July 1, 2007

As Marriage and Parenthood Drift Apart, Public Is Concerned about Social Impact

Generation Gap in Values, Behaviors

Executive Summary

Overview

Figure

Americans believe that births to unwed women are a big problem for society, and they take a mixed view at best of cohabitation without marriage. Yet these two nontraditional behaviors have become commonplace among younger adults, who have a different set of moral values from their elders about sex, marriage and parenthood, a new Pew Research Center Survey finds.

This generational values gap helps to explain the decades-long surge in births to unmarried women, which now comprise nearly four-in-ten (37%) births in the United States as well as the sharp rise in living together without getting married, which, the Pew survey finds, is something that nearly half of all adults in their 30s and 40s have done for at least a portion of their lives.

But this generational divide is only part of a more complex story. Americans of all ages, this survey finds, acknowledge that there has been a distinct weakening of the link between marriage and parenthood. In perhaps the single most striking finding from the survey, just 41% of Americans now say that children are “very important” to a successful marriage, down sharply from the 65% who said this in a 1990 survey.

Figure

Indeed, children have fallen to eighth out of nine on a list of items that people associate with successful marriages — well behind “sharing household chores,” “good housing,” “adequate income,” “happy sexual relationship,” and “faithfulness.” Back in 1990, when the American public was given this same list on a World Values Survey, children ranked third in importance.

The new Pew survey also finds that, by a margin of nearly three-to-one, Americans say that the main purpose of marriage is the “mutual happiness and fulfillment” of adults rather than the “bearing and raising of children.”

In downgrading the importance of children to marriage, public opinion both reflects and facilitates the upheavals in marital and parenting patterns that have taken place over the past several decades.

In the United States today, marriage exerts less influence over how adults organize their lives and how children are born and raised than at any time in the nation’s history. Only about half of all adults (ages 18 and older) in the U.S. are married; only about seven-in-ten children live with two parents; and nearly four-in-ten births are to unwed mothers, according to U.S. Census figures. As recently as the early 1970s, more than six-in-ten adults in this country were married; some 85% of children were living with two parents; and just one-birth-in-ten was to an unwed mother.

Figure

Americans take a dim view of these trends, the Pew survey finds. More than seven-in-ten (71%) say the growth in births to unwed mothers is a “big problem.” About the same proportion — 69% — says that a child needs both a mother and a father to grow up happily.

Not surprisingly, however, attitudes are much different among those adults who have themselves engaged in these nontraditional behaviors. For example, respondents in the survey who are never-married parents (about 8% of all parents) are less inclined than ever-married parents to see unmarried childbearing as bad for society or morally wrong. They’re also less inclined to say a child needs both a mother and father to grow up happily. Demographically, this group is more likely than ever-married parents to be young, black or Hispanic,1 less educated, and to have been raised by an unwed parent themselves.

Figure

There is another fast-growing group — cohabiters — that has a distinctive set of attitudes and moral codes about these matters. According to the Pew survey, about a third of all adults (and more than four-in-ten adults under age 50) have, at some point in their lives, been in a cohabiting relationship with a person to whom they were not married. This group is less likely that the rest of the adult population to believe that premarital sex is wrong. They’re less prone to say that it’s bad for society that more people are living together without getting married. Demographically, this group is more likely than the rest of the adult population to be younger, black, and secular rather than religious.

Marriage

But while this survey finds that people in nontraditional marital and parenting situations tend to have attitudes that track with their behaviors, it does not show that they place less value than others on marriage as a pathway to personal happiness.

To the contrary, both the never-married parents as well as the cohabiters in our survey tend to be more skeptical than others in the adult population that a person can lead a complete and fulfilled life if he or she remains single. This may reflect the fact that never-married parents as well as cohabiters tend to be less satisfied with their current lives than is the rest of the population. For many of them, marriage appears to represent an ideal albeit an elusive, unrealized one.

Along these same lines, the survey finds that low income adults are more likely than middle income or affluent adults to cite the ability to meet basic economic needs (in the form of adequate income and good housing) as a key to a successful marriage. Adults with lower socioeconomic status — reflected by either education or income levels — also are less likely than others to marry, perhaps in part because they can’t meet this economic bar.

And it’s this decline in marriage that is at the heart of the sharp growth in nonmarital childbearing. This trend has not been primarily driven — as some popular wisdom has it — on an increase in births to teenage mothers. To the contrary, those rates have been falling for several decades. Rather the sharp increase in nonmarital births is being driven by the fact that an ever greater percentage of women in their 20s, 30s and older are delaying or forgoing marriage but having children.

Figure

The Pew survey was conducted by telephone from February 16 through March 14, 2007 among a randomly selected, nationally-representative sample of 2,020 adults. It has a margin of sampling error of 3 percentage points.

