Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jun 1.
Published in final edited form as: J Fam Issues. 2021 Jul 1;43(6):1650–1668. doi: 10.1177/0192513x211030033

The Trajectory of Subjective Well-Being: A Partial Explanation of the Marriage Advantage

Alfred DeMaris *, Gary Oates **
PMCID: PMC9230772  NIHMSID: NIHMS1717551  PMID: 35755972

Abstract

Although several studies have documented a distinct marriage advantage in well-being, it is still unclear what it is about marriage that renders this benefit. We hypothesize that it is due to factors theorized to accrue to matrimony, such as elevated financial status and specific social psychological supports. We examine the trajectory of subjective well-being for 1,135 respondents from the three-wave 2010 GSS panel survey utilizing linear mixed-effects modeling. We find that about two-fifths of the marriage advantage in subjective well-being is accounted for by a mixture of control variables, finances, and emotional factors, with most of this due to elements that are associated with the marital context. Higher annual income, enhanced interpersonal trust, greater sociability, and less of a sense of loneliness and isolation appear to be responsible for a substantial component of the marital advantage. We further find that the marriage advantage is invariant to both race and gender.

The last several decades in the U.S. have witnessed a decline in the relevance of the institution of marriage for regulating intimate relationships. In particular, since the 1950s marriage has experienced a steady erosion of its power as a social institution (Cherlin, 2009, Coontz, 2005). Whether speaking of the power to constrain behavior or the possibility of offering rewards not found in other arrangements, the influence of marriage has steadily waned. One would expect that an institution that has been largely stripped of its special status would no longer have much of an influence on behavioral outcomes. Nevertheless, several studies continue to document a marriage advantage in subjective well-being, net of relevant control variables (See Lee, 2015, for a review of this research). And despite some contrary findings, most researchers also find that married persons have better mental health than comparable cohabiting couples, despite the similarity in their daily circumstances (e.g. Mikucka, 2016; Perelli-Harris & Styrc, 2018; Wadsworth, 2016). This begs the question: what is it about marriage, per se, that confers an advantage in well-being? Although there has been considerable research on the topic, we are still somewhat in the dark as to the answer to this question.

The purpose of the current study is to attempt to account for the marriage advantage in subjective well-being. Toward that end, we draw on a well-known probability sample of the U.S. population in the context of a three-wave panel study. We examine the trajectory in subjective well-being over a four-year period, comparing those who are married throughout the follow-up period with those who remain unmarried throughout the follow-up period. We include a number of relevant control variables. We also introduce some theoretically relevant social psychological mediating variables in an effort to test hypothesized processes regarding the benefits to formalizing an intimate relationship. Ultimately we wish to answer the question: how much of the marriage advantage in subjective well-being can be accounted for by known factors? And, in particular, does marriage promote well-being via mechanisms theorized to distinguish this domestic arrangement from any other?

Background

The Advantages of Marriage

Being married is held to be advantageous to health—both physical and mental—for many reasons. Marriage entails what Cherlin (2009, p. 138) refers to as “enforceable trust.” It involves a public commitment to enter into a potentially lifelong, caring relationship with one’s partner. As it is a legally enforceable commitment, it requires substantially more effort to sunder than, say, a cohabiting relationship (Perelli-Harris & Styrc, 2018). Therefore the married can be relatively secure in their spouse’s obligation to support them both financially and emotionally in times of need. They also have regular access to a confidant, a companion, and a life partner. All of these elements may contribute to a feeling of subjective wellbeing (Horwitz et al., 1996; Kim & McKenry, 2002; Marks, 1996). Married people also appear to embrace healthier and less risky lifestyles than, say, the unmarried (Pienta, Hayward, & Jenkins, 2000). This can also spill over as improvements in physical health. Due to the pooling of resources, marrieds can reap the benefits of economies of scale, improving their financial status. Socioeconomic resources are consistently found to be positive predictors of emotional well-being (Tan, Kraus, Carpernter, & Adler, 2020).

Several more subtle aspects of married life are expected to enhance psychological functioning (Ross, 1995). Marriage provides social integration and nesting inside a network of social obligation, which also protects against a sense of isolation (Stack, 1998). As Coontz (2005) has noted, what makes marriage unique among all other social systems is that it allows us to acquire in-laws. Thus, marriage has always been about the linking together of families and widening the network of kin (Coontz, 2005). Cherlin’s (2009) allusion to marriage’s providing enforceable trust also points to another social-psychological benefit. As one’s spouse is described by many as their closest friend (Grover & Helliwell, 2019), being able to trust the spouse should influence trust in others in general. One would therefore expect married people to be more trusting of others, and to feel more secure in the world as a result. This should, in turn, elevate psychological well-being (Ashleigh, Higgs, & Dulewicz, 2012). By virtue of having a committed partner, married people should also benefit from more emotional support than singles. This leads to a reduction in depression and anxiety (Rankin, Paisley, Mulla, & Tomeny, 2018). Finally, married people have been found to have higher levels of attachment to a significant other than the unmarried (Ross, 1995).

