Using data from the 1994 Health Survey for England, logistic multivariate multilevel modelling techniques are used to investigate the simultaneous effect of individual demographic characteristics and socio-structural factors on self-reported problem drinking as revealed by CAGE scores and `unsafe' levels of alcohol consumption. Whilst the influence of key socio-structural variables is broadly similar for both unsafe alcohol consumption and high CAGE scores, there are notable exceptions when results are examined by tenure group: those in the rented sector are more likely to be problem drinkers as revealed by CAGE, but less likely to consume `unsafe' amounts of alcohol. Both dimensions of drinking behaviour are influenced by the consumption patterns of others in the household, with both likelihoods increasing as the average consumption of others in the household rises. After taking into account individual compositional variables, the research indicates that there is very little evidence for geographical variation remaining in these two dimensions of drinking behaviour. It is found that the proportion of the population whose drinking behaviour may be classed as (potentially) problematic via the CAGE responses is substantially less than the proportion consuming above recommended `safe' levels. The research concludes, however, that the two measures are broadly similar in their relationship to social and structural variables. Tenure provides an exception to this conclusion and indicates a continuing need to take account of housing circumstances in developing an understanding of drinking behaviour.