Why Marriage Matters, Second Edition: Twenty-Six Conclusions from the Social Sciences

Why Marriage Matters, Second Edition: Twenty-Six Conclusions from the Social Sciences

Family Scholars

Institute for American Values, 2005 - 44 pages

Sixteen social scientists summarize marriage-related research into an extremely useful, succinct form.

Subject: Marriage

More by: Family Scholars

About this Report

In all too many communities in the United States, especially poor and minority ones, marriage is in retreat. The statistics tell part of the story. In 1960, 5 percent of children were born outside of wedlock. Today, 34 percent of children are born outside of wedlock. In 1960, more than 67 percent of adults were married. Today, fewer than 56 percent of adults are married. As a consequence, American children are much less likely to spend their entire childhood in an intact, married family than they were 50 years ago. Likewise, men and women are less likely than they were 50 years ago to get married as a young adult and stay married. The bottom line is this: The institution of marriage has less of a hold over American men, women, and children than it did earlier in the last century.

These trends are even more dramatic in minority and lower income communities. In 2002, 68 percent of African American births and 44 percent of Latino births were out of wedlock, compared to 29 percent of white births. Similarly, although only about 5 percent of college-educated mothers have children out of wedlock, approximately 25 percent of mothers without a high school degree have children outside marriage.[1] Most of the women in the latter group hail from low-income families. African Americans and men and women without college degrees are also significantly more likely to divorce than their Anglo college-educated peers.[2]

The changes that have swept over American families in the last two generations have inspired a large body of social scientific research and a growing number of marriage education programs aimed at better preparing couples for marriage and better equipping couples with the knowledge, values, and skills required for successful marriage in today's world. This report, the second edition of Why Marriage Matters, is an attempt to summarize the research into a succinct form useful to Americans on all sides of ongoing family debates – to report what we know about the importance of marriage for our families and for our society.

What does the social science tell us? In addition to reviewing research on family topics covered in the first edition of the report, this report highlights five new themes in marriage-related research.

Five New Themes

  1. Even though marriage has lost ground in minority communities in recent years, marriage has not lost its value in these communities. This report shows that African Americans and Latinos benefit from marriage in much the same way that Anglos benefit from marriage. We also present evidence that marriage matters in countries, such as Sweden, that have markedly different approaches to public policy, social welfare, and religion than does the United States. In other words, marriage is a multicultural institution.
  2. An emerging line of research indicates that marriage benefits poor Americans, and Americans from disadvantaged backgrounds, even though these Americans are now less likely to get and stay married. Among other findings, this report shows that women from disadvantaged backgrounds who marry and stay married are much less likely to suffer poverty or other material hardship compared to their peers who do not marry.
  3. Marriage seems to be particularly important in civilizing men, turning their attention away from dangerous, antisocial, or self-centered activities and towards the needs of a family. Married men drink less, fight less, and are less likely to engage in criminal activity than their single peers. Married husbands and fathers are significantly more involved and affectionate with their wives and children than men in cohabiting relationships (with and without children). The norms, status rewards, and social support offered to men by marriage all combine to help men walk down the path to adult responsibility.
  4. Beyond its well-known contributions to adult health, marriage influences the biological functioning of adults and children in ways that can have important social consequences. For instance, marriage appears to drive down testosterone in men, with clear consequences for their propensity to aggression. Girls who grow up in non-intact families – especially girls who are exposed to unrelated males in their homes – are more likely to experience premature sexual development and, consequently, are more likely to have a teenage pregnancy. Thus, marriage, or the lack thereof, appears to have important biosocial consequences for men, women, and children.
  5. We find that the relationship quality of intimate partners is related both to their marital status and, for married adults, to the degree to which these partners are normatively committed to marriage. So, claims that love, not marriage, are crucial to a happy family life do not hold up. Marriage matters even or especially when it comes to fostering high-quality intimate relationships.

In summarizing marriage-related findings, we acknowledge that social science is better equipped to document whether certain social facts are true than to say why they are true. We can assert more definitively that marriage is associated with powerful social goods than that marriage is the sole or main cause of these goods.

A Word about Selection Effects

Good research seeks to tease out "selection effects," or the pre-existing differences between individuals who marry, become unwed parents, or divorce. Does divorce cause poverty, for example, or is it simply that poor people are more likely to divorce? Good social science attempts to distinguish between causal relationships and mere correlations in a variety of ways. The studies cited here are for the most part based on large, nationally representative samples that control for race, education, income, and other confounding factors. In many, but not all cases, social scientists have been able to use longitudinal data to track individuals as they marry, divorce, or stay single, increasing our confidence that marriage itself matters. Where the evidence is, in our view, overwhelming that marriage causes increases in well-being, we say so. Where the causal pathways are not as well understood, we are more cautious.

