This paper uses data from the British Health and LifestyleSurvey to explore the relationship between women's overall assessment of their health and their occupation, employment status, household composition and household income. Logistic regression was used to predict the odds of assessing health as either 'good' or 'less-than-good'. These findings suggest that after controlling for age and the presence of a long-standing illness or disability, women's occupational group has the strongest influence on self-assessed health. The re-classification of certain occupational groups, in particular retail occupations, from their conventional socio-economic categories is also vindicated in terms of their assessed health adding in line with the suggestion that, with respect to their income and terms of emplojinent, retail workers are 'misclassified' as non-manual. This analysis shows also that women who were economically inactive were more likely to evaluate their health as 'less-thangood' than women in paid work, after controlling for the presence of a long-standing illness or disability. This is interpreted as evidence that not all of the poor health usually attached to the role of 'housewife' can be explained in terms of health selection out of the labour market.