The New Family Structures Study and the Challenges of Social Science

It should be obvious that the family—and marriage in particular—is under a great deal of popular and political scrutiny. There is a tug-of-war in the social sciences over what we know with confidence about the contemporary family, which has not changed a great deal, and what can be said about it (and who can say it).

But while scholarly norms, language, and state and federal family law can shift with remarkable speed—as they have in the West over a few short decades—the data collected from thousands of regular people who are living, or have lived, in all manner of household structures and family experiences have not changed nearly so fast. The empirical evidence still documents the pivotal importance of family stability and well-being in social life. Where we see children and young adults flourishing best, we are apt to see stable families as well. The family remains the first building block—necessary but not sufficient—in any large, decent, dynamic, and sustainable social order.

But is the family a social construction? Can it be what we wish for it to be? Much is made of the socially constructed nature of the family. Indeed, calling a social practice or pattern “socially constructed” is a popular theme in contemporary sociology, and to be sure it has some merit. Most aspects of social life, and the institutions within or under which we live, are social constructions, by which we mean that people made them, and for them to continue we must remake and reinforce them regularly. But calling the family simply a social construction is an effort to undermine the reality of it, and to suggest that its structure and functions could be radically different, or not exist at all, and we would be fine.

There are common and permanent characteristics of the family; there is a structure to it that is historically reliable and that—when functioning competently—cannot be topped in its ability to accomplish six key tasks: (1) families make things, (2) they reproduce themselves (sexually), (3) they socialize their children, (4) they do things together, meaning recreation, (5) they guard sexual access to family members—sometimes very tightly, while in other settings quite carelessly—and (6) they take care of ill or aging family members. Historically these functions varied little, even as we recall or witness examples of individuals—and even whole communities—that seemed more or less adept at each of these. For example, contemporary Americans often outsource the care of elderly family members due to necessity (such as employment obligations) or excessive expense. While plenty of families prioritize shared evening meals, other families seldom do so. Some parents oversee their adolescent children’s dating activities carefully—or curb them altogether—while other parents show little interest in discerning their youths’ romantic and sexual activities. Some outsourcing, like sending children to day care, can and may be uneventful in its immediate consequences, while other outsourcing, like reproduction (e.g., IVF, surrogacy) or sexual control (e.g., paying little attention to the romantic lives of one’s adolescent children), is commonly perceived as far more problematic and risky. 

Other hallmarks of the family are increasingly in question, including marriage itself. Much of the damage is self-inflicted. Marriage rates continue to drop precipitously among young adults ages 25 to 34—the historic “sweet spot” for marrying. Between 2000 and 2012, the share of young adults ages 25 to 34 who reported being married declined 13%, from 55% to 42%.[1] During the same period, the percentage who have never been married increased sharply, from 34% to 49%. Thus in a rather remarkably short period of time, the share of young adults in the United States who have never been married now well exceeds those who are married. This is stunning.

Alongside this flight from marriage is the legal and political debate about the structure of the marital union itself. Is marriage socially constructed to such an extent that its basic structure is malleable, subject to collective will? Can a marriage be comprised of two men or two women? Few believed so prior to 30 years ago, and only in the past few years has majority opinion in the West shifted.

But the roots of where we are today lie in the development of reproductive technology more than anything else, certainly more than the emergence of new rights. I am always a big believer in the idea that technology drives structural changes, which echo in culture. Culture, basically, is the last to know or to feel social change. British social theorist Anthony Giddens said in his landmark 1992 book The Transformation of Intimacy that the revolution began with effective contraception, which, in his words,

meant more than an increased capability of limiting pregnancy. In combination with the other influences affecting family size . . . it signaled a deep transition in personal life. For women—and, in a partly different sense, for men also—sexuality became malleable, open to being shaped in diverse ways, (something one could cultivate), and a potential “property” of the individual. . . . Now that conception can be artificially produced, rather than only artificially inhibited, sexuality is at last fully autonomous.[2]

Fully autonomous: meaning separated from embeddedness in strong (or even romantic) relationships. Giddens asserts that more change should be expected to follow: “Once there is a new terminology for understanding sexuality, ideas, concepts and theories couched in these terms seep into social life itself, and help reorder it.” We are in the middle of that reordering right now. 