Evidence from Extant Research

Several studies find that married people are physically healthier, happier, and more satisfied with life in general, compared to those who are unmarried, including cohabitors (Brown, 2000; Diener, Gohm, Suh, & Oishi, 2000; Horwitz & White, 1998; Kim & McKenry, 2002; Lamb, Lee, & DeMaris, 2003; Lee & Ono, 2012; Marcussen, 2005; Mikucka, 2016; Perelli-Harris & Styrc, 2018; Soons & Kalmijn, 2009; Soons, Liefbroer, & Kalmijn, 2009; Stack & Eshleman, 1998; Stutzer & Frey, 2006; Wadsworth, 2016; author, dates). On the other hand, the unmarried have been found to be substantially happier than those in marriages described as “not too happy” (Chapman & Guven, 2016). Undoubtedly, relationship happiness plays a key role in the extent to which marrieds’ subjective well-being exceeds that of other types of people. However, we do not address this issue in the current study. In that the dependent variable involves reports of happiness (see Methods, below), marital happiness overlaps too much with it to be a credible explanatory variable.

With the exception of formalization of the commitment, unmarried cohabitation today resembles marriages in most respects. One might expect cohabitors to be comparable to marrieds in well-being. Some studies indeed find no difference in the well-being advantage accruing to marriage vs. cohabitation (Musick & Bumpass, 2012; Ross, 1995; Zimmerman & Easterlin, 2006; Perelli-Harris, Hoherz, Lappegard, & Evans, 2019). However, a number of other studies show marriage to be superior to cohabitation in fostering subjective well-being (Mikucka, 2016; Perelli-Harris & Styrc, 2018; Wadsworth, 2016). For reasons to be explained below, this is also an issue that we do not address in the current study, where we simply distinguish the married from the unmarried. At worst, this makes for a conservative test of the marriage advantage. Many of those counted as “single” are cohabiting and therefore benefit from having a steady intimate partner after the fashion of married individuals.

Some studies have claimed to have found a gender gap, such that men benefit more in mental health from marriage than women (Brown, Bulanda, & Lee, 2005; Gove, Hughes, & Style, 1983; Mastekaasa, 1994). However, a number of other studies have found no significant gender difference in the marriage advantage (Bierman, Fazio, & Milkie, 2006; Diener et al., 2000; Dush & Amato, 2005; Hughes & Waite, 2002; Lee & Bulanda, 2005; Simon, 2002; Soons et al., 2009; Stack & Eshleman, 1998; Williams, 2003). A couple of studies have found that women actually benefit more from marriage than men. Mikucka’s (2016) analysis of 102 countries from the World Values Survey-European Values Survey dataset showed that the life satisfaction advantage of marriage decreased over time for men but remained constant for women. Grover and Helliwell (2019) examined data from a large 18-wave panel dataset from the United Kingdom. They found the life satisfaction impact of marriage to be significantly larger for women than for men. In the current study, we reassess whether there is a gender difference in the potential protective effect of marriage for subjective well-being.

It is unclear whether the marriage advantage in well-being is invariant to race. African-Americans are less likely to marry, have poorer quality marriages, and have a higher divorce rate than Whites (Bulanda & Brown, 2007). Marriage is less central to intimate relationships in the Black community than it is for majority individuals (Liu and Umberson, 2008). Given its more circumscribed nature among African Americans, it’s possible that marriage is not as protective for them in the area of well-being. Findings from research so far are mixed. Barrett (2003) finds race differences in the consequences of marital termination. In particular, separated Whites reported significantly more depressive symptoms than their Black counterparts, but divorced Blacks showed more symptoms of substance abuse or dependence than divorced Whites. Using repeated cross-sectional data from the National Health Interview Survey, Liu and Umberson (2008) found that the difference in self-reported health for the married vs. the unmarried declined more dramatically for Blacks than Whites across time. This suggested that the marriage advantage in health became smaller over time among Blacks, compared to Whites. In this study, we re-examine whether there is any difference in the marriage advantage in subjective well-being for minorities vs. Whites.

Attributing a Causal Impact to Marriage

It has often been remarked that attributing causality to marriage itself in explaining the marriage advantage in well-being is problematic. Compositional differences between groups who marry vs. those living in other arrangements are a prime competing explanation. Marriage may be selective of those who are already better off mentally and physically. After all, those with a more cheerful outlook or who are in better health would be more attractive marital partners than others. Additionally, those suffering from depression, anxiety, or other emotional difficulties tend toward inertia in facing life decisions (Baeza-Velasco, et al., 2020). Such individuals are less likely than mentally healthier people to exert the initiative to move a relationship toward marriage. Additionally, married individuals with physical or mental health issues are more likely to be abandoned by their spouses (Blekesaune, 2008)

In an earlier era, marriage was the signal event that jump-started one’s membership in the adult community. Married people were deemed more mature and responsible than singles by prospective employers. Marriage was viewed as necessary for regular sexual access to a partner and was seen as the only proper milieu for childbearing (Cherlin, 2009; Coontz, 2005). Today many couples view marriage instead as a capstone event, an emblem of having finally achieved a durable and financially stable partnership in a liaison that may already have children. For example, qualitative studies of cohabitors with and without children at home found that many were waiting to achieve financial stability before entering into matrimony. Most often this meant waiting for the male partner to settle into a permanent, well-paying job (Smock, Manning, & Porter, 2005). Even among the lower classes, marriage is only contemplated when male employment is reliable. Edin and Kefalas’ (2005) study of single mothers living in disadvantaged urban areas found many women unwilling to marry their partners until they evidenced stable behavior and an intention to put mother’s and children’s needs above their own. Thus, any SWB benefit accruing to marriage may simply be due to a filtering effect: Only the most committed and financially and emotionally stable couples move to formalize their unions.