Children

The survey finds that while children may have become less central to marriage, they are as important as ever to their parents. Asked to weigh how important various aspects of their lives are to their personal happiness and fulfillment, parents in this survey place their relationships with their children on a pedestal rivaled only by their relationships with their spouses — and far above their relationships with their parents, friends, or their jobs or career. This is true both for married and unmarried parents. In fact, relatively speaking, children are most pre-eminent in the lives of unwed parents.

Figure

The survey also finds that Americans retain traditional views about the best family structure in which to raise children. More than two-thirds (69%) say that a child needs both a mother and father to grow up happily. This question has been posed periodically over the past quarter century,2 and — even as the percentage of children who live with both a mother and father has dropped steadily during this time period — public opinion has remained steadfastly in favor of a home with a mom and a dad.

In keeping with these traditional views, the public strongly disapproves of single women having children. Among the various demographic changes that have affected marriage and parenting patterns in recent decades — including more women working outside the home; more people living together without getting married; more first marriages at a later age; and more unmarried women having children — it’s the latter trend that draws the most negative assessments by far.

Two-thirds (66%) of all respondents say that single women having children is bad for society, and nearly as many (59%) say the same about unmarried couples having children. No other social change we asked about in this particular battery drew a thumbs-down from more than half of respondents.

Divorce

Figure

While the public strongly prefers the traditional mother-and-father home, this endorsement has some clear limits. By a margin of 67% to 19%, Americans say that when there is a marriage in which the parents are very unhappy with one another, their children are better off if the parents get divorced. Similarly, by a margin of 58% to 38%, more Americans agree with the statement that “divorce is painful, but preferable to maintaining an unhappy marriage” than agree with the statement that “divorce should be avoided except in an extreme situation.”

Thus, public attitudes toward divorce and single parenting have taken different paths over the past generation. When it comes to divorce, public opinion has become more accepting.3

When it comes to single parenting, public opinion has remained quite negative.

The oddity is that rates of divorce, after more than doubling from 1960 to 1980, have declined by about a third in recent decades, despite this greater public acceptance. On the other hand, the rates of births to unwed mothers have continued to rise, despite the steadfast public disapproval. Some 37% of all births in the U.S. in 2005 were to an unwed mother, up from just 5% in 1960. This rapid growth is not confined to the U.S. Rates of births to unwed mothers also have risen sharply in the United Kingdom and Canada, where they are at about the same levels as they are in the U.S. And they’ve reached even higher levels in Western and Northern European countries such as France, Denmark and Sweden.

Public Opinion by Demographic Groups

The group differences in public opinion on these matters tend to be correlated with age, religion, race and ethnicity, as well as with the choices that people have made in their own marital and parenting lives. There are some, but not many, differences by gender. Here is a rundown of the key differences by group.

Age, Religiosity and Political Conservatism

Figure

As noted above, the Pew survey finds that older adults — who came of age prior to the social and cultural upheavals of the 1960s — are more conservative than younger and middle-aged adults in their views on virtually all of these matters of marriage and parenting. Thus, some of the overall change in public opinion is the result of what scholars call “generational replacement.” That is, as older generations die off and are replaced by younger generations, public opinion shifts to reflect the attitudes of the age cohorts that now make up the bulk of the adult population.

Even among the younger generations (ages 18 to 64), however, our survey finds substantial differences in attitudes that fall along the fault lines of religion and ideology rather than age.

White evangelical Protestants and people of all faiths who attend religious services at least weekly hold more conservative viewpoints on pretty much the whole gamut of questions asked on the Pew survey. This is true across all age groups. For example, white evangelical Protestants are more likely than other religious groups to consider premarital sex morally wrong.

Figure

They are more likely to consider the rise in unmarried childbearing and cohabitation bad for society and more likely to agree that a child needs both a mother and father to be happy. They also are more likely to say legal marriage is very important when a couple plans to have children together or plans to spend the rest of their lives together. Further, white evangelical Protestants are more likely than white mainline Protestants to say that divorce should be avoided except in extreme circumstances and to consider it better for the children when parents remain married, though very unhappy with each other. In sum, white evangelical Protestants have a strong belief in the importance of marriage and strong moral prescriptions against premarital sex and childbearing outside of marriage.

The pattern is the same among those of any faith who attend religious services more frequently, compared with less frequent attendees. And it is the same for political conservatives compared with their more moderate or liberal counterparts.

