Monday, February 17, 2014

About Those Controversial “Marriage” Programs Funded by the Government


The fur is flying once again around questions of the wisdom and effectiveness of the government funded efforts to increase and strengthen two parent families through the use of relationship education strategies. I believe that things have heated up, in part, because this issue about relationship education has now intersected with this year’s political debate about the nature of poverty and income inequality. That is too bad because politically-infused rhetoric is not conducive to the type of smart policy discussions that need to be sustained to actually make a difference in the lives of real people.

In this post, I am going to link to four articles that exemplify the issues being debated right now. My goal here is to focus in on a few specific points raised in the four articles, particularly related to the subject of government funded initiatives providing relationship education. Links are provided if you want to find and read the pieces I reference.

First up, Kay Hymowitz wrote a broad, thoughtful piece entitled How Single Motherhood Hurts Kids (New York Times on February 8th, 2014). She covers a lot of important ground. I like how she challenges some of the deeply held viewpoints of liberals and conservatives. I also like how she embraces the bi-directionality of how poverty impacts families and how romantic and sexual relationships impact poverty. I think, for some people, the latter point may get uncomfortably close to how marriage might play some causal role in poverty. For example, romantic and sexual relationships impact childbearing and also the odds of succeeding in marriage, which, in turn, all influence poverty. I believe that discussions about poverty and family policy/programs that do not embrace the obvious and complex ways that causation flows both ways are unproductive; it is better to embrace than ignore complexity when it is relevant.

Next up is a piece by Annie Lowrey, in the New York Times Magazine: Can Marriage Cure Poverty? (February 4th, 2014). Lowrey’s piece is excellent though I think it is slightly marred by the way she caricatures conservatives. But she gets some crucial points right. First, non-marriage is linked with poverty in some profound ways, and the only real debate on that subject is not if but how. Second, she acknowledges that there is some positive news in the recent, large-scale government funded evaluations testing relationship education strategies with unmarried couples in poverty. Third, she directly addresses the topic of the unbelievably dumb disincentives within government policies that punish disadvantaged couples as they try to get ahead. That’s a topic for another day, but the disincentives are probably growing and that makes me fear for our future.

On the other hand, Lowrey tilts pretty far to the unidirectional viewpoint that poverty drives problems in family structure and marriage without addressing the ways romantic partnering and sexual relationships contribute to disadvantage and risk. Hymowitz, as I noted above, digs in on this important issue while also addressing aspects of the complexity of marriage for some among the disadvantaged.

I raise these particular issues here because I want draw attention to one type of relationship education that has been utilized in some of the government supported efforts that gets little or no attention in the typical discussions. There have been a lot of services focused on helping individuals, not couples, make relationship choices that are designed to foster their odds of achieving their own life goals (including, eventually, marriage if desired) while also improving the lives of their children. Such individually-oriented relationship education is founded on the awareness that the economic, mental, and physical wellbeing of adults is deeply impacted by what happens in their love lives. Anyone close to the challenges confronting those with immense disadvantages recognizes that landing with a wrong, poorly matched, or dangerous partner can completely upend any other progress in life. Does any of this type of work get acknowledged in public discourse about government funded relationship education? Rarely. On the ground, I see liberals and conservatives united around the obvious rationality of such strategies. We do need robust evaluations of such services (they are pretty new and innovative), but the goals are important and some of the work is immensely thoughtful. 

Instead of showing any awareness of some of the innovative work being attempted, what is often suggested or implied in media stories is that government funded relationship education efforts are focused primarily on getting people in poverty to get partnered-up in marriage because that will magically cure all the ills of poverty, family functioning, and individual wellbeing. Right. Of course it will. And, of course, that’s how most of the people working on the ground in such efforts think. Magic beans, Jack.

If your sarcasm meter is not functioning, let me clarify: I’ve been around this work for a long time. I do not know anyone deeply involved who thinks so simplistically about this work. In fact, on the broader subject of thoughtfulness, I have been fortunate to be in scores of meetings with diverse scholars and experts grappling with complex issues about helping people—meetings where what happens is a lot more substantive than implied, for example, by a critic in one of the upcoming pieces I link to below who refers to “mucking around in people's lives.”

Related to complexity and thoughtfulness, I also want to suggest that anyone who wants to be taken seriously in discussing relationship education and government policies needs to think carefully about the types of outcomes that would be positive in different situations. For example, there are many cases wherein a good outcome of relationship education amounts to a damaging or destructive relationship coming to an end—not ending in marriage. Marriage is a public and private good, desired by many, but that does not mean any marriage to anybody. Strategists and workers on the ground level seem to have wholly accepted the goal of fostering healthy relationships and marriage; this fact seems to elude most critics of the efforts. This point also highlights just one of the reasons why it may be quite challenging to interpret simple statistics about marriage and divorce rates at the macro level as part of these discussions. 