Giddens goes on: “The ‘sexual revolution’ of the past thirty or forty years is not just, or even primarily, a gender-neutral advance in sexual permissiveness. It involves two basic elements,” one of which, he asserted, “is the flourishing of homosexuality, male and female.”

So Giddens draws an arrow from (1) contraception to (2) sexual malleability and (3) the flourishing of homosexuality. I didn’t say it, but it makes sense.

What Westerners have witnessed over the past several decades, then, is not so much the social construction of marriage or family simply toward different plausible ends as a product of political will, but the reality of technology-driven social change, which brings about new social structures, which then act back upon the phenomena that gave rise to them.

What has emerged in the domain of sex and relationships is not simply different norms or values but re-structured realities around the intimate life. We now value and express what Giddens calls the pure relationship. This is

where a social relation is entered into for its own sake, for what can be derived by each person from a sustained association with another; and which is continued only insofar as it is thought by both parties to deliver enough satisfactions for each individual to stay within it.[3]

What Giddens said—back in 1992—is that in a time wherein childbearing can be avoided amidst a rich and varied sexual life, what would likely result is what in fact has resulted nearly 25 years later:

Strong norms about emotional and physical satisfaction in relationships

Expectations of paired sexual activity emerge quickly in budding relationships

Sexual exclusivity is no longer assumed but rather the subject of negotiation

More shorter-term relationships, together with perceptions of commitment “phobia”

Plastic sexuality—sexual interests and directions are explored, shaped, and remodeled

An obsession with romance among many, and yet stability seems increasingly elusive

To many younger Americans, these are all self-obvious realities—the way things are.

Giddens cautions, however, saying that “it would certainly not be right to suppose that childhood has remained unaffected” by the new realities in parental romantic relationships. This is an understatement. It is also a very passive sentence, but a very loaded claim. So, how can we know?

Prior to 2010, very little population-based data analyses of same-sex households with children were available. Several, however, have emerged since then; data collection attention here has risen notably. But what these data say about same-sex households with children, and how exactly such data ought to be analyzed, has been the subject of significant emotional and political contest.

The publication of my 2012 journal article on the comparative outcomes of adult children who report having had a parent in a same-sex relationship stirred a hornet’s nest, to put it mildly.[4] Using data from nearly 3,000 people, 248 of whom reported witnessing a parental same-sex relationship, I documented that those for whom that was the case were more likely to have endured a more difficult childhood and adulthood than kids who grew up with a stable, married mother and father. It’s hardly rocket science, really. My purpose in doing that study was to evaluate the claim that there are “no differences” in outcomes between kids who grew up in same-sex households and kids who grew up with a married mother and father. 

Not everything is different between those who grow up in a stably intact household and those who do not, which is to be expected. But across different types of measures, the benefits of having grown up with a married biological mother and father who are still together (or were until one or both passed away) were obvious.

Vengeance was swift. There was a near-instant demand for the scholarly journal to retract the publication (which has not occurred, because the results are neither erroneous nor fabricated), a letter of 200 scholars against me within two days of its publication, hostile blogs updated daily for weeks, 15 freedom-of-information-act requests for all my correspondence related to the project, hostile people emailing all of my colleagues and my dean denouncing the University of Texas for employing me, etc. The journal editor, who thought perhaps this article would stir the pot a bit, was virtually assaulted online. His university was sued. 

I learned that stating apparent but unpopular things on sensitive subjects can be expensive.

On the bright side, it turned scholars’ attention to the limitations of the social science in this area. My study of course had limitations. I was unable to track the lives of very many children who grew up in long-term same-sex households because stability was uncommon. Some suggested it was unfair of me to compare stability with what amounted (in reality) to instability. But what if instability is a common experience for children in such households?

