Inside the Black Box of Interactions Between Programs and Participants: Re-conceptualizing Subgroups for Fatherhood Program Evaluation. IX. CONCLUSIONS


Fatherhood programs are designed to support fathers and foster positive behavior change in their roles as parent, partner, and provider. Service-user typology research suggests that the same fatherhood intervention might have differential impacts depending on how well the intervention addresses the number, nature, and severity of challenges that individual fathers face. Behavior change theories and public health research employing audience segmentation methods suggest that the same fatherhood intervention might have differential impacts for fathers who are more versus less ready, willing, and able to effect changes in their lives.

These hypotheses are largely untested. Research is needed on the extent to which theoretically relevant but empirically understudied psychosocial factors are important predictors of behavior change among low-income fathers enrolled in fatherhood programs, and if and how these factors shape the likelihood of their benefiting from these programs. Measurement research is also needed to better understand how best to (1) operationalize these constructs with measures that have been validated for use with fathers, especially low-income and culturally diverse samples of fathers; (2) code these variables (including defining cut-points for continuous variables) for use in creating subgroups, and (3) combine multiple theoretically relevant variables to yield promising subgroups.

Future research could explore:

·      Which theoretically relevant but empirically understudied psychosocial factors shape fatherhood-related outcomes?

·      What factors determine behavior change among low-income men? What are prerequisites of change? That is, what conditions must exist-at the individual, interpersonal, and contextual levels-before behavior change can be expected?

·      What other factors foster or impede behavior change among low-income men?

·      To what extent might these factors serve as useful variables for creating subgroups for use in impact evaluations that use an experimental design?

·      Are there established measures that could be used to assess these factors at program intake?

·      Are there additional, innovative methods for assessing these factors at program intake-such as the judgments and impressions of intake staff?

Regarding the creation of subgroups, future research could explore:

·      Does the level of risk matter? Where on the risk-need continuum might subgroup impacts be expected?

·      Do particular combinations of certain risks and readiness levels matter? What levels of need, coupled with what levels of readiness and willingness to change, might be needed for subgroup impacts to be expected?

Findings from program evaluations that examine subgroups based on factors other than demographic characteristics may aid in program development. Such findings can help fatherhood program operators think about how best to design interventions, decide those to target for program services, and assess fathers' needs at intake in the following ways:

·      Design interventions. The most effective programs may need to target not only the outcomes ultimately sought for fathers-namely, viable employment, improved parenting, and improved partner relationships-but also aspects of their lives that may influence these outcomes, such as their relationship with their child's mother, their experiences in the family of origin, and their stress and coping strategies. In addition, fatherhood programs may need to do more than provide information and skills; they may need to actively support the behavior change process. This would require explicitly addressing the conditions theorized to be necessary for change, such as the father's perceived need and ability to change, his willingness to change, and his readiness to change. Programs that provide emotional supports (such as empowerment, encouragement, and opportunities to share concerns with peers) along with practical services and supports (such as goal setting, action planning, and links to necessary services) may more effectively move fathers from contemplation to preparation to action (Prochaska et al. 1992).

·      Target services. Measures of service needs and psychosocial determinants of change such as readiness to change (Prochaska et al. 1992) might help fatherhood providers better recruit and enroll fathers most likely to benefit from program services. Fathers who (1) want to improve their employment prospects and gain viable, stable employment, (2) wish to be more involved in their children's lives but lack the skills to be an effective parent and co-parent, (3) wish to improve their relationship with their child's mother and/or a current romantic partner, and (4) are at a point in their lives where they are ready, willing, and able to make these changes may be the ideal candidate for a fatherhood program. By contrast, for example, unemployed fathers with a solid employment history and few other service needs may not need the intensive, comprehensive services typical of fatherhood programs, so these fathers may be better served by directing them to job search and job placement services in the community. At the other extreme, fathers who not only lack employment, parenting, and relationship skills but who face severe obstacles such as substance abuse, mental health issues, or domestic violence may need to address these issues first before they can successfully participate in and benefit from a fatherhood program.

·      Assess fathers' needs at intake. Baseline measures used in program evaluations could be adapted for use on intake forms to help providers identify fathers' service needs and risks, strengths and protective factors, barriers to participation, and impediments to behavior change. This information could help providers better target, triage, and sequence the appropriate set of services.

To be sure, subgroup impact analyses present methodological challenges and concerns-sample size requirements and multiple comparisons chief among them. Evaluation sample sizes must be large enough to ensure sufficient statistical power to detect impacts in subgroups (or in the subpopulation comprising the evaluation sample), and examining numerous subgroups within a single sample runs the risk of finding spurious impacts purely by chance. Despite these and other practical challenges, it is useful to think about the various "kinds" of fathers who may enroll in and might benefit from fatherhood programs and to seek strategies for identifying which of these fathers may benefit most from intervention. Opening the black box to understand what works for whom and under what circumstances can help policymakers and program providers make more effective use of resources.

View full report


"rpt_insidebb.pdf" (pdf, 1.35Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®