Cohabitation, Marriage, Divorce, and Remarriage in the United States

Report

By: Matthew D. Bramlett and William D. Mosher

Date: July 2002

Source: Centers for Disease Control. "Cohabitation, Marriage, Divorce, and Remarriage in the United States." July 2002 〈http://www.cdc.gov/nchs/data/series/sr_23/sr23_022.pdf〉 (accessed June 19, 2006).

About the Author: Matthew Bramlett and William Mosher are researchers with the Division of Vital Statistics in the U.S. Department of Health and Human Services.

INTRODUCTION

During the twentieth century, marriage underwent sweeping changes in the United States. From 1960 to 2000, more couples opted to forego marriage entirely, and the number of unmarried couples living together increased by a factor of ten. Marriages also dissolved at an increasing pace: divorce rates more than doubled from 1.5 divorces per 1,000 residents in 1910 to 3.7 in 2004. Current U.S. divorce rates remain well above those of other industrialized nations, including Canada (2.0 per 1,000) and Spain (0.6 per 1,000).

The total cost of failed marriages in the U.S. is difficult to quantify; however, in 2000 the federal government spent $150 billion on direct aid to single-parent households. The overwhelming majority of divorced mothers with children face financial hardship, often living below the federal poverty line. Children of divorced parents are more likely to drop out of school, experience general health problems, and enter the juvenile justice system.

Given the enormous direct and indirect costs associated with divorce in America, the federal government has significant incentive to understand why people choose to marry, live together, divorce, and remarry. In order to better understand the motivations for these choices, the U.S. Department of Health and Human Services has conducted the National Survey on Family Growth (NSFG). The 1995 edition of this survey was the fifth iteration, utilizing personal interviews with more than 10,000 women between the ages of 15 and 44. The result was a statistically intensive report detailing the impact of factors such as age, race, and income on a woman's likelihood to cohabit, marry, divorce, and remarry.

PRIMARY SOURCE

What are the trends? Our data show an increase in the chances that first marriages will end (in separation or divorce) for marriages that began in the 1950s through the 1970s. From the early 1970s to the late 1980s, the rates of breakup were fairly stable. The probability of remarriage following divorce has decreased slightly, and the probability that the second marriage will break up has risen from the 1950s to the 1980s.

Do the trends differ by race/ethnicity? It appears that these trends were similar for non-Hispanic white and non-Hispanic black women, but black women faced higher rates of marital breakup, lower rates of making the transition from separation to divorce, and lower rates of remarriage. Among white women, the increasing probability of first marriage breakup leveled off in the 1970s but appears to have continued rising for black women through the 1980s.

Are characteristics of communities related to success in marriage? This report shows clear evidence that community prosperity is related to successful cohabitations and marriages, and that neighborhood poverty increases the likelihood that cohabitations and marriages will fail.

Is the statistical portrait of union formation and dissolution affected if we measure unmarried cohabita-tion and separation from marriage as well as legal divorce? One major advantage of survey data on marriage is that we are not limited to examining legal marriage and divorce. The data in this report show that the probability that an intact premarital cohabitation will result in marriage is 70 percent after 5 years; that probability is associated with the woman's race, age, education, the household's income, and the economic opportunities in the community. The data also show that a great many marriages end in legal separation but not in divorce, and that looking only at divorce greatly understates marital disruption among some groups—especially non-Hispanic black and Hispanic women.

What demographic, economic, and social factors affect the chances that marriage will succeed or fail? This report shows that a number of characteristics are closely associated with the chances that a marriage will continue or break up. For first marriages, for example, marriages are less likely to break up, and more likely to succeed, if the wife grew up in a two-parent home, is Asian, was 20 years of age or over at marriage, did not have any children when she got married, is college-educated, has more income, or has any religious affiliation.

The following highlights illustrate the kinds of findings shown in this report:

The probability of first marriage is lower for non-Hispanic black women than for other women. Getting married by the 18th birthday is more likely for Hispanic and non-Hispanic white women and less likely for non-Hispanic black and Asian women. First marriage is less likely for women who report that their religion is not important. Early marriage is more likely for women in communities with higher male unemployment, lower median family income, higher poverty and higher receipt of welfare. First marriage is more likely in nonmetropolitan areas and less likely in central cities.

The probability that an intact first premarital cohabitation becomes a marriage is higher among white women and lower among black women; higher among couples with higher incomes than for couples with lower incomes; and higher for cohabiting women with any religious affiliation than for those with no religious affiliation, especially among white women. Marriage is more likely for cohabiting white women who report that their religion is either somewhat or very important than for those who report that their religion is not important.

Cohabiting women are more likely to marry if they live in communities with lower male unemployment, higher median family income, lower poverty, and lower receipt of welfare. The male unemployment rate seems to be more important among black women than among white women.

After the first 3 years of cohabitation, the probability that a first premarital cohabitation breaks up is higher among black women than among Hispanic or white women and is higher among younger than older women, especially among white women. Women who have ever been forced to have intercourse before the cohabitation began are more likely to experience the breakup of their first premarital cohabitation than women who have never been forced.