It reminds me of a story recently told to me by a University of Colorado professor, who lives next door to a woman—a mother—who has been through three same-sex relationships in the past few years that they have been neighbors. The professor told me that he will be out in his backyard playing with his kids, only to notice the neighbor’s son looking over the fence, watching, wanting. The boy’s mother, aware of this, said to the professor, “I’m doing the best I can.”

Some suggest this mother’s situation will invariably improve with greater social acceptance and legal rights. It’s hard to say. Hers, however, is not the popular face of the movement. Instead, we see the well-adjusted child of a 30-something, upper-middle-class mother and her partner, a child born by assisted reproductive technology. That is what scholars, judges, and the media demand as a comparison category. It is as if they have watched too much television and insist the world look like the ideal. I admit, I did not have the numbers of stably-coupled female households to assess them in a very large, nationally-representative study. It raises the question of whether the average experience within such households is something quite different from the ideal.

Meanwhile, I have seen no documentable change in how same-sex couples come to have children, and the population most apt to use artificial means—white women—are the population statistically least interested in having children, according to data from the National Survey of Family Growth.  And ART is expensive and not that common.

Even this—the matter of “what counts” as a same-sex household, especially one with children—is far from settled science. Should we only compare those households with two parents? How long should they have been together? Must the children know about their parents’ sexuality? 

And what should social scientists do about the household instability that has been—and may still be—quite commonly associated with same-sex households? Ignore it? Denounce any suggestions that it is relevant or perhaps even common? Or—as typically has been the case—control for it in regression models relying on the assumption that instability is randomly distributed when it may not be? 

I hold, and continue to do so despite opposition, that the answers to these questions are not simple. I have been told repeatedly that my study violated protocols in the study of same-sex household outcomes. But there is no obvious protocol. 

In reality, I suspect every scholar who works with population-based data about same-sex couples, persons, parents, and households knows that the underlying story varies little across datasets. What varies widely is the manner in which different scholars—myself included—have chosen to present, analyze, and interpret these data.

Despite competent analyses and plausible interpretations of the population-based data available, economists Joseph Price, Douglas Allen, and myself were labeled “fringe scholars” and largely unbelievable by a Michigan federal court judge. But are the arguments we raised in Michigan really so odd or unseemly? I frankly do not see it. These are legitimate questions, but we are being criticized for not doing the “decent thing” and keeping quiet.

The New Family Structures Study, however, is hardly the only data source that suggests there may be problems. New analyses of the many waves of the National Health and Information Survey are noting these problems, especially regarding emotional well-being measures.[5] Expect social scientists to do their best, however, to paper it over or explain it away by adding more and more control variables. If they do not succeed, there is a blistering blogosphere happy to remind people of the costs of disloyalty to the cause.

How does one do sociology in such an atmosphere? Well, it’s not easy. Sociology assumes a particular good that formally goes unspoken. It has its shared compass.

The wider gay-rights movement has been a very popular one among social scientists. Few social scientists have, or express, any concerns about making civil marriage available to same-sex couples. Likewise, few express concerns about possible ramifications of esteeming—not just tolerating, as in the case of single parenting—the notion of motherless or fatherless children in households headed by same-sex couples. They have their professional credibility to worry about.

I have learned some things, though: first, you can survive. You can even thrive, amid hostility. I have learned there is still academic freedom. And when you realize that your professional status within sociology has likely peaked, it is strangely freeing. No longer do I concern myself with what my peers think, wondering how an op-ed or an amicus brief will affect my next faculty evaluation. It is actually what academic freedom and tenure are for. I think we forget about that. Too many of us consider it basic job security that is afforded to no one else in the world except us, because we somehow deserve it. It has a purpose, however. It was created to sustain the pursuit of the truth, wherever it happens to lead.