Many researchers have attempted to address such selection issues through innovative modeling strategies. These include applying endogenous treatment regression, also known as the Heckman selection model (Wu, Penning, Pollard & Hart, 2003), using autoregressive modeling (Frech & Williams, 2007; Hawkins & Booth, 2005; Horwitz &White, 1998), relying on monozygotic twin pairs (Antonovics & Town, 2004), or using fixed-effects regression (Amato &Kane, 2011; Musick & Bumpass, 2012; author, date). None of these is a panacea for the selection problem, however. In the current study, a fixed-effects approach was not feasible as the prime predictor of interest, being married, was not a time-varying variable. However, the first author (date) examined the marriage advantage in well-being using the GSS panel in an earlier paper. There, marriage was treated as a time-varying covariate. Using a pseudotreatment approach, author (date) found evidence for differential selection into and out of marriage as a function of well-being. Nevertheless, using fixed-effects regression, he found that the marriage advantage remained significant even after controlling for such unmeasured heterogeneity. The Heckman model was not deemed appropriate as it is not especially developed for growth-curve modeling, the analytic technique of the current project. On the other hand, the first author used the Heckman model with the cross-sectional GSS to explore the marriage advantage in well-being in an earlier article (author citation). Again, he found a significant marriage advantage even after controlling for unmeasured heterogeneity.

In this study we control for measured heterogeneity by bringing in important control variables that have been shown to be associated with both being married and subjective well-being. Minority status is one such factor. For example, Blacks are less likely to marry than Whites and also have a higher risk of marital dissolution (Liu & Umberson, 2008). With respect to well-being, race effects appear mixed. Blacks appear to trail Whites with respect to overall life satisfaction and general happiness (Hughes & Thomas, 1998), whereas Blacks appear less likely than Whites to suffer from mood disorders (Keyes, 2009). Age is also important, as the probability of being married increases with age. Education has been found to be positively correlated with marrying, as opposed to cohabiting or staying single (Schoen & Cheng, 2006). Among other things, more educated people are more attractive as marriage partners. Not only are their social skills and conversational abilities more highly developed, but their enhanced career trajectories are an additional attraction. And compared to the less educated, more-educated people appear to suffer less from physical and mental illnesses. Finally, religiosity is an important control. More religious people are more likely to marry (Ellison, Burdette, & Glenn, 2011), as well as less likely to suffer from depression and other emotional afflictions (Howell, et al., 2019).

The Current Study

In this study we reexamine the marriage advantage in subjective well-being and attempt to account for it with covariates. We expect to find a significant marital advantage, as has been documented above, and in particular, using the GSS (author citations). We explore whether the marital advantage is dependent upon race or gender. As the findings are not clear as to what direction any such dependence would take, we make no hypotheses in this regard. We expect that some portion of the marital advantage is accounted for by covariates like race and education, reflecting compositional differences between individuals. But we also expect that some portion will be accounted for by socioeconomic status, sociability, trust, and isolation. To the extent that marriage enhances well-being via improving financial status and providing spousal emotional sustenance, controlling for these last factors should diminish the size of the marital advantage. However, this is only a sufficient condition for inferring causality of marriage. The converse does not hold: that finances, sociability, and so forth reduce the marital advantage does not necessarily mean that marriage brings about better well-being. It can also be that those with better finances, and the more sociable, trusting, and gregarious are more likely to enter into matrimony. We keep these caveats in mind.

Methods

The Data

We employ data from the 2010 GSS panel, which involved three waves of an initial survey of 2,044 individuals contacted via probability sampling of the U.S. population. The two follow-ups were done in 2012 and 2014. Of the original respondents, 1,551 responded to wave 2 and 1,304 additionally responded to wave 3 of the study. We utilize only respondents who answered all three waves of the survey and who moreover had nonmissing responses on marital status at each time. We also restricted the sample to those who were either married for all three time periods or single for all three time periods. This was done to ensure causal priority of marital status over subjective well-being. That is, we wanted to be sure that marital status at any given time was not being driven by subjective well-being from a previous wave. In preliminary analyses we waived this requirement and ran the substantive analyses with marital status as a time-varying covariate, including all those who changed martial statuses across waves. No substantive changes in our results were observed. Our final sample consisted of 1,135 respondents, of whom 513 were continuously married, and 622 were continuously single over time.

Measures

The outcome.

Our dependent variable, subjective well-being, was a perceptual measure of both emotional and physical well-being. It consisted of three items. The first was worded “Taken all together, how would you say things are these days?”, with possible responses “very happy (1),” “pretty happy (2),” and “not too happy (3).” The second variable was “In general, do you find life exciting, pretty routine, or dull?”, with “exciting,” “routine,” and “dull” coded 1 – 3, respectively. The last item was “Would you say your own health, in general, is excellent, good, fair, or poor”?, with “excellent,” “good,” “fair,” and “poor” coded 1 – 4, respectively. All three items were reverse coded so that the high score reflected maximum well-being. Then, as the variables were in different metrics, the items were first standardized, then summed to create a well-being scale. Reliabilities for the three time periods were .47, .51, and .56 for times 1 – 3, respectively. The reliabilities are admittedly low. But these scales are tapping the perception of both emotional and physical health and would therefore not be expected to exhibit high internal consistency. The scale showed good variability and little skewness across waves and has been used in other published work on the marriage advantage (author citations). Moreover, a thorough search of the GSS panel data revealed the items in this scale to be the only well-being variables measured in all three waves.

Focal predictor.

The prime predictor was marital status, tapped by a dummy for being continuously married vs. continuously single throughout the study. Using a coarse-grained measure of this nature provides a conservative test of the marriage advantage. Many of the “single” respondents are in all likelihood either involved in a dating relationship or cohabiting with an intimate partner. As a result, they are reaping many of the same benefits as marrieds. Additionally, as argued by author (date), some of the married individuals are probably in the throes of marital disintegration. Although dating and cohabiting relationships also experience disintegration, they are easier to exit than marriages. This would have the effect of including in the sample more marrieds, than singles, whose well-being is deteriorating under relationship strife.

