problems in the mothers’ lives, parenting classes, and classes on life skills, such as communication and decision making (Quint, Bos, and Polit, 1997). Over half of the overall sample at the outset of the study were at risk of clinically significant depression based on CESD scores. The percentage at risk for depression decreased in both the experimental and control groups, but the intervention did not have a significant effect on depression relative to the control group. CESD scores improved more for mothers in the control group than for mothers in the New Chance group, and, at 42-month follow-up, the mothers in the program group were at higher risk of depression than those in the control group. This was hypothesized to result from the greater instability in living arrangements in the program group as well as from raised expectations for improvement that remained unmet. The program group also reported higher levels of parental stress and rated their children as having more behavior problems, an effect that was concentrated among those mothers who were at risk of depression at baseline.

At the 18-month follow-up, children in the New Chance group did have more favorable home environments than the control group based on the Home Observation for Measurement of the Environment (HOME) scale, which evaluates aspects of the home environment, such as cognitive stimulation, safety, emotional support in mother-child interactions, and harshness of discipline. By the 42-month follow-up, however, this effect persisted only for the subset of mothers who were not at initial risk of depression. An in-depth examination of parenting behavior and child development in a subset of families, using coded videotaped observations of interactive tasks for the mother and child, showed that, at an average of 21 months after enrollment, the program had positive effects on parenting, but these were not accompanied by a difference in child developmental outcomes measures (Zaslow and Eldred, 1998).

Moving to Opportunity: In a randomized housing mobility experiment as part of a demonstration project called Moving to Opportunity, families living in high-poverty public housing projects in five U.S. cities were given vouchers and counseling to help them move to private housing in neighborhoods with lower levels of poverty. The study population was comprised predominantly of African American or Hispanic female heads of household with children. The families were offered housing vouchers, allowing them to move to neighborhoods that were safer and had significantly lower poverty rates than those of control families who were not offered vouchers. There were significant reductions in psychological distress and in the probability of a diagnosis of a major depressive episode, with systematically larger effect sizes for the group experiencing larger changes in neighborhood poverty rates (Kling et al., 2004; Kling, Liebman, and Katz, 2007).

Taken together, these findings support the principle that policies and