An analysis of the performance of public elementary schools in New York City during 2001--2005 from a geographical perspective
Item
-
Title
-
An analysis of the performance of public elementary schools in New York City during 2001--2005 from a geographical perspective
-
Identifier
-
d_2009_2013:4b416c92a33f:10183
-
identifier
-
10391
-
Creator
-
Kellici, Ylli,
-
Contributor
-
Sean C. Ahearn
-
Date
-
2009
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Geography | Statistics | Social sciences education | Elementary education | Jackknife | Public Elementary Schools | School District | School Performance | Spatial Regression | Square Combining Table
-
Abstract
-
This study examines from a geographical perspective the factors that impact the performance of public elementary schools in New York City during 2001-2005, a period when its schools were undergoing major reforms at both the local and national level. Education reforms have focused their attention on schools by increasing their responsibility and autonomy concurrent with an increase in accountability. This increased focus on schools as the main agent of change presents a challenge for geographers to investigate the impact of the geographical context on school performance. Although school achievement has a geographical component, the contribution of geographers on this issue has been limited.;There are several major findings of this research. Despite a general increase in educational achievement in the public elementary schools in New York City during the period of this study, the race/ethnicity achievement gap remains substantial and does not change in space or over time. Findings indicate that there is spatial clustering in school performance the majority of which is explained by the spatial pattern of students' socioeconomic characteristics. There is no major spatial variation in the relationship between school performance and student and school characteristics and such relationships do not change over time. The presence of spatial dependence in school performance, not accounted for by student and school characteristics, necessitates the use of spatial regression models. The spatial model estimation indicates that in the first three years of the study period the spatial error model is a better fit of the data whereas in the last two years the spatial lag model is a better one. The switch from an error to a lag model, occurring in the first academic year (2003-2004) when education reforms were implemented in New York City, is an indication of a global change which can be identified with the effect of reforms throughout the City's public schools. Regression analysis shows that some school districts have an impact on school performance, after accounting for the student and school characteristics. From a policy perspective, the concentration of social disadvantage in space and over time should be taken into consideration in policies regarding allocation of resources that should be spatially focused. Furthermore, the education reforms should consider not only schools but also school districts in their accountability system.;In addition to its findings, this research contributes to the geographic literature by introducing a robust framework to explore the impact of the geographic context. This framework, with the Square Combining Table method at its core, includes also Bisquare Weights and Multiple Comparison procedures. Another important methodological contribution is the introduction to educational literature of the Jackknife technique to examine the spatial variation of the relationship between school performance and student and school characteristics. The methods applied on the jackknifed coefficient values to study their variation in space and over time can similarly be used in other geographical phenomena.
-
Type
-
dissertation
-
Source
-
2009_2013.csv
-
degree
-
Ph.D.
-
Program
-
Earth & Environmental Sciences
