice are discussed as are steps for future research.

In this dissertation, I examine the mechanisms that create qualified engineers. I am motivated by the important role that engineers play in the economic and social development of nations around the world. First, I ask whether the use of computers in formal and informal settings leads to higher problem solving achievement for low-income high school students. I use propensity score matching with PISA to recover causal estimates of the effect of computer use. Paper two highlights the in- and out-of-school factors that lead to engineering achievement for college students in Brazil. It uses a large national sample of first- and final-year students in one of the first rigorous quantitative estimates of the effects of inputs for higher education achievement. Paper three adds vital qualitative perspective to the other two secondary datasets and follows chronologically, investigating the post-college plans of engineers from South Africa. I gather new data (a survey I have designed) to determine the predictors of the pursuit of a local or global engineering career. Paper one finds that school use of computers has a positive effect on problem-solving achievement in the two large, diverse, high-income countries studied; home use has no effect or a negative effect; and use elsewhere is positive at low levels of use. Paper two finds that a student's home environment and the schooling she was exposed to before college predict her score, even within institutions. University factors also matterthere is growth from first to final year, and factors such as large classes and reports of bad teachers are related to lower scores. Paper three finds that, while engineering students are intrinsically motivated, unless they have exposure to hands-on opportunities in relevant local engineering, they personally are not motivated to work in that space. This dissertation makes contributions in its focus on social context, its employment of novel datasets, its introduction of the theoretical concept of technological capital, and its implications for policy. If barriers within the engineering pipeline are removed for underserved students and incentives to persist into relevant engineering practice are increased, these students will access, apply, and benefit from technological capital.

Although school leadership is widely recognized as an essential component of highly effective schools, we know little regarding the qualities and characteristics of the teachers who choose to become school leaders or how these qualities and characteristics correspond to the types of schools they choose to lead. An examination of the existing research on effective leadership, effective teachers, and effective schools suggests that, upon moving to administrative positions, highly effective teachers or teachers who worked in highly effective schools may be more likely to display the leadership behaviors associated with improving student outcomes. It follows then, that introducing measures of school and teacher quality to analyses of administrative labor supply may yield fresh insights regarding school leadership. Results show that teacher and school quality are positively associated with matriculation into school leadership when comparisons are made among all teachers. However when comparisons are made among only teachers with administrative certificates these quality measures are not longer significant. Results also show that high quality teachers and teachers from high quality schools are more likely to move into leadership positions in other high quality schools. Additional descriptive findings are discussed, as are policy implications and possibilities for future research.

Though there is widespread evidence that teachers matter, a more challenging problem exists in attempting to measure teacher effectiveness. It can be argued that student feedback is an important consideration in any teacher evaluation system as students have the most contact with teachers and are the direct consumers of a teachers service. The current paper outlines the development and preliminary validation of a student survey on teacher practice. Using data from a large-scale pilot in Georgia, the analysis finds that teacher scores on a student survey have a positive and marginally significant relationship to value-added estimates of teacher effects on student achievement. Further, there is a strong link between teacher scores and measures of academic student engagement and student self-efficacy. Finally, the paper investigates policy related issues that are pertinent to implementing student surveys as a component of teacher evaluation.

English language learners are one of the most rapidly growing demographic groups of students in the United States. A key goal for this group of students is to transition out of English language learner status by demonstrating English proficiency. Previous research has demonstrated that the timing of when a child is reclassified as English proficient may influence students educational outcomes, yet little is known about what facilitates or hinders the reclassification process. Using a rich longitudinal student-level statewide dataset from Texas, this dissertation employs event history analysis to investigate factors that influence English language learners probability of reclassification. Specifically, this dissertation examines the role that English proficiency and achievement assessments play in the reclassification process and explores the degree to which students social demographic characteristics, educational profile, schooling environment and local policy context influence reclassification decisions. In doing so, this dissertation makes a timely and unique contribution to the research literature on improving educational access and equity for English language learners by disentangling how state assessments, student characteristics and local context drive the rate of the reclassification process, which may in turn determine how quickly English language learners are granted access to valuable educational resources such as more advanced academic tracks, higher quality teachers and meaningful social networks with peers who are proficient in English.