Control variables.

Control variables were gender, minority status, age, education, number of children, and religiosity. Gender was represented by a dummy for being female. Minority status was represented by a dummy for being an ethnic minority, with Whites as the reference group. One might argue that this conflates African Americans and other more privileged groups, such as Asians, under the same umbrella. Therefore we re-ran all analyses restricting the sample to just Whites and Blacks. No substantive differences in the results were observed when we utilized this strategy. Age in years was respondents’ age in the first time period. Education was measured in years of formal schooling completed. Number of children ever borne at wave 1 was used to tap the presence of children at home. Because of the positive skewness in this variable, it was recoded from 0 to 4, with 4 reflecting 4 or more children. Finally, religiosity was a scale consisting of three variables. The first asked how often the respondent prayed, with responses ranging from 1 for “several times a day” to 6 for “never.” This variable was reverse-coded. The second was the frequency of church attendance, ranging from 0 for “never” to 8 for “several times a week.” The third was the respondent’s certainty of belief in God, with responses ranging from “I don’t believe in God (1)” to “I know God really exists and I have no doubts about it (6).” Because the items were in different metrics they were standardized first, then summed to form a religiosity scale. Reliabilities at times 1 – 3 were .79, .80, .80, respectively.

Mediating variables.

The variables used as mediators of the marriage advantage fell into four categories: socioeconomic assets, interpersonal trust, sociability, and aspects of loneliness and isolation. All except the loneliness/isolation measures are time-varying covariates. Socioeconomic assets consisted, first, of annual family income in ten-thousands of dollars. This factor was inflation-adjusted and couched in terms of year 2000. The second asset was a dummy for home ownership, with renting or other domicile arrangement as the reference group. Interpersonal trust was based on three items. The first was phrased “Would you say that most of the time people try to be helpful, or that they are mostly just looking out for themselves?”, with response categories “try to be helpful (1),” just look out for themselves (2),” and “depends (3).” These responses were recoded so that “helpful” was 3, “look out for selves” was 1, and “depends” was 2. The second item was “Do you think most people would try to take advantage of you if they got a chance, or would they try to be fair?” Responses were “would take advantage of you (1),” “would try to be fair (2),” “depends (3).” These were recoded so that “try to be fair” was 3 and “depends” was 2. The third question was “Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people?” Responses were “most people can be trusted (1),” “can’t be too careful (2),” “other, depends (3).” This item was recoded in the same fashion as the first item. These measures were then summed to form a scale of interpersonal trust. Reliabilities for the scale at times 1 – 3, respectively, were .67, .67, .71. The sociability scale consisted of three items asking respondents how often they: spent a social evening with relatives, spent a social evening with someone who lives in their neighborhood, spent a social evening with friends outside the neighborhood? Responses for each item ranged from 1 for “almost every day” to 7 for “never.” All items were reverse-coded and summed to form the scale. Reliabilities at times 1 – 3 were .41, .43, .44. These low values simply reflect that the scale is a simple count of the frequency of various types of socializing. These measures are not necessarily tapping any single underlying dimension. The final measures were individual items, all measured only at Time 3. The first was “How often do you feel that you lack companionship?” The second was “How often do you feel left out?” And the third was “How often do you feel isolated from others?” All were coded from 1 for “often” to 4 for “never.” We left the coding as is, so that the high score represented the degree to which respondents never experienced the particular feeling. Descriptive statistics for all variables in the analysis are shown in Table 1.

Table 1.

Descriptive Statistics for Study Variables (Number of Person-Periods Ranges from 2322–3405)

Variable M SD Min Max
Subjective Well-Being* 0.019 2.464 −6.750 4.251
Time in Years* 2.000 1.633 0.000 4.000
Married 0.452 0.498 0.000 1.000
Femalea 0.563 0.496 0.000 1.000
Minorityb 0.218 0.413 0.000 1.000
Age at Time 1 48.337 16.261 18.000 89.000
Education at Time 1 13.852 2.884 2.000 20.000
Number of Children at Time 1 1.714 1.343 0.000 4.000
Religiosity* 0.014 2.521 −5.792 3.332
Annual Family Income in 10,000s* 5.104 4.447 0.037 17.871
Owns Home*c 0.671 0.470 0.000 1.000
Interpersonal Trust* 6.097 2.241 3.000 9.000
Sociability* 12.168 3.582 3.000 21.000
Never Isolated at Time 3 3.159 0.893 1.000 4.000
Never Friendless at Time 3 2.950 1.032 1.000 4.000
Never Feel Left Out at Time 3 3.004 0.895 1.000 4.000
Open in a new tab
*

Time-varying variables.

a

Male is contrast category.

b

White is contrast category.

c

Rents or has other domicile arrangement is contrast category.

Statistical Analysis

In that measurement occasions were nested inside respondents, we employed the linear mixed-effects model to analyze the predictors of subjective well-being. This technique utilizes a combination of fixed and random effects to model the kinds of multilevel data found in cross-section time-series data. By declaring certain effects random, the serial correlation and heteroskedasticity characteristic of panel data is readily accommodated (Fitmaurice, et al., 2004; Singer & Willett, 2003). In our analyses, the intercept and slope of time were random effects. Coefficients are estimated by restricted maximum likelihood and are interpreted exactly as they are in ordinary least squares (OLS) regression. For each of our models, we also present the model R2 suggested by Singer and Willett (2003) as an analogue of the OLS R2.

