Research has consistently found that teachers effectiveness improves during the first few years on the job. These increases are typically attributed to improvements in the novice teachers teaching practices. Yet, we know little about which practices new teachers are most likely to improve during their first years of teaching. This study addresses gaps in existing teacher quality research by examining the development of beginning middle school math teachers' classroom environment and instructional practices, the relationship between these practices, and the association of these practices to teacher effectiveness. I found that beginning teachers began teaching with higher ratings of classroom environment practices and were more likely to improve those practices during their first three years. The beginning teachers who demonstrated higher levels of mastery of their classroom environment practices were significantly more likely to both assign and implement math-specific instructional practices. I also found that the beginning teachers who demonstrated greater proficiency and improved most in their math-specific instructional practices saw greater increases in student learning.

The recent adoption of career and college ready standards also coincides with the adoption of new statewide accountability assessments aligned to these standards. While the development of assessments aligned to these updated and more rigorous standards represents an opportunity for states to orient their assessment and accountability systems to measuring and promoting ambitious goals for teaching and learning, it is unclear whether these tests are in fact more likely to measure or reward ambitious instruction in the classrooms. My analysis of a longitudinal data set with scores from more than 39,000 student-year observations, nested in 244 teachers in 27 schools in two districts across seven years provides evidence which suggests that in one district (District D), prior to the adoption of CCR standards and assessments, the relationship between classroom measures of ambitious math instruction and student test score gains were not statistically distinguishable from zero at conventional levels of significance. However, in the three years subsequent to the move to CCR standards and assessments, more ambitious mathematics instruction in the classroom was associated with positive and statistically significant gains in student test scores in this district. However, in the other district (District B), even after the change to an ostensibly more rigorous set of standards and state assessments, there was no statistically significant change detected in the relationship between ambitious teaching of mathematics and estimated teacher-level contributions to student test scores on these assessments. Implications for policy and future research are discussed.

Despite a national call from advocacy groups and the U.S. Department of Education for expanded choice, students residing in segregated urban neighborhoods are often locked out of school options due to limited access to the transportation networks required to optimize school options. Consequently, studentsâ choices are often limited to their assigned school comprised of students from proximal segregated neighborhoods. The intersection of school choice, racial integration, and the geography of opportunity is addressed in this dissertation. I examine the enrollment patterns of public secondary school students in Metropolitan Nashville Public Schools (MNPS) as an illustrative case of this convergence. MNPS has created a robust school choice system; however, school buses are provided only for assigned zone schools and select charter schools. <p> A unique administrative dataset with student addresses from 2011 to 2015, census tract data from the American Community Survey, and travel times on public transportation calculated with Google Maps are used to analyze distinctions in enrollment patterns by neighborhood characteristics, assignment to an integrated school, and enrollment in an integrated school. In late 2014 MNPS partnered with the Metro Transit Authority to provide public high school students with a bus pass (StrIDe). Descriptive statistics, GIS analysis, OLS and logistic regression, as well as differences-in-differences analyses are conducted to evaluate enrollment patterns and changes as students are granted access to bus passes. <p> Findings suggest students enroll in schools that demographically reflect their residential neighborhood. Even with school choice options, a larger proportion of students zoned to an integrated school enroll in an integrated school than the proportion of their counterparts zoned to a racially identifiable school. Eligibility for StrIDe is associated with attending a school with more academically proficient peers and a higher likelihood of attending a choice school, an integrated school, and a school with a high value-added score. The StrIDe policy is associated with changes for students near transit routes, but expanded policies may be necessary to reach students residing far from the city center.

