Weighted Census-based deprivation indices: their use in small areas

The concept of deprivation refers to the conditions experienced by people who are poor, whereas the concept of poverty relates to the lack of income and other resources that make these conditions difficult to break away from. The usual Census-based deprivation indices are often based on equally weighted (and highly correlated) variables. This is in spite of the fact that different social groups have been shown in numerous studies to have differing probabilities of suffering from deprivation. A weighted deprivation index based on individual level data and Census data produces a more accurate and more easily understandable method of estimating deprivation within an area, as it reflects the differences between social groups. This paper will describe how a technique used by Gordon (J Epidemiol Commun Hlth 1995; 49(Suppl. 2): S39-S44) can be applied to smaller areas to give estimates of the number of deprived living within them, producing a weighted index.

Breadline Greenwich data together with Census data were used to estimate the percentage of poor households in Greenwich wards using similar methods to those of Gordon.

The results produced were very similar to the numbers estimated using the Breadline Britain weightings but there were differences in individual weightings for some of the variables.

Gordon's approach can be applied to smaller areas to give estimates of the number of deprived using the nationally derived weightings. More accurate local estimates, subject to the different local conditions, can be easily derived if a similar survey to the Breadline Britain survey is conducted locally.