Cohabiting women are more likely to experience the breakup of their first premarital cohabitation if they live in communities with higher male unemployment, lower median family income, and higher rates of poverty and receipt of welfare.

Black women are more likely to experience first marital disruption and Asian women are less likely to experience first marital disruption, compared with white or Hispanic women. First marriages of women who are 20 years of age or over at marriage are less likely to break up than marriages of teenaged brides; but there is no significant difference by age at marriage among Hispanic women. Women whose religion is somewhat or very important are also less likely to experience a breakup of their first marriage than those whose religion is not important. Women who lived with both parents throughout childhood are less likely to experience the breakup of their first marriage than women who were not raised with two parents throughout childhood.

SIGNIFICANCE

Interpreting the National Survey of Family Growth is challenging; the actual report spans 103 pages, of which more than half are numeric tables. While this level of detail is necessary for thorough statistical analysis, most readers are more interested in a synopsis of the findings, which is provided near the beginning of the report. In summary the report's authors conclude that divorce rates are presently stable, that marriages fail more often in poorer communities and neighborhoods, and that better education, higher age and income, and religious affiliation are associated with lower rates of divorce in a first marriage.

General findings such as these do not require exhaustive data analysis and are often available elsewhere. The value of a study such as the NSFG lies in its ability to answer very specific questions. For example, consider a social service agency attempting to reduce divorce rates in a specific urban community. The workers in this agency know that demographic factors play a role in divorce; they also understand that the specific causes of divorce vary among ethnic groups. In such cases the NSFG allows investigators to determine specific factors relevant to their community.

For example, national statistics on first marriage breakup show that divorce rates climbed more slowly during the 1970s. However, this improvement was not consistent across ethnic groups, and rates continued to rise more sharply among black women through the following decade. Such information could help the previously mentioned agency decide where to focus their efforts.

An exhaustive analysis such as the NSFG is also useful in answering some long-debated questions. Two distinct schools of thought exist regarding cohabitation, or living together while unmarried. Couples who choose to cohabitate often characterize it as a low-risk test of compatibility, potentially less risky and less complicated than a marriage and subsequent divorce. Conservative religious groups tend to be critical of such arrangements, arguing that they are emotionally unhealthy and make commitment more difficult, ultimately harming marriages.

Which perspective on cohabitation does the study data support? As often occurs in data analysis, the answer is complex. First, the relationship between premarital cohabitation and marriage is influenced by factors such as race and income level; for example, white women and those with higher incomes are more likely to move from cohabitation to marriage than black women and those earning less. In general, the study finds that after three years of cohabitation, white couples have a 60 percent chance of marrying, while the rate for Hispanics is 50 percent and the rate for blacks is 35 percent. During the first ten years of cohabitation, couples who began cohabiting while under age twenty-five face a 60 percent chance of ultimately separating.

When interpreting data from the NSFG and similar reports, care must be taken to correctly interpret the results. For example, most research on marriage and divorce measures only actual divorces and not separations. Since Americans sometimes choose to end their marriages by separating but not legally divorcing, the actual number of failed marriages is probably higher than the figures provided by the NSFG indicate. Another concern in such a study deals with social desirability bias, or the tendency for respondents to lie about socially unacceptable behaviors such as infidelity. This bias may result in inaccurate assessments of the frequencies of specific behaviors.

While the NSFG provides voluminous information on when and why Americans marry, divorce, and remarry, it does not prescribe solutions for the problems it illuminates. The study's purpose is to provide comprehensive data for use by organizations that provide solutions, allowing them to more efficiently apply their resources.

FURTHER RESOURCES

Books

Tucker, M. and C. Mitchell-Kernan. The Decline in Marriage among African Americans. New York: Russell Sage, 1995.

Waite, L. J. and M. Gallagher. The Case for Marriage: Why Married People Are Happier, Healthier and Better Off. New York: Doubleday, 2000.

Zill, N. and M. Gallagher. Running in Place: How American Families Are Faring in a Changing Economy and an Individualistic Society. Washington, D.C.: Child Trends, 1994.

Periodicals

Amato, P. R. "The Consequences of Divorce for Adults and Children." Journal of Marriage and the Family 62 (2000): 1269-1287.

Duncan, G. and S. Hoffman. "A Reconsideration of the Economic Consequences of Marital Dissolution." Demography 22 (1985): 485-497.

Kunz, J. "The Intergenerational Transmission of Divorce: A Nine Generation Study." Journal of Divorce and Remarriage 34 (2000): 169-175.

Web sites

Discovery Health. "Debunking Divorce Myths." 〈http://health.discovery.com/centers/loverelationships/articles/divorce.html"〉 (accessed June 19, 2006).

Journal of Family Violence. "Divorce Related Malicious Mother Syndrome." 1995 〈http://www.fact.on.ca/Info/pas/turkat95.htm"〉 (accessed June 19, 2006).

Los Angeles County Law Library. "California Divorce Pathfinder." 〈http://lalaw.lib.ca.us/divorce.html"〉 (accessed June 19, 2006).