The purpose of this study was to test the hypothesis that quality improvements in early childhood centers experience diminishing returns as the quality of the classroom rises with regards to concurrent socio-emotional outcomes. This hypothesis lies at the core of Scarr's argument that public policy should concentrate on improving low quality settings rather than improving settings that already have acceptable quality. The study detected sizeable effect sizes linking process quality in the good to excellent range with reduction of existing socio-emotional risk factors (d=0.51) and prevention of the emergence of new socio-emotional risk factors (d=-0.41). These effect sizes are substantially larger than those reported by other studies investigating quality environments in the poor to good quality range (d=0.16), and larger than Durlak and Wells meta-analytic effect size for universal preventive interventions (d=0.35). Therefore, the hypothesis that as quality increases the benefits for children increase but at a diminishing rate was rejected for concurrent socio-emotional outcomes in urban populations.