Race and Ethnicity

The racial and ethnic patterns in public opinion on these matters are more complex. Blacks and Hispanic are more likely than whites to bear children out of wedlock. And yet these minority groups, our survey finds, also are more inclined than whites to place a high value on the importance of children to a successful marriage. Indeed, they place higher value than whites do on the importance of most of the ingredients of a successful marriage that this survey asked about — especially the economic components. But blacks and Hispanics are less likely than whites to be married. One possible explanation to emerge from this survey is that many members of these minority groups may be setting a high bar for marriage that they themselves cannot reach, whether for economic or other reasons.

Figure

As noted above, there are sharp generational differences in views about the morality of unwed parenting. However, there is no significant difference on this front by race or ethnicity; blacks, Hispanics and whites are about equally likely to say it is wrong for unmarried women to have children. There are small differences along racial and ethnic lines when it comes to evaluating the impact on society of the growing numbers of children born out of wedlock. Hispanics are somewhat less negative about this phenomenon than are whites and blacks, between whom there is no statistically significant difference.

When it comes to the relationship between marriage and children, Hispanics again stand out. They are more inclined than either whites or blacks to consider having and raising children to be the main purpose of marriage (even so, however, a majority of Hispanics say that adult happiness and fulfillment is the main purpose of a marriage). Also, Hispanics — more so than either blacks or whites — consider children “very important” for a successful marriage.

But when considering a broader range of characteristics of a successful marriage, it is whites who stand apart. They are much less likely than either blacks or Hispanics to consider adequate income, good housing and children to be “very important” to a successful marriage. And they are somewhat less likely to rate various measures of compatibility (see chart) as being important as well. To some degree all these racial and ethnic differences reflect the differing socioeconomic circumstances of whites, blacks and Hispanics. People with higher incomes and education levels — regardless of their race and ethnicity — tend to rate these various characteristics as less important to marriage than do people with a lower socio-economic status.

Figure

When it comes to views about divorce, whites and, especially, Hispanics are more likely than blacks to say that divorce is preferable to maintaining an unhappy marriage. However, about two-thirds of all three groups say that it is better for the children if their very unhappy parents divorce rather than stay together.

Views about cohabitation are similar for blacks and whites, while Hispanics are a bit less negative about the impact of cohabitation on society. But the similarities between blacks and whites masks divisions of opinion within each group. Among whites, the difference of opinion between generations is particularly sharp — with 55% of whites ages 50 and older saying that living together is bad for society, compared with 38% among younger whites, a difference of 17 percentage points. The comparable difference between older and younger blacks is just 9 percentage points. Among older blacks and whites, the balance of opinion is tilted in the negative direction. For younger whites (ages 18 to 49), a plurality hold a neutral assessment of the impact on society of couples living together without marrying. Among younger blacks, opinion about cohabitation is more divided; 48% of this group considers living together bad for society while 45% take a neutral position and 5% say it is good for society.

To some degree, views about cohabitation reflect differing moral assessments of premarital sex. Blacks are more likely than whites and Hispanics to say that premarital sex is always or almost always morally wrong — and this is true even after group differences in age are taken into account. Those who consider premarital sex wrong also tend to consider cohabitation bad for society, while those who say premarital sex is not wrong or is only wrong in some circumstances are more likely to say the cohabitation trend makes no difference for society.

Figure

When it comes to marital and parenting behaviors (as opposed to attitudes), a number of racial and ethnic patterns stand out.

More than eight-in-ten white adults in this country have been married, compared with just seven-in-ten Hispanic adults and slightly more than half (54%) of all black adults. Among blacks, there is a strong correlation between frequent church attendance, moral disapproval of premarital sex and the tendency to marry. Among whites (who marry at much higher rates) this relationship is not as strong.

Among those who have ever been married, blacks (38%) and whites (34%) are more likely than Hispanics (23%) to have been divorced. Blacks also are somewhat more likely than whites or Hispanics to have cohabited without marriage. But all three groups, this survey finds, are equally likely to have had children.

Blacks and Hispanics are more likely than whites to have children out of wedlock. For all groups, this behavior also is strongly correlated with lower educational attainment. For blacks and Hispanics (more so than for whites), frequent church attendance correlates negatively with the likelihood of being an unwed parent.

Gender

The Pew survey finds a great deal of common ground between men and women on issues surrounding marriage and parenting. There are some small differences, however. While men and women are about equally likely to see unmarried parenting as a problem for society, men are a bit more negative than women about unmarried parenting when no male partner is involved in raising the children. Similarly, men are a little more likely than women to believe that children need both a mother and father to be happy. Women, on the other hand, are a bit more likely than men to consider divorce preferable to maintaining an unhappy marriage; they also believe more strongly than men that divorce is the better option for children when the marriage is very unhappy. On other matters — such as the main purpose of marriage or the characteristics of a successful marriage — there are few differences.

