Several methods are available to measure social inequalities in health. This paper discusses the advantages and disadvantages of different approaches, in particular the odds ratio, the slope and alpha. These methods are illustrated using data from subjects in the 1958 British birth cohort. The inequality measures are compared using health status at ages 23 and 33. Six health indicators are examined, including self-rated health, limiting long-standing illness, psychological health, respiratory symptoms, asthma and obesity. Two social indicators are compared, namely class at birth and educational qualifications. Conclusions do not differ substantially using the three methods for measuring inequality. However, consistent differences were evident between the measures of social position, with greater inequalities apparent for educational qualifications. Choice of social indicator therefore appears to be of primary importance in measuring health inequality.