In this Sunday's Washington Post magazine is a lengthy and largely laudatory story about marriage and relationship education programs. In those programs, couples typically participate in workshops in which they learn relationship skills such as learning how to disagree without being disagreeable. In my next post, I'll tell you what I think the reporter (Ellen McCarthy) got right and got wrong in her story, and what she overlooked. As a social scientist, though, the first thing I want to know is this: Based on the best available science, do marriage and relationship education programs work?
Results of many such programs have been reported in the scientific journals. There's money for research like that, especially since the days of Wade Horn's tenure as assistant secretary in the Department of Health and Human Services under George W. Bush. Horn was the Director of the Administration for Children and Families (ACF). That division is tasked with promoting the social and economic well-being of vulnerable children and families. Among those helped by ACF are children in foster care, children with developmental disabilities, and children in Head Start. Horn succeeded in getting $100 million redirected from existing programs into marriage education.
I'll describe here the results of two reports. One is a review of 143 studies of the effects of marriage and relationship education programs on couples' communication skills. The other is an executive summary of an ACF project on Building Strong Families, involving more than 5,000 couples recruited from 8 different parts of the country. The 8 versions of the Building Strong Families program also measured couples' communication skills. In addition, researchers determined whether the couples had stayed together or gotten married, whether they had experienced abuse, what the quality of their co-parenting was like, and how involved the fathers were with their children.
Do Marriage and Relationship Education Programs Improve Couples' Communication Skills?
Across the 143 relevant studies, the key comparison is between the couples who did participate in a marriage and relationship education program, and those who did not. On the average, the program participants spent between 9 and 20 hours in the training workshops. In this review of the studies, the authors were primarily interested in the couples' communication 6 months or more after their participation in the training. If there were any effects of the program, did they last at least a half-year?
There were two different ways of evaluating the couples' communications. In one, the couples were videotaped as they discussed some problem. The researchers coded their behaviors, to see, for example, whether they were listening closely or arguing in a fair way. By that measure, marriage education was clearly successful: Couples who participated in the program communicated better than couples who had not participated.
The second way of evaluating the program was by asking the participants directly about their experiences as a couple. They answered questions such as "Does your spouse insult you when he (she) gets angry with you?" On those measures, marriage education was totally irrelevant. Across the experimental and quasi-experimental studies, program participants reported communication that was no better than that of the couples who had not participated in the program.
No one can yet say for sure why the results were different for the two measures. One possibility is that when the couples were videotaped again 6 months later by the same research team that conducted the workshops, they knew how they were supposed to behave and they played nice with their partners. The rest of the time, though, off camera, they behaved no differently than the couples who were left alone.
Does Marriage Education Keep Couples Together, Decrease their Intimate Violence, or Improve Parenting?
In the Building Strong Families (BSF) project, in which studies were conducted in 8 different locations, the participants were unmarried couples who were expecting a baby or just had one. There were three program components:
- Participation in group sessions focused on relationship skills, such as managing conflict, expressing affection, and considering marriage. Couples attended these sessions for an average of 14 hours.
- A "family coordinator" was assigned to each couple, to provide support and encourage the couple to participate in the sessions.
- The couples were offered referrals to services to help them with housing, employment, mental health, and so forth, though they were not obligated to utilize any of the services.
The researchers looked at 14 different ways that the couples who participated in the program could have differed, 15 months later, from those who did not. Those assessments included whether the couples stayed together, the quality of their relationship, the quality of the parenting, and the level of intimate violence.
Here's the bottom line from the scholars who summarized the results from the 5,000+ couples: "Fifteen months after entering the program, the relationship outcomes of BSF couples were, on average, almost identical to those of couples in the control group."
Now let me tell you some of the details (from Table ES.1, p. 4):
- In 4 of the 8 locations (Baton Rouge, Florida counties, Houston, and San Angelo TX), there were no differences whatsoever between the couples who participated in the program and those who did not.
- In Atlanta, the program made no difference in 13 of the 14 assessments; with regard to the use of constructive behaviors during conflicts, there was a positive effect.
- Oklahoma City was the one success story. On 9 of the 14 assessments, participation in the program was linked to positive outcomes, especially with regard to the quality of the couples' relationships. On the other measures, there were no differences. For example, people who participated in the program were no more likely than people who did not to be living together or married 15 months later.
- In the Indiana counties, couples who participated in the program were LESS likely to be living together or married 15 months later. In all of the other ways, the participants did not differ from the non-participants.
- In Baltimore, couples who participated in the program had WORSE outcomes on 7 of the measures than did couples who did not participate. (There were no differences on the other assessments.) Specifically, the Baltimore couples who participated in the Building Strong Families program were less likely to be romantically involved 15 months later, they were less likely to offer support and affection, the mothers were more likely to report severe physical assaults, the quality of co-parenting deteriorated, and the fathers were less likely to live with their children, spend substantial time with their children, or provide financial support to them.
There were differences not just by location but also by race. Specifically, for couples in which both members were African Americans, relationship quality improved. Other couples, though, were more likely to break up if they participated in the program than if they did not.
Here is the summary provided by the authors of the report. It sounds accurate to me:
"The variation in impacts across the local BSF programs and across populations suggests that programs like BSF can have positive effects. However, the results also indicate that these programs can have negative effects on relationships in certain circumstances, including increasing the rate at which couples break up and experience intimate partner violence."
In my next post, I'll discuss the Washington Post story on marriage education. Read it first if you are interested, and see if you think the reporter - who mentions both reports - actually read any more than a one-paragraph abstract (summary) of each.