In recent years a number of research projects have reported on the level of alcohol use by young people. The research to date, however, has tended to overlook the ethnic origins of the young people as a factor associated with the level of alcohol use. Against this background, this paper reports findings on the use of alcohol by 15-16-year-olds in Leicestershire with specific reference to ethnic group. A survey of over 1000 young people revealed significant differences between ethnic groups in terms of their attitudes to alcohol consumption and the frequency of alcohol consumption. As expected, South Asians tended to hold less favourable attitudes to drinking alcohol than their White counterparts and reported a far lower frequency of alcohol consumption. The extent of the divergence between the two ethnic groups is detailed, revealing a stark contrast between the groups at age 15-16 years. Attitudes to alcohol consumption and reported frequency of alcohol consumption were also analysed in relation to the religion of the respondents. This analysis was undertaken to see if there were differences within the broader ethnic groups which would be disguised by aggregating the results under the headings of 'South Asian' and 'White'. Within the South Asian group, some differences in attitudes to the consumption of alcohol were found between Hindus, Sikhs and Muslims, with the Muslims exhibiting particular sensitivity to their religion's proscription of drinking alcohol. Despite this, the reported level of drinking by the Hindus and Sikhs was not much greater than that of the Muslims. The three groups tended to have a similar frequency of alcohol consumption which was markedly lower than that reported by the White 15-16-year-olds. The findings from the DART research have certain substantive and certain methodological implications in relation to epidemiological studies of alcohol consumption. Substantively, the findings reinforce the need to include ethnic group as a key factor and highlight the danger of aggregating the results from different ethnic groups. In the case of Leicestershire, the effect of ignoring the ethnic dimension would have been to underestimate significantly the extent of drinking among the White 15-16-year-olds. Methodologically, the findings offer some support for the continued use of 'South Asian' as an analytic category since the variations observed within the category were relatively minor, especially in comparison with results for the White 15-16-year-olds.