Table 4.

Odds Ratios for Prediction of Child Well-being From Final Logistic Regression Model*

Social Capital Index (0 to 5)0.25291.29 (1.04,  1.59)
Mother figure has high school diploma/GED0.23111.26(.74,  2.15)
Family income category0.11541.12 (1.01,  1.25)
Maternal depression (CES-D ≥ 16)−0.54870.58(.33,  1.03)
Foster care placement−1.83630.16(.03,  .74)
Indicator for Baltimore§ 0.13451.14(.62,  2.10)
Indicator for San Diego§ −0.86180.42(.18,  .97)
Indicator for Seattle§ −0.58990.55(.28,  1.11)
• * χ2 test comparing fitted model (intercept plus covariates) to the null model (intercept alone): χ2 = 35.3, (8 df), P < 0.0001.

• Logistic regression produces the OR for each variable adjusted for all other variables in the model.

• OR for ordinal variable indicates change in the odds of doing well for each unit change in predictor variable. Social capital units extend from 0 to 5 reflecting the number of indicators present. Family income units increase in $5000 increments from <$5000 to >\$50 000 per year.

• § Indicator variable for site has a value of 1 for children belonging to the specific site, and a value of 0 for other children.