Are preventive early childhood interventions effective in improving home environments, as assessed with the HOME inventory (Caldwell & Bradley, 1984)? The authors traced 48 published articles, presenting 56 intervention effects (N = 7,350). The combined effect size on the HOME total score was d = 0.20 (p less than 0.001). Randomized intervention studies were effective, but the combined effect size was limited (d = 0.13). Nonrandomized studies showed inflated effects (d = 0.58). Interventions with middle-class, non-adolescent parents showed higher effect sizes than interventions with low-SES or adolescent samples. Effective interventions used a moderate number of sessions in a limited period and were home-based. Learning Materials, Involvement, and Responsivity showed significant intervention effects. Families in better living conditions profited more from parent education (the Matthew effect). (Contains 4 tables and 1 figure.)