The central question addressed by this article is the effect of state-level marriage initiatives on divorce and childhood poverty rates. State divorce rates have been problematic for researchers because of variation across states in the way they are compiled. This research takes a different approach, measuring instead the prevalence of divorce rather than the number of divorces granted in a given state or year. The authors use this indicator, derived from Current Population Survey data, as an outcome measure in a test of marriage initiatives, and as an independent variable in a childhood poverty analysis. The quasi-experimental design employs time-series and cross-section regression analysis. Results show a significant negative effect from marriage initiatives on divorce prevalence, and a significant positive association between divorce prevalence and childhood poverty rates.