The use of multiple regression models to determine weights for the prediction of UK local authority spending targets is criticised. It is argued that predictions based upon such regression weights are not valid estimates of ‘need to spend’ and in particular that they should not be used as the basis for decisions about ‘capping’.
Get full access to this article
View all access options for this article.
References
1.
DerbyshireM E, 1987, “Statistical rationale for Grant-Related Expenditure Assessment (GREA) concerning personal social services”Journal of the Royal Statistical Society A150309–333 (with discussion)
2.
DES, 1984, “School standards and spending: Statistical analysis: A further appreciation”, Statistical Bulletin 13/84, Department of Education and Science, Elizabeth House, York Road, London SE1 7PH
3.
DoE, 1989, “Application of regression analysis to needs assessment”, NSG:NASG(89) 58 CORE(3) March 1989, Department of the Environment, 2 Marsham Street, London SW1E 5HE
4.
DoE, 1990, “Standard spending assessments, background and underlying methodology”, Local Government Finance Policy Directorate, Department of the Environment, 2 Marsham Street, London SW1E 5HE
5.
MaystonDSmithP, 1990, “Analysing the need to spend on education”Journal of the Operational Research Society41125–131
6.
TaylorS J, 1989, “Measuring need to spend on education”Journal of the Operational Research Society40177–186
7.
WoodhouseGGoldsteinH, 1988, “Educational performance indicators and LEA league tables”Oxford Review of Education14301–320
