Bella DePaulo Ph.D. Nov 27, 2016
Older women are psychologically healthy with or without romantic partners
When adults get into their mid-fifties and beyond, how much does a romantic partner matter to their psychological well-being? Matthew Wright and Susan Brown of Bowling Green University, authors of a study recently published online at the Journal of Marriage and Family, expected to find a hierarchy of good outcomes. They predicted that married people would enjoy the greatest psychological well-being. Cohabiters, they thought, would do next best, and daters would follow in third place. They expected unpartnered single people to be worst off, psychologically. That is not what they found.
Instead they found that for women, partnership status made no difference. Whether the women were married, cohabiting, dating, or single and unpartnered, there were no statistically significant differences in their experiences of depression, stress, or loneliness. There were some nonsignificant trends in the data, but even those were not always consistent with the authors’ predictions. For example, the women who were dating tended to experience more stress than the single women without a romantic partner.
For the men, having a romantic partner mattered more than it did for the women, but again, not exactly in the ways the authors predicted. The authors thought that the unpartnered single men would do worse than the single men who were dating on every measure, but that never happened. The men who were dating did not differ significantly from the unpartnered single men in their experiences of depression or stress or loneliness.
The cohabiting men were predicted to do less well than the married men, but that never happened, either. The married men were more likely to report frequent depressive symptoms. They were also slightly more likely to experience stress than the cohabiting men. Marriage was also no protection against loneliness, as married men were no less lonely than cohabiting men. Cohabiting men also did well in comparison to the dating or unpartnered men on two measures of well-being: They were less likely to report frequent depressive symptoms or loneliness.
The unacknowledged psychological strengths of older adults who are not married
The hierarchy that the authors predicted is consistent with the prevailing narrative about marriage and coupling in contemporary society. Married people are supposed to do better than everyone else. Cohabiters should do next best, then people who are dating. Single people with no partners should, in theory, do worse than everyone else. The theory is that these four groups form a continuum of social attachment (with married people enjoying more social ties, and more social and emotional support, and unpartnered singles the least) and also a continuum of commitment (again, with married people showing the most commitment and unpartnered singles the least).
For the older women, though, partnership status did not make a lick of a difference. Married women, cohabiting women, dating women, and unpartnered single women – they were all about the same in their experiences of depression, stress, and loneliness.
For the older men, there were some differences, but not exactly the ones the authors predicted. Married men never did significantly better than everyone else, on any of the measures. In fact, when there was a difference between the married men and the cohabiting men, it favored the cohabiting men. According to the theoretical perspectives that guided the authors’ work, the dating men should have done better than the unpartnered single men, because they supposedly had greater social attachment and greater commitment. But again, the men who were dating did not do significantly better than the unpartnered single men in any way.
For at least 10 reasons, the psychological well-being of the unmarried people in this study – particularly the unpartnered single people – is especially impressive.
- Married people are the recipients of more than 1,000 federal benefits and protections given only to them. They include tax breaks, Social Security benefits, special considerations under the Family and Medical Leave Act and much more. Unpartnered single people get none of this largesse, and neither do daters or cohabiters. Why is it that married people do not do better than everyone else psychologically when they have these extraordinary advantages?
- Married people also enjoy a vast array of social, cultural, economic, and political privileges simply because they are married. These, too, should have catapulted them far above everyone else in their psychological well-being. But they didn’t.
- Among the participants in Wright and Brown’s study, the unpartnered single people – both men and women – were far less well off financially than the married or dating people. The married women’s assets (household assets minus debts) were more than twice those of the unpartnered single people, and for the men, the difference was almost as great. (The cohabiters’ assets were similar to those of the unpartnered single people’s – yet they did as well or better than the married people on every measure of psychological well-being.)
- The unpartnered single people in the study were the least likely to be employed. For the men, the difference between the unpartnered and everyone else was especially large.
- Unpartnered women in the study were least likely to have some college education. Among the men, only the cohabiters were less likely than the unpartnered singles to have some college education.