Missing data.

Each of our summated scales was created by taking the mean of the items and multiplying by the number of items. This is equivalent to summing the items, but it also has the advantage that the mean of the items answered is automatically substituted for any missing item (this is easily shown mathematically). This approach tends to minimize the amount of missing data while taking maximal advantage of existing data. Each scale has a valid score provided that at least one of the items is answered. When converted to person-period data, complete data are represented by an N of 3405 (1135 × 3). The outcome only had two missing person-periods. Most other variables had no missing data. Exceptions were family income, home ownership, interpersonal trust, sociability, and the three loneliness/isolation factors. The most missing data occurred for interpersonal trust, which only had 2,322 valid person-periods. To address this problem, we employed multiple imputation to replace missing data in all analyses, requesting 20 imputations for each missing value. Coefficients in our tables represent the average over 20 replications of the analyses, as outlined by Allison (2002). The reported model R2s are also the average values across the 20 replications.

Results

Table 2 presents four linear mixed-effects models of subjective well-being (SWB). The first includes only the marriage dummy, the linear time effect (intercept and slope for Time) and the marriage × time interaction. The second model adds the control variables gender, race, age, education, number of children, and religiosity. The third model adds the set of socioeconomic mediators, annual family income and home ownership. The fourth model adds the set of social psychological mediators—interpersonal trust, sociability, and the loneliness/isolation factors.

Table 2.

Linear Mixed Effects Coefficients for Models of Subjective Well-Being (Standard Errors)

Predictors Model 1 Model 2 Model 3 Model 4
Intercept −0.565*** (0.091) −2.044*** (0.348) −1.850*** (0.350) −5.428*** (0.413)
Time in Yearsa 0.036 (0.024) 0.037 (0.024) 0.034 (0.024) 0.038 (0.025)
Married 1.300*** (0.136) 1.198*** (0.139) 1.026*** (0.148) 0.788*** (0.144)
Married × Timea −0.087* (0.036) −0.088* (0.036) −0.086* (0.036) −0.092* (0.036)
Femaleb 0.067 (0.118) 0.093 (0.117) 0.122 (0.107)
Minorityc −0.259† (0.146) −0.231 (0.146) −0.177 (0.135)
Age at Time 1 −0.004 (0.004) −0.004 (0.004) −0.000 (0.004)
Education at Time 1 0.132*** (0.021) 0.105*** (0.022) 0.076*** (0.020)
Number of Children at Time 1 −0.069 (0.049) −0.067 (0.049) −0.104* (0.045)
Religiositya 0.087*** (0.021) 0.092*** (0.021) 0.080*** (0.020)
Annual Family Income in 10,000sa 0.049*** (0.013) 0.031* (0.013)
Owns Homead 0.015 (0.112) −0.002 (0.107)
Interpersonal Trusta 0.086*** (0.025)
Sociabilitya 0.060*** (0.013)
Never Isolated at Time 3 0.274*** (0.076)
Never Friendless at Time 3 0.302*** (0.063)
Never Feel Left Out at Time 3 0.345*** (0.077)
R 2 0.053 0.087 0.097 0.204
Open in a new tab
†

p < .1.

*

p < .05.

***

p < .001.

a

Time-varying variables.

b

Male is contrast category.

c

White is contrast category.

d

Rents or has other domicile arrangement is contrast category.

The first model shows that average initial SWB for the unmarried is −.565. For this group, the trajectory of SWB over time is flat, with a nonsignificant slope of .036. Being married adds, on average, 1.3 units to initial SWB, a significant increment. In that a half of a standard deviation of SWB (see Table 1) is 1.232, this increment represents just over half a standard deviation of SWB. On the other hand, SWB declines linearly with time for the married at a rate of .036 − .087 = −.051, or by .051 units per year. Another perspective on this is to see that the model estimates the marriage advantage to be 1.3 −.087 × Time. That is, the marriage advantage in SWB declines at the rate of .087 units per year. At the end of the follow-up period, in 2014, the marriage advantage is estimated to be .953 (1.3 − .087 × 4), and still very significant (analysis not shown). About five percent of the variability in SWB is explained by this model.

Model 2 adds the control variables. A few have significant effects on SWB. Minority status is marginally significant and negative, indicating that minorities have lower SWB than Whites, net of other factors. On the other hand, those with greater education and those higher in religiosity have higher SWB, compared to others. The marriage advantage, at 1.198, has been reduced by eight percent, but remains significant. The R2 indicates that 8.7% of the variance in SWB is explained by this model.

Model 3 adds annual family income in 10,000s of dollars as well as home ownership. The latter has a nonsignificant effect. But, not surprisingly, annual income significantly boosts SWB. Each ten thousand dollar increment to income adds, on average, .049 units to SWB, net of model covariates. The marriage advantage is still significant but is reduced by another 14.4% compared to the previous model. The R2 increases again to .097.

Model 4 adds the remaining mediators representing social-psychological benefits to marriage. They are all positive and significant. Thus, those with greater interpersonal trust, who experience greater sociability, and who tend to never feel isolated, friendless, or left out have higher SWB than others. The final marriage advantage is reduced to .788 units, or a 23.2% reduction from the previous model. Nevertheless, it is still significant. In all, from the baseline model (Model 1) to the final model (Model 4), the marriage advantage dropped by 100 × [(1.3 − .788)/1.3] = 39.4%. Another phrasing of this is that we have accounted for about 39% of the marriage advantage using control variables and mediators. A further breakdown shows that of the .512 units change from first to final model in the marriage advantage, 20% is from the control variables, 34% is from socioeconomic mediators, and 46% is from the social psychological mediators. The model R2 suggests that about 20.4 % of the variability in SWB is accounted for by our final model. Net of all model covariates, is the marriage advantage still significant at the end of the follow-up period? After four years, the marriage advantage is reduced to .421, but is still significant (p < .01; analysis not shown).