American children of lower-income, Black, or Hispanic parents are more likely to suffer from poorer health, poorer academic performance, and ultimately poorer earnings, higher imprisonment, and earlier deaths than children of higher-income parents. They are also substantially less likely to have access to quality healthcare and highly effective teachers. Research has consistently confirmed correlations between disparate access to quality educational and social services, and poor life outcomes. However, scholars and policymakers need to know more about the effectiveness of policies designed to counteract these inequities. In this three-essay dissertation, I evaluate the effectiveness of a set of initiatives that seek to reduce health and educational barriers to the success of economically disadvantaged children. The range of obstacles to academic success for children living in poverty is broad. However, I focus here on policies targeting two disparities: a lack of access to healthcare and high-quality instruction. Specifically, the studies evaluate the impacts of three policies: (a) expanded childrenâs health insurance, (b) school-based health centers (SBHCs), and (c) selective retention bonuses (SRBs) for teachers in high-poverty schools. Using national data from the Early Childhood Longitudinal Survey (ECLS-K), I find evidence that gaining health insurance coverage increases the frequency of doctor and dentist visits, and decreases the number of school days missed by students. Using rich longitudinal administrative data maintained by the Tennessee Education Research Alliance (TERA), paired with health service records from the Tennessee Office of Coordinated School Health (TNCSH), I also find that increased exposure to health services from SBHC openings improve both the attendance and test scores of students in Tennesseeâs non-urban schools. Finally, I find that student test scores improve in response to the implementation of a selective retention bonus program targeting the highest-rated teachers in Tennesseeâs lowest performing schools, which are concentrated in high-poverty urban school districts. In conclusion, the results of the three studies provide promising, though modest evidence that interventions designed to improve access to healthcare and effective teachers can decrease barriers to academic success for low-income children.

Charter school policy is driven by a combination of empirical concerns about their effectiveness as well as ideological motivations. However, there has thus far been a relative lack of evidence on the latter, making it difficult to identify ways to shift the policy debate. This dissertation begins to fill this gap by pursuing three related research projects. First, I conduct a qualitative investigation of two opposing social movements in East Nashville to break down their positions on charter schools and school choice. Here, I find that there are significant differences in perspectives on choice and charter schools, including perspectives on the role of teachers and leaders, community, and no-excuses environments. However, there are also important points of commonality. Second, I employ topic modeling methods to organize and identify the language and concepts associated with charter school coverage in United States news media. First, in earlier years, there was significant emphasis on the publicness of charter schools and their independence from the public system. Later, there was more of an emphasis on the financial resources aspect of charter school policy. Lastly, drawing on lessons from political psychology, I present the results from an information search exercise designed to illuminate the types of information that people find most important when evaluating charter school policy. I find evidence that people tend to seek out pro-attitudinal information and that peopleâs search behaviors differ by ideology and schooling background. Altogether, the results from these studies have important implications for researchers and policymakers alike. The most important takeaway is that charter school attitudes are not always primarily driven by concerns about their effectiveness. This dissertation demonstrates several ways in which the other ideological considerations are indeed observable, opening up a new avenue for exploration.

This study examined how student characteristics could be used to predict whether or not a community college student would persist in an online course. The research question guiding the study was, “Which student characteristics can be used to best predict the persistence of community college students in online courses?” The student characteristics examined were learning style, locus of control, computer experience and access, previous online experience and demographics. A survey instrument consisting of two previously developed instruments and a Computer Experience scale that was created by the researcher specifically for this study, was administered to online students at one Florida community college for the pilot study and five additional Florida community colleges for the full study. Confirmatory and exploratory factor analysis were conducted on the computer experience scale to determine if there was an underlying hidden structure. Stepwise logistic regression was completed to determine the student characteristics that were significant predictors of online persistence, as well as an equation that could be used to predict whether or not a community college student would persist in an online course. Confirmatory and exploratory factor analysis revealed that the Computer Experience scale consisted of three underlying subscales. The researcher named the three subscales based on the similarities of the variables that were associated with each factor: Factor one (basic computer skills); Factor two (Internet/email skills); Factor three (interactive computing skills). Three of the initial 25 predictor variables were found to be significant predictors of community college online persistence: GPA, auditory learning style, basic computer skills. An increase in both auditory learning style and basic computer skills was associated with a decrease in the odds of course persistence. On the other hand, an increase in GPA was associated with an increase in the odds of course persistence. Additionally, an equation to predict whether or not an online community college student would persist in an online course was developed. Implications for community college administrators as well as recommendations for future studies are also provided in the study. / Dissertation / PhD