We recognize that, absent random assignment to marriage, divorce, or single parenting, social scientists must always acknowledge the possibility that other factors are influencing outcomes. Reasonable scholars may and do disagree on the existence and extent of such selection effects and the extent to which marriage is causally related to the better social outcomes reported here.

Nevertheless, scholarship is getting better in addressing selection effects. For instance, in this report we summarize two divorce studies that follow identical and non-identical adult twins in Australia to see to what extent the effects of divorce on their children are genetic and to what extent the effects of divorce on their children seem to be a consequence of divorce itself. Methodological innovations like these, as well as complex analyses using econometric models, are affording us greater confidence that family structure exercises a causal influence for some outcomes.

Of course individual circumstances vary.[3] While divorce is associated with increased risks of serious psychological and social problems for children, for example, about 75 percent of children of divorce do not suffer such problems (compared to approximately 90 percent of children from intact families).[4] While marriage is a social good, not all marriages are equal. Research does not generally support the idea that remarriage is better for children than living with a single mother.[5] Unhappy marriages do not have the same benefits as the average marriage.[6] Divorce or separation provides an important escape hatch for children and adults in violent or high-conflict marriages. Families, communities, and policy makers interested in distributing the benefits of marriage more equally must do more than merely discourage legal divorce.

Despite its inherent limitations, good social science is a better guide to social policy than uninformed opinion or prejudice. The public and policy makers deserve to hear what research suggests about the consequences of marriage and its absence for children and adults. This report represents our best judgment of what the current social science evidence reveals about the importance of marriage in our social system.

Fundamental Conclusions

Here are our three fundamental conclusions:

  1. Marriage is an important social good, associated with an impressively broad array of positive outcomes for children and adults alike.
  2. Marriage is an important public good, associated with a range of economic, health, educational, and safety benefits that help local, state, and federal governments serve the common good.
  3. The benefits of marriage extend to poor and minority communities, despite the fact that marriage is particularly fragile in these communities.

Family structure and processes are of course only one factor contributing to child and social well-being. Our discussion here is not meant to minimize the importance of other social and economic factors, such as poverty, child support, unemployment, teenage childbearing, neighborhood safety, or the quality of education for both parents and children. Marriage is not a panacea for all of our social ills. For instance, when it comes to child well-being, research suggests that family structure is a better predictor of children's psychological and social welfare, whereas poverty is a better predictor of children's educational attainment.[7]

But whether American society and, indeed, the world, succeeds or fails in building a healthy marriage culture is clearly a matter of legitimate public concern. In particular, marriage is an issue of paramount importance if we wish to help the most vulnerable members of our society: the poor, minorities, and children.


  1. David Ellwood and Christopher Jencks, 2004. "The Spread of Single-Parent Families in the United States Since 1960," in D.P. Moynihan et al. (eds.) The Future of the Family (New York: Russell Sage), 25-65.
  2. Steven P. Martin, 2005. "Growing Evidence for a 'Divorce Divide'? Education and Marital Dissolution Rates in the U.S. Since the 1970s" (College Park, MD: University of Maryland Department of Sociology), unpublished manuscript.
  3. See, for example, R.E. Heyman et al., 2001. "The Hazards of Predicting Divorce without Crossvalidation," Journal of Marriage and Family 63: 473-479.
  4. E. Mavis Hetherington and John Kelly, 2002. For Better or For Worse: Divorce Reconsidered (New York: W.W. Norton & Co.).
  5. For example, Hanson et al. find that remarriage decreases parental supervision and lowers college expectations for children. McLanahan and Sandefur show that children whose mothers remarried had no higher rate of high school graduation (or lower levels of teen childbearing) than did children living with single mothers. T. Hanson et. al., 1998. "Windows on Divorce: Before and After," Social Science Research 27: 329-349; Sara McLanahan and Gary Sandefur, 1994. Growing Up With a Single Parent: What Hurts, What Helps (Cambridge: Harvard University Press).
  6. See, for example, Kristi Williams, 2001. "Has the Future of Marriage Arrived? A Contemporary Examination of the Effects of Marital Status and Marital Quality on the Psychological Well-Being of Women and Men," paper presented at the Annual Meeting of the National Council on Family Relations (Rochester, NY).
  7. Sara S. McLanahan, 1997. "Parent Absence or Poverty: Which Matters More?" in G. Duncan and J. Brooks-Gunn (eds.) Consequences of Growing Up Poor (New York: Russell Sage), 35-48.

About this Report

Why Marriage Matters, Second Edition comes from a team of family scholars chaired by W. Bradford Wilcox of the University of Virginia, William Doherty of the University of Minnesota, Norval Glenn of the University of Texas, and Linda Waite of the University of Chicago. The project is sponsored by the Institute for American Values.

© 2005, Institute for American Values


Institute for American Values, 420 Lexington Avenue, Room 300, New York, NY 10170-0399