- Unpartnered men and women in the study were least likely to have private health insurance. For the men, the difference between them and everyone else was especially large.
- People who are not white, who have the challenges of racism in addition to singlism, were disproportionately represented among those who were not married. Among the women in the study, the proportions followed the proposed hierarchy precisely, with the fewest people who were not white among the married, then the cohabiting, then the dating, and the greatest proportion among the unpartnered. For the men, the distribution was similar, except that the dating men included a greater percentage of people who were not white than the unpartnered men.
- The unpartnered single people in the study were also disadvantaged by the authors’ decision to lump together all unpartnered single people, regardless of whether they were divorced or widowed or had always been single. (They were constrained by the sample size, but other researchers do the same thing.) People who are divorced and widowed, especially if their marriages ended fairly recently, may feel especially depressed, stressed, and lonely. Lifelong single people often do quite well psychologically. By including the previously married in with the lifelong single people, the study likely underestimated the true psychological well-being of lifelong single people.
- The single people in the study were also disadvantaged methodologically in another way. The married group included only those people who were currently married. The previously married, who often do less well than the married people and the lifelong single people, got assigned to one of the other groups. The authors’ prediction was that the people who got married would do the best – but the people who got divorced and widowed did get married. They just didn’t stay that way. The authors – like just about everyone else who studies marital status – gave the married group an unfair advantage by excluding from that group everyone who got married, hated it, and got divorced. They compounded the unfairness by including the previously married with other groups, such as the lifelong single people, whose psychological well-being may have looked even better if they were studied on their own.
- Even if the authors had found exactly what they predicted, with married people doing the best and unpartnered singles the worst, the design of their study would not allow them to make the claim that the married or partnered people were doing better because they were married or partnered. The people in the four groups were all different people. Consider, for example, the unpartnered single people who chose to be single, who perhaps were single at heart. If they were to marry or cohabit, they might not experience any improvement in psychological well-being – and in fact could end up doing less well, psychologically – even if the people who chose to marry or cohabit did better. (In this study, especially among the women, they generally did not do better.)
- The authors are not alone in giving short shrift to the methodological issues that disadvantage single people, and they are not alone in citing uncritically claims about the benefits of marrying that have been extensively critiqued. The methodological issues are basic ones but rarely fully acknowledged in the research on marital status. (This is explained in more detail here, here, and here.)
The authors offered one possible explanation for why partnership status didn’t matter for the women: Wives do more caregiving than husbands, so they may not get the same benefits from their partnerships that men do. That explanation, though, overlooks all the research showing that single people do more of the work of caring for their aging parents than married people do, and that single people are more likely to step in to help people who need help for three months or more, even when those people are not family members. So single women are in some ways doing more of the work of caregiving than married people, yet, in this study, they are doing just as well as partnered women in their psychological well-being.
Considering the profoundly important ways that the unpartnered single people are disadvantaged – both in the society at large, in this particular study, and in the specific methodology of this study and so many others – perhaps the most significant question raised by the present study (and the research on marital and relationship status more broadly) is, how is this possible? How is it possible that the unpartnered single people in this study were disadvantaged economically, educationally, in their employment status, in their access to private health insurance, and faced more racial stigma and discrimination as well, yet they held their own on every measure of psychological well-being? (In some of their analyses, the authors tried to control for these differences and a few others, but it made little difference.) With so much going against them, why didn’t the unpartnered single people do worse than everyone else, as the authors predicted they would? How is it possible that single people are stereotyped, stigmatized, marginalized, and discriminated against, and still live happily ever after? That’s a question hardly anyone addresses.
What mattered more than having a romantic partner
The focus of the article was on romantic partnerships and their purported benefits. But for the women especially, the predicted benefits of having a spouse or cohabiting partner or dating partner simply were not there.
Those romantic partnerships were supposed to provide the social attachments and commitments that enhanced people’s psychological well-being. For the most part, they didn’t. The data the authors reported actually did show the importance of social ties, just not the narrow romantic ones that were at the center of their interest.
