ABSTRACT Objective: Suboptimal sleep is associated with disruptive behaviors in childhood. We evaluate associations of mean and variability (SD) of sleep duration, quality, and timing with emotion regulation, impulsivity, and prosocial and antisocial behavior in children. Methods: Intervention Nurses Start Infants Growing on Healthy Trajectories, a randomized controlled trial designed for obesity prevention, compared a responsive parenting intervention delivered in the first 2.5 years after birth with a home safety control group. At age 6 years, children wore an actigraphy device for 7 days and participated in behavioral tasks evaluating behavioral control, emotion regulation, and prosocial and antisocial behaviors. Separate linear regression models examined associations between sleep and behavioral variables, adjusting for study group, child sex, and household income. Moderation analysis investigated whether the study group moderated relationships between sleep and positive age-appropriate behavior. Results: Children (N = 143, age 6.7 ± 0.3 years) were predominantly non-Hispanic White (95%). Mean actigraphic sleep duration, quality, and timing were not associated with behavioral variables. By contrast, greater variability in sleep onset timing was associated with greater impulsivity (B = 0.85, p = 0.004) and poorer emotion regulation (B = −0.65, p = 0.01). Greater variability in sleep midpoint timing was associated with greater impulsivity (B = 0.80, p = 0.03). The study group moderated the effect of sleep onset variability on behavior; only the home safety control group exhibited a significant negative relationship between variability in sleep onset timing and emotion regulation (B = −1.28, p = 0.0002). Conclusion: Findings support the importance of consistency in sleep timing and how this may play a greater role in children's behavioral and emotional outcomes than mean actigraphic sleep duration and quality.