A new "marriage gap" in the United States is increasingly aligned with a growing income gap.
Marriage, while declining among all groups, remains the norm for adults with a college education and good income but is now markedly less prevalent among those with less income and education, according to a new Pew Research Center study. The report is based on a new nationwide survey, conducted in association with TIME, and complemented by an analysis of demographic and economic data from the U.S. Census Bureau.
The survey finds that those who are less well-off are as likely as others to want to marry, but they place a higher premium on economic security as a condition for marriage. And this is a bar that many may not meet.
The survey also finds that Americans are both accepting and uneasy about the transformative trends of the past 50 years that have led to a sharp decline in marriage and a rise of new family forms, from single parenthood to blended families.
Against a backdrop of declining marriage rates, the survey finds that nearly four in 10 Americans say marriage is becoming obsolete. But even as marriage shrinks, family -- in all its emerging varieties -- remains resilient. The survey finds that Americans have an expansive definition of what constitutes a family. And the vast majority of adults consider their own family to be the most important, most satisfying element of their lives.
Key findings include:
- The Class-Based Decline in Marriage. About half (52 percent) of all adults in this country were married in 2008; back in 1960, seven in 10 (72 percent) were. This decline has occurred along class lines. In 2008, there was a 16-percentage-point gap in marriage rates between college graduates (64 percent) and those with a high school diploma or less (48 percent). In 1960, this gap had been just four percentage points (76 percent vs. 72 percent). The survey finds that those with a high school diploma or less are just as likely as those with a college degree to say they want to marry, but they place a higher premium than college graduates (38 percent versus 21 percent) on financial stability as a very important reason to marry.
- Is Marriage Becoming Obsolete? Nearly four in 10 survey respondents (39 percent) say that it is; in 1978, when Time magazine posed this question to registered voters, just 28 percent agreed. Those most likely to agree include those who are a part of the phenomenon (62 percent of cohabiting parents) as well as those most likely to be troubled by it (42 percent of self-described conservatives). Despite these growing uncertainties, Americans are more upbeat about the future of marriage and family (67 percent say they are optimistic) than about the future of the country's educational system (50 percent optimistic), its economic system (46 percent optimistic) or its morals and ethics (41 percent optimistic).
- An Ambivalent Public. The public's response to changing marital norms and family forms reflects a mix of acceptance and unease. On the troubled side of the ledger: Seven in 10 (69 percent) say the trend toward more single women having children is bad for society, and 61 percent say that a child needs both a mother and father to grow up happily. On the more accepting side, only a minority say the trends toward more cohabitation without marriage (43 percent), more unmarried couples raising children (43 percent), more gay couples raising children (43 percent) and more people of different races marrying (14 percent) are bad for society. Relatively few say any of these trends are good for society, but many say they make little difference.
- Group Differences. Where people stand on the various changes in marriage and family life depends to some degree on who they are and how they live. The young are more accepting than the old of the emerging arrangements; the secular are more accepting than the religious; liberals are more accepting than conservatives; the unmarried are more accepting than the married; and, in most cases, blacks are more accepting than whites. The net result of all these group differences is a nearly even three-way split among the full public. A third (34 percent) say the growing variety of family arrangements is a good thing; 29 percent say it is a bad thing and 32 percent say it makes little or no difference.
- The Resilience of Families. The decline of marriage has not knocked family life off its pedestal. Three-quarters of all adults (76 percent) say their family is the most important element of their life; 75 percent say they are "very satisfied" with their family life, and more than eight in 10 say the family they live in now is as close as (45 percent) or closer than (40 percent) the family in which they grew up. However, on all of these questions, married adults give more positive responses than do unmarried adults.
- The Definition of Family. By emphatic margins, the public does not see marriage as the only path to family formation. Fully 86 percent say a single parent and child constitute a family; nearly as many (80 percent) say an unmarried couple living together with a child is a family; and 63 percent say a gay or lesbian couple raising a child is a family. The presence of children clearly matters in these definitions. If a cohabiting couple has no children, a majority of the public says they are not a family. Marriage matters, too. If a childless couple is married, 88 percent consider them to be a family.
- The Ties that Bind. In response to a question about whom they would assist with money or caregiving in a time of need, Americans express a greater sense of obligation toward relatives -- including relatives by way of fractured marriages -- than toward best friends. The ranking of relatives aligns in a predictable hierarchy. More survey respondents express an obligation to help out a parent (83 percent would feel very obligated) or grown child (77 percent) than say the same about a stepparent (55 percent) or a step or half sibling (43 percent). But when asked about one's best friend, just 39 percent say they would feel a similar sense of obligation.
The survey was conducted from Oct.1-21, 2010, on landlines and cell phones with a nationally representative sample of 2,691 adults 18 and older. The margin of sampling error is plus or minus 2.6 percentage points for the full sample. Interviews were done in English and Spanish by Princeton Survey Research Associates International.