Some indicators of socio-economic status may not be reliable and use of indices with these data could worsen equity.

Socio-economic status (SES) indices are increasingly being used to characterise (in)equity, with the assumption that SES indices are reliable. However, the accuracy of such SES indices is questionable if they are unreliable. We examined the inter-rater and test-retest reliability of a range of indicators commonly used to generate SES indices as well as the composite SES indices. Results from research in southeast Nigeria showed considerable variation, with some indicators having only low to moderate reliability (reliability coefficients 0.25-0.77). Inter-rater and test-retest reliability of SES indices was 0.63 in both cases. Many households were misclassified because of the unreliable SES indices. Analyses of the distribution of resources based on such indicators could lead to inaccuracies in benefit incidence estimates and policy decisions based on low to moderately reliable SES indicators could worsen equity in access to and use of resources. Greater rigour is needed in conceptualising as well as undertaking SES measurements. Copyright 2005 John Wiley & Sons, Ltd.