The most incisive guide to issues facing the American family today . . . An invaluable resource for anyone wishing to stay on the cutting edge of research on family trends.
-W. Bradford Wilcox
Associate Professor of Sociology, University of Virginia
The Book of Ecclesiastes observes that “of the making of books there is no end.” Likewise, of the research quantifying the negative outcomes of bearing children out of wedlock there is no end. Delivering some of the most recent findings, a study by a team of sociologists led by Kristi Williams of Ohio State uncovers through extensive analysis a strong negative relationship between the health of black and white women at midlife and premarital childbearing. In fact, the correlation is so robust that only a subsequent and enduring marriage to the child’s biological father demonstrates potential to moderate the health penalty.
The researchers analyzed data representing 3,391 mothers from the National Longitudinal Survey of Youth to measure the health of women at age 40 who experienced a premarital first birth before age 36, relative to their married peers whose first birth also occurred before age 36 but within the protective bonds of marriage. In their first set of regressions, the scholars found that these unwed mothers reported significantly lower levels of self-reported health (p<.05), even when controlling for background characteristics. Those background characteristics, which the researchers characterize as “highly significant predictors” of unwed childbearing, included health limitations at baseline, mother’s educational level, and scores on the Armed Forces Qualification Test.
Additional statistical models only confirmed the correlation. In tests that adjusted for racial and ethnic differences, the health penalty of premarital childbearing remained significant for both black and white women, although not for Hispanic women. Moreover, multivariate propensity-score matching (PSM)—a methodology designed to account for “selection effects” that might account for premarital births—yielded little differences in the results. Those covariates included such things as family status at age 14, poverty status at baseline, health limitations and cognitive abilities, self-efficacy, gender-role orientation, and future expectations as reported in 1979.
As if these tests weren’t enough, the researchers also sought to measure, using a sample limited to the 1,275 mothers with a non-marital first child, whether those who subsequently married or cohabited were able to improve their health relative to those who did not marry or cohabit. They found that only those mothers who married and stayed married to the biological father of their child born prior to marriage reported better midlife health than their peers who remained single, and the coefficients were significant in ordinary-least square (OLS) regressions (p<.01) as well as PSM regressions (p<.05). However, in the African-American subsample, the coefficients were only significant in the OLS model (p<.05).
Because the most sophisticated model (PSM), which teased out variables that normally go together, does not show statistically significant health improvements at midlife for black mothers that later married and remained married, the Williams researchers (all of who are women) unfortunately reveal their feminist hands. They claim their study provides no support for public policies that uphold marriage as the social (and health) ideal, particularly recent “healthy marriage” programs that serve populations that include African Americans. Such efforts involve “succumbing” targeted populations to “a tyranny of averages.” As these scholars assess the data, “the empirical evidence does not support axiomatic assumptions about the universal benefits of marriage.” At the same time, they see a need to “urge caution in generalizing these results to more recent cohorts of women for whom nonmarital childbearing is more common,” presuming without a shred of evidence that their clear findings of negative health outcomes of non-marital childbearing, even among African Americans, would wane if more American women followed that dubious pattern.
This downplaying of the marriage health premium is nothing more than disingenuous posturing. If the researchers were honest with the data, they would not allow a subordinate finding, based upon a smaller sample, that subsequent marriage does not in every situation reverse or temper the negative consequences of unwed childbearing to trump their central finding, based upon the full sample, that huge health disparities at midlife separate women who bear children with a wedding ring and husband from their peers who foolishly think they can go it alone.
(Kristi Williams et al., “Nonmarital Childbearing, Union History, and Women’s Health at Midlife,” American Sociological Review 76.3 [June 2011]: 465–86.)