Spatial variations in school performance

A local analysis using geographically weighted regression

Stewart Fotheringham, M. E. Charlton, C. Brunsdon

Research output: Contribution to journalArticle

97 Citations (Scopus)

Abstract

In Britain, the performance of all state primary schools is assessed by students' attainment levels in a set of standardized tests administered to pupils at the ages of 7 and 11 (the so-called Key Stages 1 and 2, respectively). These data are analysed for 3687 schools in northern England. In particular, school performance is linked to the number of students taking the test at each school and to various socioeconomic indicators of the estimated school catchment areas. The latter are based on a geographical weighting function that links census data, an areal coverage, to school locations, a point coverage. Following a traditional global regression analysis, spatial variations in the relationships are examined with geographically weighted regression (GWR) to reveal some interesting geographical variations in the results.

Original languageEnglish (US)
Pages (from-to)43-66
Number of pages24
JournalGeographical and Environmental Modelling
Volume5
Issue number1
DOIs
StatePublished - 2001
Externally publishedYes

Fingerprint

spatial variation
regression
school
performance
coverage
student
socioeconomic indicator
weighting
primary school
geographical variation
pupil
regression analysis
census
analysis
test

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Earth and Planetary Sciences(all)

Cite this

Spatial variations in school performance : A local analysis using geographically weighted regression. / Fotheringham, Stewart; Charlton, M. E.; Brunsdon, C.

In: Geographical and Environmental Modelling, Vol. 5, No. 1, 2001, p. 43-66.

Research output: Contribution to journalArticle

@article{6e77790b07c745a39299984ad6e30b1f,
title = "Spatial variations in school performance: A local analysis using geographically weighted regression",
abstract = "In Britain, the performance of all state primary schools is assessed by students' attainment levels in a set of standardized tests administered to pupils at the ages of 7 and 11 (the so-called Key Stages 1 and 2, respectively). These data are analysed for 3687 schools in northern England. In particular, school performance is linked to the number of students taking the test at each school and to various socioeconomic indicators of the estimated school catchment areas. The latter are based on a geographical weighting function that links census data, an areal coverage, to school locations, a point coverage. Following a traditional global regression analysis, spatial variations in the relationships are examined with geographically weighted regression (GWR) to reveal some interesting geographical variations in the results.",
author = "Stewart Fotheringham and Charlton, {M. E.} and C. Brunsdon",
year = "2001",
doi = "10.1080/13615930120032617",
language = "English (US)",
volume = "5",
pages = "43--66",
journal = "Geographical and Environmental Modelling",
issn = "1361-5939",
publisher = "Carfax Publishing Ltd.",
number = "1",

}

TY - JOUR

T1 - Spatial variations in school performance

T2 - A local analysis using geographically weighted regression

AU - Fotheringham, Stewart

AU - Charlton, M. E.

AU - Brunsdon, C.

PY - 2001

Y1 - 2001

N2 - In Britain, the performance of all state primary schools is assessed by students' attainment levels in a set of standardized tests administered to pupils at the ages of 7 and 11 (the so-called Key Stages 1 and 2, respectively). These data are analysed for 3687 schools in northern England. In particular, school performance is linked to the number of students taking the test at each school and to various socioeconomic indicators of the estimated school catchment areas. The latter are based on a geographical weighting function that links census data, an areal coverage, to school locations, a point coverage. Following a traditional global regression analysis, spatial variations in the relationships are examined with geographically weighted regression (GWR) to reveal some interesting geographical variations in the results.

AB - In Britain, the performance of all state primary schools is assessed by students' attainment levels in a set of standardized tests administered to pupils at the ages of 7 and 11 (the so-called Key Stages 1 and 2, respectively). These data are analysed for 3687 schools in northern England. In particular, school performance is linked to the number of students taking the test at each school and to various socioeconomic indicators of the estimated school catchment areas. The latter are based on a geographical weighting function that links census data, an areal coverage, to school locations, a point coverage. Following a traditional global regression analysis, spatial variations in the relationships are examined with geographically weighted regression (GWR) to reveal some interesting geographical variations in the results.

UR - http://www.scopus.com/inward/record.url?scp=0035061114&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035061114&partnerID=8YFLogxK

U2 - 10.1080/13615930120032617

DO - 10.1080/13615930120032617

M3 - Article

VL - 5

SP - 43

EP - 66

JO - Geographical and Environmental Modelling

JF - Geographical and Environmental Modelling

SN - 1361-5939

IS - 1

ER -