Browsing by Subject "Multilevel analysis"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item The effect of teacher quality on student achievement in urban schools : a multilevel analysis(2015-05) Kim, Su-Yun; Saenz, Victor B.; Reyes, Pedro; Valenzuela, Angela; Paxton, Pamela; Brown, KeffrelynThe No Child Left Behind Act recognizes the importance of quality teachers in improving student achievement in that it mandates that all students have to be taught by “a highly qualified teacher”. The increasing demand for highly qualified teachers has led to a shortage of qualified teachers. In the United States, however, an uneven distribution of high quality teachers exists. A closer look at urban areas reveals that the problem is more severe in those localities than the national average. In order to address the teacher shortage problem in urban areas, more than 40 states initiated an alternative certification route for candidates who hold a bachelor’s degree (Darling-Hammond, 2000; Heilig, Cole, & Springel, 2011), although teachers’ certification or licensing status play an important role in differentiating teacher quality. The purpose of this study was to examine how high quality teachers are distributed across a large, urban district in Texas according to student’s characteristics, school characteristics and student achievement. In addition, more importantly, this study explored how teacher’s quality influences student achievement and, more specifically, on achievement of students with limited English proficiency (LEP). Due to the differences of characteristics in student characteristics and a school system, elementary schools and middle schools were separated in the analyses. In order to examine which students were allocated to high quality teachers and to determine the effect of teacher quality on students’ achievement in an urban district, Southeast Independent School District (SISD), which is a major urban district in Texas with more than 200,000 students, was chosen. As the largest public school system in Texas, SISD has large shares of minority and low-income students. Student data utilized in this analysis came from the Public Education Information Management System (PEIMS), which is data collection and reporting system produced by the Texas Education Agency (TEA) for the public schools of Texas. SISD provided individual level teacher’s data, students’ data, and a matching file so that teacher’s and their students’ data could be linked. All data that SISD provided are protected by using masked identification. To address the research questions, the study involved three statistical approaches – descriptive analysis, Analysis of variance (ANOVA) and three-level hierarchical linear models (HLM). Results from ANOVA indicated unequal distribution of high quality teachers across an urban school district. Economically disadvantaged students, minority students, and students with limited English proficiency were more likely to be allocated to alternatively certified teachers. It implies that students with economically and socially disadvantaged backgrounds lose a chance to have fully-certified or highly qualified teachers. The test scores of students who had fully-certified teachers were higher than the test scores of under-certified teachers’ or alternatively-certified teachers’ students. Campus accountability ratings were also significantly lower for schools that had more Teach for America (TFA) teachers than schools that had more fully-certified teachers. There were also clear distinctions among teacher’s qualifications, student characteristics, and school conditions between elementary schools and middle schools. There were more alternative certified teachers and less fully-certified teachers in middle schools. Middle schools served a higher percentage of students that are economically disadvantaged, at-risk of dropping out, were LEP, and Hispanic. The average campus accountability rating was also lower in middle schools than elementary schools. Overall, school conditions in middle schools were more inferior than in elementary schools among urban schools in Texas. In order to explain the effect of teacher quality and school condition besides student’s characteristics on student performance, a multilevel analysis was necessary to explain each variance of students, teachers, and schools. Through multilevel analyses (or three-level hierarchical linear modeling (HLM)), I confirmed that student background or ability was the strongest predictor of student achievement as many previous studies have found. The results showed that student achievement significantly differed by students’ background or ability even when they have the same reading teachers. However, HLM results also showed that teacher’s and school’s effects on student achievement were not negligible based on their proportions of variances. It implied that student achievement could be differentiated by teacher’s quality or school’s conditions. Among variables regarding teacher qualifications, the fully-certified teacher variable was a solely significant and positive factor of student achievement in middle schools. That is, students who had fully-certified teachers were more likely to achieve higher test scores than those who had under-certified and alternatively certified teachers after controlling all variables. However, in elementary schools that had 95 percent of fully-certified teachers did not show the significant differences of student achievement by teacher’s qualifications. The years of teaching experience and teacher educational attainment was not significant factors to explain student performance. Among