BACKGROUND Lack of sleep and increased consumption of energy-dense foods and sugar-sweetened beverages (SSBs) have all been suggested as factors contributing to the increased prevalence of overweight and obesity. OBJECTIVE To evaluate whether objectively measured sleep duration (average and day-to-day variability) as well as parent-reported sleep problems are independently associated with proposed dietary risk factors for overweight and obesity in 8-11-year-old children. METHODS In this cross-sectional study, data on sleep duration and day-to-day variability in sleep duration were measured in 676 Danish, apparently healthy children by an objective measure (actigraphy) for 8 nights, and the Children's Sleep Habits Questionnaire (CSHQ) was filled out by the parents. Diet was recorded using a web-based food record for 7 consecutive days. Fasting blood samples were obtained for measurements of plasma leptin and ghrelin levels. RESULTS Sleep duration (h per night) was negatively associated with energy density (ED) of the diet (β = -0.32 kJ g(-1)), added sugar (β = -1.50 E%) and SSBs (β = -1.07 E%) (all P ≤ 0.003). Furthermore, variability in sleep duration (10-min per night) was positively associated with SSBs (β = 0.20 E%, P = 0.03), independent of sleep duration, and CSHQ score was positively associated with ED (β = 0.16 kJ g(-1), P = 0.04). All of these associations were independent of potential confounders (age, sex, pubertal status, height, weight, screen time, moderate-to-vigorous physical activity and parental education and ethnicity). CONCLUSIONS Our study suggests that short sleep duration, high sleep duration variability and experiencing sleep problems are all associated with a poor, obesity-promoting diet in children.