CSDE Nonexperimental Evaluation

The nonexperimental effort in the CSDE had two main components.

Ethnographic Research

Intensive interviews with African American fathers in Milwaukee explored fathers' knowledge of the pass-through policy and its influence on their payments, their employment patterns, and their involvement with the children. The ethnographic work described fathers who were involved with their children, despite sporadic employment. They had little understanding of the basics of the child support system, generally viewing enforcement efforts as hostile. The report highlighted policy implications emerging from these interviews.

Quantitative Evaluations

Although an experimental design is powerful, it is also limited; it provides information only on the comparison of the policy regimes actually tested and cannot be used to assess the effects of other potential policies. Thus nonexperimental approaches are needed to explore potential effects of pass-through and disregard policy more generally. The CSDE nonexperimental studies included four quantitative evaluations.

Paralleling the ethnographic research, comparative quantitative analyses of administrative and survey data provided information on the fathers of children participating in W-2. The analyses suggest that these fathers have limited economic resources and often face substantial barriers to providing for themselves and their families.

In Volume III, three nonexperimental studies evaluated policy effects using different counterfactuals.

The first nonexperimental study used data from the federal Office of Child Support Enforcement (OCSE) and relied on cross-state and over-time variation in disregard policy. The second study used data from the Current Population Survey (CPS) and policy variation similar to that in the OCSE study. The third study, using Wisconsin Court Record Data (WCRD), considered a different type of variation—changes in the policy regime facing an individual family as the custodial parent moved on and off AFDC. Despite substantial variation in the type of data, unit of analysis, locations, and time periods covered, it proved possible to conduct a rough comparison of the results for similar outcomes: paying or receiving any child support, the amount of child support paid, and rates of paternity establishment.

Taken as a whole, the results of these three studies support the conclusion that increasing the pass-through/disregard will increase the payment and the receipt of child support. It is encouraging that the results from the CSDE experiment have generally been confirmed by nonexperimental studies that rely on national data. Although the CSDE experimental results also suggest increases in amounts of child support paid, the sole nonexperimental study to address these outcomes, using OCSE data, found no effect. Finally, both the CSDE experimental evaluation and the nonexperimental analysis suggest that paternity establishment proceeds more quickly for children eligible for a full pass-through.

Publications

W-2 Child Support Demonstration Evaluation: Report on Nonexperimental Analyses, March 2002

Volume I: Comparative Summary of Quantitative Nonexperimental and Experimental Analyses
Volume II: Fathers of Children in W-2 Families
Volume III: Quantitative Nonexperimental Analyses: Background Reports