We sociologists operate under the false assumption that a sociological perspective on human social behavior will invariably lead us to commend more and greater freedoms, rights, and equalities, regardless of the domain or the question. Rather, that is the arc of a Western sociology devoid of a compelling anthropology of personhood and community. What are people? is a question we ought to ask. The dominant answer, however, is whatever—and whomever—they wish to be. What ought people do? is answered with “whatever they wish to do, so long as no immediate or significant harm is done to themselves or to those around them.” And we can talk about a definition of harm later. Those are all answers that do not bode well for the long term, sustainable reproduction of flourishing human societies.

Hence, tradition is under suspicion. This is somewhat ironic given the longstanding interest within sociology of the study of ethnic culture and codes. The ways of the Irish Catholic or the black Chicagoan used to be fodder for fascinating sociological study. That has largely receded now. But what would replace tradition, collective culture, and code after we deem them burdensome and conformist? In reality, we see the rise in the imperative—and it feels exactly like that—the imperative of creating and expressing a “personal culture.”

The solidarity the Irish, or African Americans, or women, or Catholics once felt toward each other is dissolved. We are on our own—it would seem—since kinship, neighborhood, faith, and common culture can no longer be counted upon and are openly undermined. You invent yourself; you pick your sex, your gender expression, your faith, you name it. This is actually a remarkable and destructive burden, resulting in an explosion of fragile associations requiring our management rather than our reliance. No wonder we fear giving offense. What we have constructed in place of tradition is so unbelievably fragile that it requires new rules, norms, and even laws to prop it up. And thus we spend far more time reinventing ourselves, seeking a promised freedom that does not, because it cannot, materialize. 

Instead of resting, if even uncomfortably, in social structures not of our own making, we cannot rest at all; the self—the primal unit of postmodern society—requires renewal and remaking and, as always, the pursuit of acceptance and love. And yet amid all this new dynamism, social theorist Margaret Archer remarks, “there is no growth in real personal differentiation, and thus in the heterogeneity of the population.”[6]It’s true. We are not really creating genuinely interesting diversity, only rumors of it. Facades of it. Diversity is largely a masquerade. In reality, it is really about the fashioning of sameness, the McDonaldization of human culture.

There seem to be two key hopes among those who welcome same-sex marriage, and they are pretty distinctive hopes. The first is that it would stabilize the relationships (and perhaps the families) of those same-sex couples that get married. Another quite different hope is that it makes marriage less necessary, that this is a signal that we no longer need a marrying culture. They are are two quite different hopes, of course. We will see how each hope fares, because the reality is nearing. 

No one disputes that same-sex households have historically displayed breakup rates in excess of those in opposite-sex, especially married, households. Why this is the case is not often discussed, apart from the assumption that legal same-sex marriage will cure that. Any claims about the likely persistence of this empirical phenomenon are dismissed as either irrelevant or mean-spirited. But I have weighed in with an argument that suggests we should not presume gay and lesbian marriages will fare like opposite-sex marriages. Why not?  

Two reasons: men are men and women are women, and I am not convinced that the experience of homosexuality really alters that at a primal level.

Two sociologists who examined the same-sex parenting literature back in 2000 noted statistically higher breakup rates among lesbian couples, even in nonprobability samples where you could conceivably find the most stable participants. They reasoned:

Not only do same-sex couples [in this case, women] lack the institutional pressures and support for commitment that marriage provides, but qualitative studies suggest that they tend to embrace comparatively high standards of emotional intimacy and satisfaction.[7]

Of course women do this. My wife does this. Her standards for emotional connection are higher than mine. But the difference is that she is married to a man, a fact that forces her to navigate difference. She is married to someone who does not exhibit comparable standards around emotion, who has to be coaxed and taught to value them. And the reverse is true: I am not married to a man, meaning that she will fail to value some things that I wish she would value more.

Hence what you expect from marriages between women are exactly what social scientists find: greater valuing of emotional intimacy, greater preference for egalitarianism, a decline in sexual frequency over time, and higher breakup rates. 