All participants answered questions about social support, assessing the extent to which they could open up to their friends and family, and rely and friends and family when they have a problem. Social support did matter, for both women and men, in almost every way.
Although romantic partnership never mattered for women, social support from friends and family always did. Women with more social support were less likely to report frequent depressive symptoms, they were less likely to experience stress, and they were less likely to be lonely. Social support from friends and family mattered to men, too, though not quite as much as it did for the women. Men who had more social support were less likely to report frequent depressive symptoms and they were slightly less likely to experience stress.
In the section of the article in which the authors discussed the meaning and implications of their findings, they never mentioned what they found about social support. They never suggested, for example, that perhaps social support from friends and family, and the reciprocal support participants likely provided in return, constituted the very social attachments and commitments theorized to be so special to romantic partners. They never urged their colleagues to consider the possibility that support from friends and family is more important than romantic partnerships. Instead, they said their findings “demonstrate the need to consider the benefits of non-marital unions for older adults.”
Other research has already shown the significance of ties beyond romantic ones to people of different marital statuses. It is single people, more than married people, who maintain ties with friends, neighbors, siblings, and parents. When people marry, they become more insular. They lean on one particular social attachment and commitment, the one to their spouse. The hierarchy perspective considers one kind of relationship, a romantic relationship, to be paramount, and marginalizes all the other significant people and relationships in our lives. Maybe that’s one of the reasons the authors did not find what they thought they would.
Details of the study
Participants were about 1,000 people, ages 57 through 85, from a representative national sample from the National Social Life, Health, and Aging Project. The project was a longitudinal study but the authors of the present study analyzed the three dependent measures (depressive symptoms, perceived stress, and loneliness) from only one point in time, Wave II of the data. The other variables in the study, including partnership status, demographic characteristics, socioeconomic resources, and social support, were measured at Wave I.
Depression was assessed by the Center for Epidemiologic Studies Depression Scale. It includes items such as “felt depressed,” “felt everything was an effort,” and “did not feel like eating.” The authors created a measure that separated people into those who experienced depressive symptoms frequently and those who did not experience symptoms as frequently.
Perceived stress was measured by participants’ answers to questions such as “I felt difficulties were piling up so high I could not overcome them” and “I was unable to control important things in my life.” The authors created a measure that separated people into those who rarely or never experienced those stressors in the past week, and those who experienced them more often.
Loneliness was assessed by participants’ answers to three questions: How often do you feel that you lack companionship? How often do you feel isolated from others? How often do you feel left out? (When single answer that last question, they may be describing more than a feeling, as, for example, when coupled people exclude their single friends because they are single.)
There actually is a hierarchy of value and respect, and it is just the one the authors described: married people are valued and respected the most, and given the most benefits and protections; cohabiters are in second place, followed by single people who are dating. Single people without romantic partners are the most stereotyped and stigmatized. The authors thought that psychological well-being would follow the same hierarchy, with married people enjoying the most and unpartnered single people the least. There are many reasons why it would be reasonable to expect that, including all the ways in which married people are advantaged and single people are not – both in society in general and in the specific way this study, and many other studies of marital status, are designed and analyzed. But that’s not what they found. Partnership status made no difference whatsoever for the women: the married, cohabiting, dating and unpartnered single women were all about the same in their experiences of depression, stress, and loneliness. Partnership made some difference to the men, but not always in the way the authors expected. For example, married men never did significantly better than the cohabiting men in any way. The dating men also did no better than the unpartnered men on any of the measures of their psychological well-being. The key question left unanswered by this study is: How is it that single people do so well psychologically, when they have so much stacked against them?
The fact that partnership status did not matter to women (and did not always matter to men, either) doesn’t mean that social ties and interpersonal commitments don’t matter. They do matter, when ties beyond the narrow romantic ones are considered. The men and women with more social support from friends and family experienced less depression and less stress, and the women less often felt lonely.
Currently available online; will appear in print later.