Exhibit 3. Marginal effects of neighborhood collective efficacy on adolescent depressive and anxiety symptoms by sub-groups.
| Depressive symptoms scale | Anxiety symptoms scale | |
|---|---|---|
| By neighborhood socio-economic disadvantage | ||
| High (1 standard deviation above national mean) | -.070 (.023)** | -.075 (.023)** |
| Average (national mean) | -.089 (.029)** | -.057 (.029)* |
| Difference | .019 (.020) | -.018 (.020) |
| By family income | ||
| Low (100% federal poverty line) | -.075 (.024)** | -.081 (.024)** |
| Moderate (300% federal poverty line) | -.079 (.029)** | -.057 (.029)* |
| Difference | .005 (.023) | -.024 (.023) |
| By child gender | ||
| Female | -.081 (.032)* | -.095 (.052)** |
| Male | -.066 (.030)* | -.052 (.030)+ |
| Difference | .016 (.042) | .043 (.042) |
Source: Authors' analysis of data from the Fragile Families and Child Wellbeing Study. Notes: Marginal effects for each sub-group are obtained by interacting sub-group variables with the neighborhood collective efficacy scale. Continuous variables are used for neighborhood disadvantage (national z-score) and family income (income-to-needs ratio); marginal effects are estimated at the values shown in parentheses. Models control for all socio-demographic and mental health history variables present in Exhibit 1, including sample city at birth.
p<.10
p<.05
p <0.01