school-level predictors, campus accountability ranking was a positively significant factor to predict student achievement in both elementary schools and middle schools. The percentage of economically disadvantaged students in campus was negatively associated with student achievement in middle schools. Since the study focused on reading achievement, the effect of teacher’s quality on the achievement of LEP students was particularly concerning. To address research questions, an interaction effect between teacher certification status and the achievement of LEP students was added on the three-level model. Results from the analysis showed that after accounting all variables LEP students who had fully-certified teachers achieved 0.1 scores higher on the TAKS reading test in the middle schools. Considering that LEP students typically achieved lower than their peers, the results implied that fully-certified teachers mitigate the effect of LEP on TAKS reading. The finding showed a positive effect of fully-certified teachers for students in need and corresponded with previous studies that high quality teachers played a more important role for socially and economically disadvantaged students. To sum up, this study found that teacher quality is a significant factor to predict student achievement, yet highly qualified teachers are unequally distributed across an urban school district. Socially and economically disadvantaged students were less likely to be taught by fully-certified teachers and were more likely to be taught by alternatively certified teachers. Furthermore, their achievement was significantly lower than their peers who were taught by highly qualified teachers. These aspects were more noticeable in middle schools.Item Investigating Childhood Overweight and Obesity in Rural Settings(2012-12-01) Serrano, Katrina 1983-Children?s risk for overweight and obesity is particularly high in rural areas of the United States. Many health, psychosocial, and economic consequences are associated with childhood overweight and obesity, which concerns health researchers and professionals. But how and why might rural children be more at risk for being overweight and obese? This dissertation investigates childhood overweight and obesity in rural settings through three separate studies. First, a systematic literature review was conducted to identify determinants and mechanisms of childhood obesity-related behaviors that are specific to rural locations. The findings from the review show that lack of health resources and poverty within the rural environment may impact children?s social environment and individual factors. However, results are inconclusive and there continues to be a lack of studies focusing on linking environmental influence with individual factors. Second, a meta-analysis of current research evidence was conducted to assess the efficacy of rural interventions designed to reduce childhood overweight and obesity. Results showed that interventions have been efficacious yet modest, with a mean effect size of 0.18. Moderating variables were also examined. Mean intervention effect size was moderated by children?s age and intervention duration. Last, secondary data were used to examine the association between rural food stores and availability and affordability of fresh fruits and vegetables. A multilevel analytical approach was used to determine if rural location was associated with availability and affordability of fresh fruits and vegetables. After controlling for other variables, results showed that rural location was not associated with fruit and vegetable availability and affordability. The findings from this dissertation suggest that the area of rural childhood overweight and obesity remains understudied. More research is needed in order to understand the mechanisms of social ecological influences on diet, physical activity, and childhood overweight and obesity. This area of research, however, is rife with opportunities for public health education and promotion. Public health educators can help promote and advocate for environmental conditions that support healthy lifestyles.Item Longitudinal analysis on AQI in 3 main economic zones of China(2014-05) Wu, Kailin; Powers, Daniel A.In modern China, air pollution has become an essential environmental problem. Over the last 2 years, the air pollution problem, as measured by PM 2.5 (particulate matter) is getting worse. My report aims to carry out a longitudinal data analysis of the air quality index (AQI) in 3 main economic zones in China. Longitudinal data, or repeated measures data, can be viewed as multilevel data with repeated measurements nested within individuals. I arrive at some conclusions about why the 3 areas have different AQI, mainly attributed to factors like population, GDP, temperature, humidity, and other factors like whether the area is inland or by the sea. The residual variance is partitioned into a between-zone component (the variance of the zone-level residuals) and a within-zone component (the variance of the city-level residuals). The zone residuals represent unobserved zone characteristics that affect AQI. In this report, the model building is mainly according to the sequence described by West et al (2007) with respect to the bottom-up procedures and the reference by Singer, J. D., & Willett, J. B (2003) which includes the non-linear situations. This report also compares the quartic curve model with piecewise growth model with respect to this data. The final model I reached is a piece wise model with time-level and zone-level predictors and also with temperature by time interactions.