Monday, February 20, 2012

Is Marriage Irrelevant?

I’ve been pre-occupied with some projects and travel, so it’s taken me awhile to swing back to this article about marriage being overrated that I noted in my prior post. I have now carefully read the paper and have some observations to share.

THE RESEARCH IS SOLID

The paper by Musick and Bumpass is well executed. The methods are clear and represent a strong approach to answering the hypotheses the authors set out to address. The researchers are solid and respected, and the paper is about as clear as it can be given the complex nature of the sample and the decisions the authors made in how to use it. The most important thing these scholars did that is different from many prior studies is that they looked at changes within individuals as they moved into either marriage or cohabitation rather than comparing people who are either in one or the other groups already. The latter type of analysis exaggerates the differences found in such studies compared to what these authors did in examining how transition into one or the other (or both) affected the individuals.

There are some important weaknesses in this study, noted by the authors. First, they had to include in the category of “single,” people who could have been in very substantial, high quality, dating relationships but who were not living together. They could not form a really clear contrast for what “single” means, therefore. Also, the data set is quite old now for addressing questions on contemporary relationship patterns (nearly 20 years so). Maybe the most important weakness here is that the before and after relationship transition measurements on these individuals were captured around 6 years apart. There are a lot of interesting things that happen in six years in romantic relationships, including break-ups and how those affect what you are measuring as effects at the second time point. These issues create complexities I will not go into further, but these do create challenges in addressing the hypotheses. Overall, though, I have no major quibbles with the methods, analyses, and, within some boundaries, the conclusions.

THE CONCLUSIONS SEEM SOUND, BUT ARE THEY SURPRISING?

You likely caught the sarcasm in the title of my last posting. That’s not directed at this study, per se, but at the type of headlines the study led to around the world. While I believe the findings are sound, the factoid in the headline is anything but—and such headlines foster ignorance about important aspects of life. I was recently in London, and people there where very well aware of this study and the conclusion in the fast-food type of headlines this study generated. In fact, one of the more clear and concise reports on the study had one of the most unfortunate and misleading headlines: “Marriage is overrated and health and happiness benefits for wedded couples are a MYTH.”

Headlines such as this are digested and overgeneralized in the mind of the average person, contributing to a growing belief among those in the next generation that marriage really does not matter. What does this study actually show? In a nutshell:

1. Quoting from the abstract: “The effects of marriage and cohabitation are found to be similar across a range of measures tapping psychological well-being, health, and social ties.”
2. The strongest, positive effects of moving into marriage or a cohabiting relationship are early on and dissipate within a few years.
3. Moving into cohabiting compared to marriage was associated with increases, on average, in self-esteem.
4. Contrary to some suggestions in past literature about marriage, movement into either cohabitation or marriage was associated with reduced contact with friends and family.

Here is the “so what?” point. Keep in mind the nature of the headlines that this study garnered. The headlines were everywhere and they “educated” millions of young people that marriage provides few benefits over cohabiting. Yet, what the results actually showed is that being in love, and moving closer to a partner—physically, and presumably emotionally—is typically associated with gains personal wellbeing. And, at least in the relative short-run, these effects are similar for cohabitation and marriage. Is that surprising when viewed in that light? Being in love and feeling connected to another is usually associated with increases in one’s personal sense of wellbeing. This is a lot different, though, from the headlined inference that marriage may be irrelevant.

THE GLARING OMISSION (at least in media discussions)

What’s missing? I did not see this issue raised, much less discussed, anywhere in the journal article itself or in any media story related to it: Does the status of a couple’s relationship matter for family stability? Does it matter at all for children?

As it happens, in the time between my last post and this one, the New York Times broke a story about the fact that most births to women under 30 in the US are now to unmarried mothers. This story by Jason DeParle and Sabrina Tavernise, published on February 17th, puts much of this entire discussion in perspective. (DeParle is a highly respected chronicler of poverty and families in America. I encourage you to read that article after finishing this.)

The fact is that a child born to married parents is much more likely to be raised by his or her two parents than a child born to cohabiting parents. For example, a baby born to cohabiting parents is about five times more likely than a baby born to married parents to experience the dissolution of his or her parents relationship by age two (research by Child Trends presented in Galston’s 2008 report, The Changing Twenties). Data presented last year, by Sheela Kennedy and Larry Bumpass at the 2011 Population Association meeting, suggested that, by age 12, a child born to cohabiting parents was roughly 2.5 times more likely to experience the dissolution of his or her parent’s relationship than a child born to married parents. (See also the recent review by Smock and Greenland (2010) in the Journal of Marriage and Family, and a paper by Manning, Smock, and Majumbar, 2004, in Population Research and Policy Review).

For the elegance of simplicity, you cannot beat this one liner by one of the most respected scholars in the study of fragile families, Sara McLanahan, in a forthcoming chapter: “Martial status at birth is a reasonably good proxy for whether children will grow up in a stable household.” (McLanahan, in press)

I will repeat what I wrote about in earlier posts about cohabitation and marriage: Such findings do not mean that married people are better parents or even better able to be good parents. What these findings really mean is that a couple who has married before having children is more likely to have settled their intentions to be a family and be committed to one another for the long-term. Does that guarantee they make it? Of course it does not. But marriage does provide a clearer, more settled foundation for family life. This is, itself, important because so many studies show that children who are raised by their own two parents have, on average, a leg up in life.

So, here’s the real issue at the heart of the media flash about the study by Musick and Bumpass. It gave a strong, consistent impression that marriage really does not have benefits over cohabitation. Whether it does or doesn’t for adults, and on what type of outcomes, and in what circumstances, are interesting academic questions and fodder for social debates. (I would argue that, because of how marriage can be used to clarify commitment, it is also likely to provide a stronger foundation for long-term benefits for adults in comparison to cohabitation.) The really obvious omission from the media discussions is how marriage and cohabitation figure in the stability and early development of children. Marriage seems to matter a great deal in child outcomes. Maybe marriage is not “overrated” as a context for bearing and raising children. The individual wellbeing of adults is important, but it’s not the only thing that matters in evaluating marriage.

*