And marriages between men—what do we find? Less concern for egalitarianism, less concern for emotional intimacy, more frequent sexual behavior and a lower bar to extramarital liaisons—all completely in step with what anyone should expect from men, not because they are gay but because they are men. 

These differences reveal that marriage between men and women—for all its historical and cultural diversity—exhibits traits less apt to be seen consistently in same-sex relationships, and that is navigating sexual difference, learning to love, and to like, someone quite different from you.

Longstanding marriages—especially those with children—are less likely to model what Giddens predicted about the “pure relationship,” in part because many such marriages remain complementary in gender role distinctions and hence more “functional” than “symbolic” in their marital meaning. While certainly such marriages are becoming a minority, quickly, they will never become extinct. Social scientists should expect the “pure relationship” to characterize more same-sex than opposite-sex unions in any era. Giddens explicitly said as much. They are vanguards, he claimed, of the “pure relationship” model. 

So while the first hope, that same-sex marriages will be more secure than cohabitation, is likely to be realized, it does not mean that the second hope, the receding of a marriage culture, will not also occur at the same time. We may not notice, however, because the receding of a marriage culture happens when there is a loss of shared assumptions about what marriage is and what it means, and that—as Giddens warned us back in 1992—has been happening for decades. This all raises some questions about marriage itself:

How central is sexual fidelity to the meaning of marriage? Can spouses cut deals with each other about sleeping with other people? Is that still a marriage? 

How central are children to the meaning of marriage? How should you have them? Should you control all aspects of childbearing? 

Are egalitarian ideals—equality—sensible within the average marriage in the United States and elsewhere in the world? Does scorekeeping foster happiness?

In the end, no one is trying to prevent adults who profess love for each from being together. I would not dream of contesting that. The question remains what is marriage? And what ought communities esteem, and why?

One of the reasons so many have been unprepared to answer these questions is because there was never a need to for millennia. Forming a stable sexually complementary partnership was how people survived and how cultures continued.

These are very new questions, ones that only the technological developments of the past 50 years enabled. No pill? No in-vitro fertilization? I would not be here talking to you today if it weren’t for them and for the revolution in relationships that Anthony Giddens describes emerging in their wake. 

In the end, the cultural turn away from the biological family in the academy and legal community is remarkable. Even while the evidence for its strengths is incontrovertible, and the costs borne by communities in its absence obvious, it is increasingly politically unpalatable to go to bat for the nuclear family while the pitched battle over the emergence of new family forms and meanings rages on.

The legal squabble over some of these things may be drawing to a close. But the cultural contest may be just beginning. 

Mark Regnerus, Ph.D., is associate professor of sociology at the University of Texas at Austin, a research associate of the university’s Population Research Center, and a senior fellow at the Austin Institute for the Study of Family and Culture.

[1] U.S.  Census Bureau, American Community Survey 2014.  These are estimates of the rates among unmarried women over the age of 15 having married in 2014. 

[2] Anthony Giddens, The Transformation of Intimacy (Stanford, CA: Stanford University Press, 1992), 27. 

[3] Ibid., 61.

[4] Mark Regnerus, “Parental Same-Sex Relationships, Family Instability, and Subsequent Life Outcomes for Adult Children: Answering Critics of the New Family Structures Study with Additional Analyses,” Social Science Research 41 (2012): 1367–77.

[5] D. Paul Sullins, “Emotional Problems among Children with Same-Sex Parents: Difference by Definition,” British Journal of Education, Society, & Behavioural Science 7 (2015): 99–120.

[6] Margaret Archer, “‘Caritas in Veritate’ and Social Love,” International Journal of Public Theology 5 (2011): 273–95, at 281. 

[7] Judith Stacey and Timothy J. Biblarz, “(How) Does the Sexual Orientation of Parents Matter?” American Sociological Review 66 (2001): 159–83, at 177.