Linking Point and Area Data to Model Primary School Performance Indicators
The introduction of standardized tests for 10- and 11-year-old children in England and Wales and the subsequent publication of the results for children at different schools have led to controversy about the relationship between academic achievement, school effectiveness and social background. An obvious approach to disentangling the relationships between these factors is to compare schools' test scores with measures of social composition for the local area. Although many relevant variables are available from the 1991 Census, there are major problems in determining how results for schools (for which point locations are available) can be compared with census data (available only for areal units). Three approaches to linking the data sets-Voronoi polygons, weighted Voronoi polygons, and a constrained allocation method-are compared for schools in Lancashire. Ordinary Least Squares and logit regression models are then fitted to the data using all three approaches. In all cases, the results show relationships between test scores and several socio-economic variables, including ethnicity, unemployment, housing tenure, and educational qualifications of the adult population. The allocation method shows the clearest results, and appears to be the most effective of the techniques.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
Document Type: Regular Paper
Publication date: 2001-05-01