Discussion

In this study, we have attempted to deconstruct the marriage advantage in the trajectory of SWB, based on supposed benefits conferred upon individuals through marrying. We found a marriage advantage estimated to be about half a standard deviation of SWB. Our major contribution is that we are able to explain about two-fifths of this advantage via a combination of control variables, socioeconomic benefits, and social-psychological benefits from marriage. We find, moreover, that fully 80% of this explanation is due to marriage benefits. That is, the marriage boost in SWB seems to be mostly accounted for by marriage’s association with higher family income, greater interpersonal trust, more sociable interactions, and lower feelings of isolation, friendlessness, and exclusion. A second contribution is in showing that this advantage not only is invariant to gender, but also to race. Whether male or female, minority or majority, marriage is associated with an enhancement to quality of life. We wish to emphasize that we are not arguing that marriage causes better SWB via the “mediating” variables we have employed. Rather, the marriage advantage in SWB is partially explained by its association with these variables. Whether that association is due to the context of marriage, as opposed to the composition of the marriage “group” cannot be adjudicated in our analysis.

Our work is consistent with most other studies in showing that the married have an advantage over the unmarried in subjective well-being (Brown, 2000; Diener, Gohm, Suh, & Oishi, 2000; Horwitz & White, 1998; Kim & McKenry, 2002; Lamb, et al., 2003; Lee & Ono, 2012; Marcussen, 2005; Mikucka, 2016; Perelli-Harris & Styrc, 2018; Soons & Kalmijn, 2009; Soons, et al., 2009; Stack & Eshleman, 1998; Stutzer & Frey, 2006; Wadsworth, 2016; author, dates). As expected theoretically, marriage appears to be associated with elevated financial well-being, at least in terms of annual income (Antonovics & Town, 2004). The married seem also to benefit from greater sociability, experiencing more frequent socializing with relatives, friends, and neighbors. Throughout most of human history, marriage has remained a unique way of adding other family members to one’s kin network (Coontz, 2005). In this manner, married individuals are more likely to be involved in interconnected nexuses of social bonds. Cherlin (2009) speaks of marriage providing “enforceable trust.” As a result, we expected that the married would be higher on interpersonal trust than the unmarried, which we found. Many family scholars have suggested that being married would be a hedge against feelings of loneliness and isolation (Stack, 1998). Again, we were able to support this aspect of marriage in our results.

Limitations.

As with all research, this study has a number of limitations that must be kept in mind. Chief among them is that we are not able to rule out selection effects as the reason for the association of being married with higher SWB. That is, it is possible that those who are subjectively better off to begin with were more likely to get and stay married in this study. We have controlled for some factors that could account for this selectivity: age, education, and religiosity, in particular. We have also restricted the sample to the continuously married and the continuously single over our four-year follow-up period to help rule out reverse causation. In other words, in our longitudinal study at least, SWB in a previous wave cannot be responsible for an individual subsequently marrying in the next wave. Additionally, a previous analysis of these data with marital status as a time-varying factor found the marriage advantage to persist despite employing fixed effects to rule out selection effects (author citation). To those who might be tempted to infer support for marriage promotion policies from our findings, there is another important caveat. Our study includes only those whose marriages remained intact over the follow-up period. As the average age of our sample in the beginning of the survey was 48, this represents marriages in the middle years, many of which have likely survived for quite some time. (We could not include marital duration as a model control, since age at marriage is unfortunately not available in the GSS.) Thus, our findings do not speak to the fate of marriages undergone earlier in the life cycle, when most people enter first marriages. We also have not made a more fine-grained discrimination of the unmarried into widowed, divorced, separated, never married, or cohabiting categories. Several others have undertaken such discriminations and found, nevertheless, that the married are better off than these other groups in well-being. Therefore, we do not believe our cruder categorization is a major drawback.

Conclusion.

Marriage appears to be associated with greater subjective well-being. However, there is still much to learn about why. In this study, we were able to marshal evidence that financial security as well as social and social psychological factors are responsible for some of this advantage. Ongoing research with more elaborate measures of such constructs as trust, confidence in the reliability of the intimate partner, social support provided by the larger kin network, and similar elements would be helpful. Marriage has lost much of its original power to orchestrate intimate relationships and constrain behavior. Regardless, something about marriage still seems to make it beneficial to individuals’ lives. It behooves us to understand why.

