Lessons From 15 Years of the Healthy Marriage and Relationship Education Initiative

Since 2006, the U.S. Administration for Children and Families (ACF) has provided significant grants to community organizations to provide educational and support services to lower income couples and young adults to help them form and sustain healthy relationships and stable marriages. This federal support for the Healthy Marriage and Relationship Education initiative over the first 15 years had a price tag of about $1.25 billion. Elsewhere, I estimated that these relationship education efforts have reached nearly 200,000 individuals a year at an overall cost of about $400 per participant. Most of the participants were lower income and non-white. And participants report highly valuing these programs.

From the beginning of this controversial policy initiative, ACF has also funded a serious evaluation process of the effectiveness of these kinds of programs. During this time, I have carefully followed this policy initiative and reported on it and the effectiveness of these educational programs. (See here and here and here and here.) I have been both a friendly critic and a skeptical enthusiast. That is, I think this policy experimentation deserves a fair test (other scholars have disagreed with me). But I know these kinds of policy initiatives—especially in the early years—struggle to show positive effects. 

It’s unusual for the federal government to launch a serious evaluation effort right from the start of a new social policy initiative. But the architect of the initiative, Dr. Wade Horn, former Assistant Secretary for Children and Families under President George W. Bush, knew that the initiative would be controversial and reasoned that the best antidote for skepticism is good data. The subsequent Obama administration’s strong push for evidence-based policy reinforced and extended the emphasis on evaluation for this new policy initiative. 

Recently, I assembled all of the publicly available evaluation reports of these programs for adult couples (through May 2020 near the end of the third wave of funded programs) and systematically reviewed them to get a big-picture read on how effective or ineffective this social policy initiative has been.¹ This meta-analytic study was recently published in the journal Family Process (in a special issue that highlighted other studies of specific ACF-funded family programs). This study focused primarily on the 32 reports with the most rigorous designs: randomized controlled trials that randomly assigned couples to treatment or control groups.²

So, what have we learned from this body of evaluation research over the past 15 years? Overall, our systematic review found a range of small but significant positive effects for couple relationship quality (d = .114) and relationship skills (d = .132). We found smaller—but still statistically significant—positive effects for mental health (d = .074) and coparenting (d = .033). Unfortunately, we found non-significant effects for relationship stability. That is, participation in these programs, on average, does not appear to help couples stay together, although there were a few programs that reported small stability effects. Also, we found non-significant effects of these programs for parenting outcomes. Some of these programs target parenting skills as well as couple relationship skills, but many focus just on couple relationships, so the lack of improvement in parenting skills isn’t too surprising. No significant effects were found, on average, for child well-being, either, although it may take longer for these child effects to emerge. However, a handful of studies (including ACF-funded programs and non-ACF-funded programs) have shown how couple relationship education programs can impact child well-being through improved couple relationships and parenting.³

These “average” program effects, however, masked some noteworthy variations. For instance, most of the earlier programs (2009-2014) struggled to produce significant effects, but later programs (2015-2020) had stronger effects (d = .19).⁴ Also, programs that included both married and unmarried couples in the same classes (rather than targeting one group or the other) produced stronger effects (d = .23). And studies that reported effects based on participants who had a strong dose of the program—rather than including all participants regardless of their program dosage—had the strongest effects (d = .34).  

The results of our systematic review of these programs can bolster both critics’ concerns and supporters’ hopes. Critics will point to the lack of effects on key outcomes such as relationship stability and the small effects for other relationship outcomes. They may argue that this policy experiment has failed to achieve its ends and should be retired. But caution is needed here because the effect sizes observed in this review are essentially equivalent to the effects seen in other federal social policy initiatives (Head Start, for example, and home-visiting programs), some funded at much greater expense over a longer period of time. In fairness, an argument to terminate ACF’s Healthy Marriage and Relationship Education policy initiative would need to consider a similar fate for other family policy initiatives. (And some recent research on educational interventions relevant to federal policy suggests that the effects seen for these couple relationship education programs are generally above the median observed effect.⁵

Proponents could argue that this new policy initiative is young compared to those other initiatives and more recent research appears to be producing stronger effects than the first wave of evaluation work. There were concerted efforts by ACF over the past 15 years to communicate lessons learned with other programs through various channels. These efforts may be paying off. And with a recent new round of funding ($75 million a year, 2021-2025), 30 community organizations continue to build on this rapid learning process and evaluate their efforts. 

The Healthy Marriage and Relationship Education policy initiative is not as controversial as it was 10 years ago, perhaps due in part to a serious, sustained evaluation agenda.⁶ But given the problem it addresses—the effects of relationship instability and lower relationship quality on disadvantaged children’s well-being⁷—continual improvement in these programs is needed. I look forward to the next wave of evaluation research that can help guide program improvement and steer policy in fruitful directions. 

Alan J. Hawkins is a professor of family life at Brigham Young University who has published extensively on the effectiveness of relationship education. 

---

1. I’m grateful to a dedicated team of student research assistants who worked with me on this study: Hawkins, A. J., Hokanson, S., Loveridge, E., Milius, E., Crawford, M. D., Booth, M., & Pollard, B. (2022). How effective are ACF-funded couple relationship education programs? A meta-analytic study. Family Process. 

2. 19 other reports evaluated treatment groups over time without a control-group comparison.

3.  Cowan et al. (2022) found that reductions in both parents’ destructive couple communication were associated with reductions in negative parenting qualities and children's behavior problems; Feinberg and Jones (2018) found that child behavior problems were impacted by a set of changes in both coparenting and parenting outcomes; Pruett et al. (2019) found a significant pathway from reduced couple conflict to reduced harsh parenting then to reduced child behavior problems; Zemp et al. (2016) found significant direct paths between improved couple relationships and child behavior problems (without exploring coparenting specifically); Sterrett-Hong et al. (2018) found significant direct paths between decreased couple conflict and improved child mental health (without exploring coparenting). For a meta-analytic review of these and other similar studies, see: Hawkins et al. (2022). Do couple relationship education programs affect coparenting, parenting, and child outcomes? A meta-analytic study. Journal of Child and Family Studies.  

4. These earlier studies were dominated by 8 program sites in the Building Strong Families study, sponsored by the ACF Office of Planning, Research, and Evaluation, that showed few significant results. 

5. See: Kraft, M. A. (2020). Interpreting effect sizes of education interventions. Educational Researcher, 49, 241-253. 

6. The legalization of same-sex marriage removed the institution of marriage from the front lines of the culture wars, and no doubt this likely contributed to a reduction of the controversy surrounding the policy initiative. 

7. See Sawhill, I. V. (2014). Generation unbound: Drifting into sex and parenthood without marriage. Washington D.C.: Brookings Institution.