OBJECTIVE: Poor-quality diet, regarded as an important contributor to health inequalities, is linked to adverse health outcomes. We investigated sociodemographic and lifestyle predictors of poor-quality diet in a population sample. DESIGN: A cross-sectional analysis of the Survey of Lifestyle, Attitudes and Nutrition (SLÿN). Diet was assessed using an FFQ (n 9223, response rate = 89 %), from which a dietary score (the DASH (Dietary Approaches to Stop Hypertension) score) was constructed. SETTING: General population of the Republic of Ireland. SUBJECTS: The SLÿN survey is a two-stage clustered sample of 10 364 individuals aged 18 years. RESULTS: Adjusting for age and gender, a number of sociodemographic, lifestyle and health-related variables were associated with poor-quality diet: social class, education, marital status, social support, food poverty (FP), smoking status, alcohol consumption, underweight and self-perceived general health. These associations persisted when adjusted for age, gender and social class. They were not significantly altered in the multivariate analysis, although the association with social support was attenuated and that with FP was borderline significant (OR = 1·2, 95 % CI 1·03, 1·45). A classical U-shaped relationship between alcohol consumption and dietary quality was observed. Dietary quality was associated with social class, educational attainment, FP and related core determinants of health. CONCLUSIONS: The extent to which social inequalities in health can be explained by socially determined differences in dietary intake is probably underestimated. The use of composite dietary quality scores such as the DASH score to address the issue of confounding by diet in the relationship between alcohol consumption and health merits further study.