Acknowledgments

This research was supported in part by the Center for Family and Demographic Research, Bowling Green State University, which has core funding from the Eunice Kennedy Shriver National Institute of Child health and Human Development (R24HD050959-01)

References

  1. Allison PD (2002). Missing data. Thousand Oaks, CA: Sage. [Google Scholar]
  2. Amato PR, & Kane JB (2011). Life-course pathways and the psychosocial adjustment of young adult women. Journal of Marriage and Family, 73, 279 – 295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Antonovics K, & Town R (2004). Are all the good men married? Uncovering the sources of the marital wage premium. The American Economic Review, 94, 317 – 321. [Google Scholar]
  4. Ashleigh MJ, Higgs M, & Dulewicz V (2012). A new propensity to trust scale and its relationship with individual well-being: implications for HRM policies and practices. Human Resource Management Journal, 22, 360–376. [Google Scholar]
  5. Baeza-Velasco C, Guillaume S, Olié E, Alacreu-Crespo A, Cazals A, Courtet P (2020). Decision-making in major depressive disorder: Subjective complaint, objective performance, and discrepancy between both. Journal of Affective Disorders, 270, 102 – 107. [DOI] [PubMed] [Google Scholar]
  6. Barrett AE (2003). Race differences in the mental health effects of divorce: A reexamination incorporating temporal dimensions of the dissolution process. Journal of Family Issues, 24, 995 – 1019. [Google Scholar]
  7. Bierman A, Fazio EM, & Milkie MA (2006). A multifaceted approach to the mental health advantage of the married: Assessing how explanations vary by outcome measure and unmarried group. Journal of Family Issues, 27, 554 – 582. [Google Scholar]
  8. Blekesaune M (2008). Partnership transitions and mental distress: Investigating temporal order. Journal of Marriage and Family, 70, 879 – 890. [Google Scholar]
  9. Brown SL (2000). The effect of union type on psychological well-being: Depression among cohabitors versus marrieds. Journal of Health and Social Behavior, 41, 241 – 255. [PubMed] [Google Scholar]
  10. Brown SL, Bulanda JR, & Lee GR (2005). The significance of nonmarital cohabitation: Marital status and mental health benefits among middle-aged and older adults. Journal of Gerontology: SOCIAL SCIENCES, 60B, S21 – S29. [DOI] [PubMed] [Google Scholar]
  11. Bulanda JR, & Brown SL (2007). Race-ethnic differences in marital quality and divorce. Social Science Research, 36, 945 – 967. [Google Scholar]
  12. Chapman B, & Guven C (2016). Revisiting the relationship between marriage and wellbeing: Does marriage quality matter? Journal of Happiness Studies, 17, 533 – 551. [Google Scholar]
  13. Cherlin AJ (2009). The marriage-go-round: The state of marriage and the family in America today. New York: Vintage Books. [Google Scholar]
  14. Coontz S (2005). Marriage, a history: From obedience to intimacy or how love conquered marriage. New York: Viking. [Google Scholar]
  15. Diener E, Gohm CL, Suh E, & Oishi S (2000). Similarity of the relationship between marital status and subjective well-being across cultures. Journal of Cross-Cultural Psychology, 31, 419 – 436. [Google Scholar]
  16. Dush CMK, & Amato PR (2005). Consequences of relationship status and quality for subjective well-being. Journal of Social and Personal Relationships, 22, 607 – 627. [Google Scholar]
  17. Edin K, & Kefalas M (2005). Promises I can keep: Why poor women put motherhood before marriage. Berkeley: University of California Press. [Google Scholar]
  18. Ellison CG, Burdette AM, & Glenn ND (2011). Praying for Mr. Right? Religion, family background, and marital expectations among college women. Journal of Family Issues, 32, 906 – 931. [Google Scholar]
  19. Fitzmaurice GM, Laird NM, & Ware JH (2004). Applied longitudinal analysis. Hoboken, N. J.: Wiley. [Google Scholar]
  20. Frech A, & Williams K (2007). Depression and the psychological benefits of entering marriage. Journal of Health and Social Behavior, 48, 149 – 163. [DOI] [PubMed] [Google Scholar]
  21. Gove WR, Hughes M, & Style CB (1983). Does marriage have positive effects on the psychological well-being of the individual? Journal of Health and Social Behavior, 24, 122 – 131. [PubMed] [Google Scholar]
  22. Grover S, &n Helliwell JF (2019). How’s life at home? New evidence on marriage and the set point for happiness. Journal of Happiness Studies, 20, 373 – 390. [Google Scholar]
  23. Hawkins DN, & Booth A (2005). Unhappily ever after: Effects of long-term, low-quality marriages on well-being. Social Forces, 84, 445 –465. [Google Scholar]
  24. Horwitz AV, White HR, & Howell-White S (1996). Becoming married and mental health: A longitudinal study of a cohort of young adults. Journal of Marriage and the Family, 58, 895 – 907. [Google Scholar]
  25. Horwitz HV, & White HR (1998). The relationship of cohabitation and mental health: A study of a young adult cohort. Journal of Marriage and the Family, 60, 505 – 514. [Google Scholar]
  26. Howell AN, Carleton RN, Horswill SC, Parkerson HA, Weeks JW, & Asmundson GJG (2019). Intolerance of uncertainty moderates the relations among religiosity and motives for religion, depression, and social evaluation fears. Journal of Clinical Psychology, 75, 95 – 115. [DOI] [PubMed] [Google Scholar]
  27. Hughes M, & Thomas ME (1998). The continuing significance of race revisited: A study of race, class, and quality of life in America. American Sociological Review, 63, 785 – 795. [PubMed] [Google Scholar]
  28. Hughes ME, & Waite LJ (2002). Health in household context: Living arrangements and health in late middle age. Journal of Health and Social Behavior, 43, 1 – 21. [PMC free article] [PubMed] [Google Scholar]