Moving on to the other two articles. Both deal evidence of effectiveness of services provided to couples in some of the most intensive type of government efforts, with a particular focus on two large, federally funded, multi-site studies of program impacts for comprehensive services provided to disadvantaged couples who are either unmarried and having a child (The Building Strong Families Study: BSF) or married with children (The Supporting Healthy Marriage Study: SHM).

One lengthy piece was written by Thomas Bartlett of The Chronicle of Higher Education: The Great Mom & Dad Experiment (January 20th, 2014). There are some elements of Bartlett’s piece that I take issue with, but, for this type of policy discussion, the piece seems to me to be relatively fair. He does a good job describing a range of complex issue and areas of serious disagreement among some of those who have thought about these issues. Another piece also covers a good deal of ground. It’s by Emily Alpert Reyes of McClatchy News: Federal Programs to Improve Marriage Don't Work (February 10, 2014). As with others here, there are complex issues raised in this piece that I may come back to in a later piece, but I mention both of these pieces here because I want to wrap up this post by making three points related to those two large government studies noted above. Many commentators of these two studies discuss the results as if there were no impacts at all. Sometimes, modest impacts are acknowledged but with an assumption that, surely, the government would not and should not continue to fund programs in any area where there were similarly modest findings in large, randomized trials.

In making these points, I take nothing from the fact that the actual findings and implications are complex.

1.  The one site among eight in the BSF study (Family Expectations in Oklahoma City) that successfully delivered a serious “dose” of the services to the participating couples demonstrated a range of statistically significant, positive impacts at the 15 month follow-up. Further, there was a statistically significant impact on family stability (a 20% increase) at the 3 year point follow-up. It is true that there was an absence of evidence for other types of significant impacts. However, obtaining a relatively large impact on a core, long-term outcome of policy significance (family stability) seems pretty important. Such an outcome is rare in this type of study.

2. Those conducting the large, SHM study, learned a great deal from the BSF study. As a result, all eight sites in that study delivered a substantial dose of the services to most couples in the program groups. What did they find? The results from the 12 month follow-up showed statistically significant, though small, impacts on a number of variables related to marital quality. (There are findings coming out soon on the 3 year follow-up.)   

3. Most large-scale studies of most government programs find no impacts. In 1987, a sociologist deeply acquainted with government evaluations, Peter Rossi, came up with The Iron Law of Evaluation, which he stated thusly: “The expected value of any net impact assessment of any large scale social program is zero.” Translation? In most such studies, one can expect no significant impacts; yet there were interpretable, significant impacts in the two studies noted above of these early efforts.  

I could list a wide variety of government programs and services that have poor evidence of effectiveness but that, nevertheless, receive vastly greater funding than the nascent attempts to help individuals and couples in their relationships. There are also many smaller studies, including some with very strong methodology, that show consistent evidence of positive impacts from relationship education. (Not my topic today.)

Am I pleased with modest impacts of those two large studies? No. As someone who works in this field, I see the current studies as promising, but like many others, I want to see us learn from current efforts and strengthen impacts.

In all of this, I have been puzzled by calls to defund what seem to me to be, by historical standards, promising findings coming early in a relatively new arena for government involvement. Why don’t the same critics explicitly call for defunding services and programs that receive hundreds of times the funding, services that happen to have long trails of meager or non-significant findings? I cannot read the minds of others, but I think some of the answer comes back to where I started. I think some of the energy currently rushing into this space, but certainly not all, is really about other issues.   

* *

Disclosure:  I think that just about everyone (but not everyone) writing and commenting on these matters has some type of bias or values animating their arguments, and it is worth considering what those may be when thinking through the points being bandied about. I try to be very above board, so here is my disclosure. You should know that Howard Markman and I (and a host of colleagues) have developed a variety of relationship education approaches that were used in some of the sites in the BSF and SHM studies and that are also used in a number of the projects funded by the government around the U.S. I receive income from our company called PREP. Further, I have been a long-time adviser for the efforts in Oklahoma. Of greater weight within me is the fact that I do believe in trying to build on promising studies, practices, and program models in these areas I focus on here. You are entitled, of course, to disregard any of my viewpoints based on these facts, but I hope those with serious interest would grapple with the ideas and consider where there may be inherent merit.


Reference for Rossi:  Rossi, P. H. (1987). The Iron Law of Evaluation and other metallic rules. Research in Social Problems and Public Policy, 4, 3-20.  (for an interesting piece on this, click here