Deprivation scales derived from multiple, typically dichotomous, indicators, are widely used to monitor households standards of living, and to complement measures of living standards based on income. We use an item response modelling (IRM) framework to address several issues concerning the derivation of deprivation scales in general and the use of sum-score deprivation indices in particular. Although we favour the IRM approach over the sum-score one in principle, we find in an illustrative analysis of basic lifestyle deprivation in Britain in the mid-1990s that both approaches provide very similar pictures of households circumstances. We conclude with further discussion of the relative merits of the two approaches and highlight some topics for future research.