  29. Kim HK, & McKenry PC (2002). The relationship between marriage and psychological well-being: A longitudinal analysis. Journal of Family Issues, 23, 885 – 911. [Google Scholar]
  30. Keyes CLM (2009). The Black-White paradox in health: Flourishing in the face of social inequality and discrimination. Journal of Personality, 77, 1678 – 1705. [DOI] [PubMed] [Google Scholar]
  31. Lamb KA, Lee GR, & DeMaris A (2003). Union formation and depression: Selection and relationship effects. Journal of Marriage and Family, 65, 953 – 962. [Google Scholar]
  32. Lee GR (2015). The limits of marriage: Why getting everyone married won’t solve all our problems. Lanham, MD: Lexington Books. [Google Scholar]
  33. Lee GR, & Bulanda JR (2005). Change and consistency in the relation of marital status to personal happiness. Marriage & Family Review, 38, 69 – 84. [Google Scholar]
  34. Lee KS, & Ono H (2012). Marriage, cohabitation, and happiness: A cross-national analysis of 27 countries. Journal of Marriage and Family, 74, 953 – 972. [Google Scholar]
  35. Liu H, & Umberson DJ (2008). The times they are a changin’: Marital status and health differentials from 1972 to 2003. Journal of Health and Social Behavior, 49, 239 – 253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Marcussen K (2005). Explaining differences in mental health between married and cohabiting individuals. Social Psychology Quarterly, 68, 239 – 257. [Google Scholar]
  37. Marks NF (1996). Flying solo at midlife: Gender, marital status, and psychological well-being. Journal of Marriage and the Family, 58, 917 – 932. [Google Scholar]
  38. Mastekaasa A (1994). Marital status, distress, and well-being: An international comparison. Journal of Comparative Family Studies, 25, 183 – 205. [Google Scholar]
  39. Mikucka M (2016). The life satisfaction advantage of being married and gender specialization. Journal of Marriage and Family, 78, 759 – 779. [Google Scholar]
  40. Musick K, & Bumpass L (2012). Reexamining the case for marriage: Union formation and changes in well-being. Journal of Marriage and Family, 74, 1 – 18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Perelli-Harris B, & Styrc M (2018). Mental well-being differences in cohabitation and marriage: The role of childhood selection. Journal of Marriage and Family, 80, 239 – 255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Perelli-Harris B, Hoherz S, Lappegard T, & Evans A (2019). Mind the “happiness” gap: The relationship between cohabitation, marriage, and subjective well-being in the United Kingdom, Australia, Germany, and Norway. Demography, 56, 1219 – 1246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Pienta AM, Hayward MD, & Jenkins KR (2000). Health consequences of marriage for the retirement years. Journal of Family Issues, 21, 559 – 586. [Google Scholar]
  44. Rankin JA, Paisley CA, Mazheruddin MM, & Tomeny TS (2018). Unmet social support needs among college students: Relations between social support discrepancy and depressive and anxiety symptoms. Journal of Counseling Psychology, 65, 474–489 [DOI] [PubMed] [Google Scholar]
  45. Ross CE (1995). Reconceptualizing marital status as a continuum of social attachment. Journal of Marriage and the Family, 57, 129 – 140. [Google Scholar]
  46. Schoen R, & Cheng YA (2006). Partner choice and the differential retreat from marriage. Journal of Marriage and Family, 68, 1 – 10. [Google Scholar]
  47. Simon RW (2002). Revisiting the relationships among gender, marital status, and mental health. American Journal of Sociology, 107, 1065 – 1096. [DOI] [PubMed] [Google Scholar]
  48. Singer JD, & Willett JB (2003). Applied longitudinal data analysis: Modeling change and event occurrence. New York: Oxford University Press. [Google Scholar]
  49. Smock PJ, Manning WD, & Porter M (2005). Everything’s there except money: how money shapes decisions to marry among cohabitors. Journal of Marriage and Family, 67, 680 – 696. [Google Scholar]
  50. Soons JPM, & Kalmijn M (2009). Is marriage more than cohabitation? Well-being differences in 30 European countries. Journal of Marriage and Family, 71, 1141 – 1157. [Google Scholar]
  51. Soons JPM, Liefbroer AC, & Kalmijn M (2009). The long-terem consequences of relationship formation for subjective well-being. Journal of Marriage and Family, 71, 1254 – 1270. [Google Scholar]
  52. Stack S (1998). Marriage, family, and loneliness: A cross-national study. Sociological Perspectives, 41, 415 – 432. [Google Scholar]
  53. Stack S, & Eshleman RJ (1998). Marital status and happiness: A 17-nation study. Journal of Marriage and the Family, 60, 527 – 536. [Google Scholar]
  54. Stutzer A, & Frey BS (2006). Does marriage make people happy, or do happy people get married? The Journal of Socioeconomics, 35, 326 – 347. [Google Scholar]
  55. Tan JJX, Kraus MW, Carpenter NC, & Adler NE (2020). The association between objective and subjective socioeconomic status and subjective well-being: A meta-analytic review. Psychological Bulletin, 146(11), 970–1020. [DOI] [PubMed] [Google Scholar]
  56. Wadsworth T (2016). Marriage and subjective well-being: How and why context matters. Social Indicators Research, 126, 1025 – 1048. [Google Scholar]
  57. Williams K (2003). Has the future of marriage arrived? A contemporary examination of gender, marriage, and psychological well-being. Journal of Health and Social Behavior, 44, 470 – 487. [PMC free article] [PubMed] [Google Scholar]
  58. Wu Z, Penning MJ, Pollard MS, & Hart R (2003). In sickness and in health: Does cohabitation count? Journal of Family Issues, 24, 811 – 838. [Google Scholar]
  59. Zimmerman AC, & Easterlin RA (2006). Happily ever after? Cohabitation, marriage, divorce, and happiness in Germany. Population and Development Review, 32, 511 – 528. [Google Scholar]

RESOURCES