Education and Income

College-educated adults and high-income adults marry at higher rates and divorce at lower rates than do those with less education and income. They are also less likely to have children outside of marriage.4

However, despite the sharp differences by socio-economic status in marital and parenting behaviors, there are only minor differences by socio-economic status in values and attitudes about marriage and parenting. Adults with higher incomes and more education tend to be slightly less inclined than others to say that premarital sex and nonmarital births are always morally wrong. The college educated also are slightly less inclined than the less educated to say it is very important for couples to legally marry if they plan to spend their lives together. Similarly, those with a college education are a little more likely to say that a man or woman can lead a complete and happy life if he or she remains single.

There are no more than minimal differences by education or income when it comes to views about the impact on society of unmarried childbirths and of cohabitation.

Family Background

The Pew survey finds some strong correlations between the kinds of family arrangements that respondents experienced growing up and their own behaviors in adulthood. For example, among respondents who are themselves products of parents who never married, about a third (32%) are themselves never-married parents. By comparison, just 5% of the general adult population are products of never-married parents.

Family background in childhood plays a smaller role, however, in predicting adult attitudes (as opposed to behaviors) about whether unmarried parenting is bad for society and morally wrong. Once age differences are taken into account, those whose parents never married are just a bit less negative than those whose parents married and never divorced about the impact of unmarried childbearing on society.

Respondents with parents who divorced are just as likely as other respondents to take the position that divorce is painful but preferable to maintaining an unhappy marriage. Similarly, among people ages 18 to 49, the now grown children of divorce hold about the same views as those who grew up in a traditional-married-parent arrangement on whether divorce is better for children than parents staying in an unhappy marriage. On the other hand, those respondents whose parents divorced are less likely than other respondents to believe that a child needs a home with both a mother and a father to grow up happily.

Moral Beliefs, Attitudes and Behaviors

Figure

There are close relationships between behaviors, attitudes and moral beliefs when it comes to the subjects of unwed parenting and cohabitation, the Pew survey finds. For example, those who have fewer moral reservations about premarital sex and are positive or neutral about the impact of living together on society also are more likely to have lived with a partner themselves. Similarly, those who are positive or neutral about the social impact of unmarried parenting and less likely to consider it morally wrong are also more likely to be in this situation themselves. It is not possible from this survey to disentangle which came first — the moral beliefs, the attitudes, or the behaviors — but it is clear they tend to go hand-in-hand.

Statistical analysis of these survey findings shows that having less education and being black or Hispanic are traits associated with being a never-married parent. Attending religious services less often also is associated with being an unmarried parent, particularly among blacks and Hispanics.

Figure

On the other side of the coin, those who believe that having children without being married is wrong are less likely to be a never-married parent. Also, those who consider the rise in unmarried parents bad for society are less likely to be unmarried parents.

A statistical analysis of factors correlated with ever having lived with a partner outside of marriage shows that cohabiters are younger, more likely to be black, and, after controlling for other demographic factors, less likely to be Hispanic. They are also less likely to attend religious services frequently. There is a strong relationship between moral beliefs about premarital sex and cohabitation history; those who consider premarital sex always wrong are less likely to have cohabited than others. They are also less likely to have cohabited than those who say living together is bad for society — suggesting that the more powerful stigma against cohabitation comes from concerns about morality rather than from concerns about social consequences.

A different pattern emerges when looking at differences between married people who have — and haven’t — been divorced. Here, the demographic and attitudinal factors do little to predict the probability of experience with divorce.

Figure

There are a few exceptions, however. Catholics are bit less likely than members of other religious groups to have been divorced. And there is a modest correlation between having been divorced and believing that divorce is better for the children than maintaining a very unhappy marriage.

But in the main, experience with divorce cuts across all demographic subgroups more evenly than does experience either with unmarried parenting or with cohabitation. The belief that divorce is preferable to maintaining an unhappy marriage is widely shared by both those who have and have not been divorced.

Read the full report for more details.

  1. Throughout this report, the term blacks or whites refers to non-Hispanic blacks or whites, respectively. Hispanics are of any race. The survey included an oversample of blacks and Hispanics. Interviews were conducted in both English and Spanish.
  2. See the World Values Surveys conducted in the U.S. in 1982, 1990, 1995 and 1999.
  3. See for example, Thornton, Arland and Linda Young-DeMarco. 2001. "Four Decades of Trends in Attitudes Toward Family Issues in the United States: The 1960s Through the 1990s." Journal of Marriage and Family, 63: 1009-1037.
  4. Bianchi, Suzanne M. and Lynne M. Casper. 2000. "American Families." Population Bulletin 55(4). Population Reference Bureau.