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
When Mae West noted that “a man in the home is worth two in the streets,” the actress was acknowledging the domesticating effect of wedlock on young men, who left to their own devices would ordinarily be content on chug-a-lugging, hitting on women in bars, and roaming the streets in search of trouble. Now comes a study by economists at the University of Toledo, Mir Ali and Olugbenga Ajilore, which puts numbers behind that common-sense observation, finding that marriage among young African Americans leads to significant reductions in anti-social, risky healthy behaviors like drinking and substance abuse.
The economists examine data on 2,518 African-American young people who were interviewed in all threes waves (1994, 1996, and 2002) of the National Longitudinal Study of Adolescent Health to determine the effects of marriage on smoking, drinking, and drug use. To prevent selection effects from influencing the results, the researchers used propensity-score matching to adjust for pre-treatment observable differences between the treatment group (those who married) and a control group (those who did not marry). Furthermore, to reduce biases that might be generated by unobservable differences, they included control variables to improve the quality of the match between the two groups so that the effects of matrimony would be isolated from such differences. Their statistical models also control for past participation in risky behaviors in addition to demographic, health, and parental characteristics.
Using this rigorous methodology, the researchers calculate the effects of marriage using several matching methods as well as through ordinary least squares (OLS), finding that African Americans who are married are significantly less likely—relative to if these same individuals had remained single—to engage in excessive alcohol use, getting drunk, and drug use. The coefficients in the OLS estimates of these three risk behaviors were all statistically significant: on average, married individuals were 10 percent less likely to get drunk (p<.001); were 4.4 percent less likely to engage in heavy drinking (p<.01); and 5.3 percent less likely to use drugs (p<.05). While not all the matching estimates of the effects of marriage on these three behaviors were statistically significant, they confirmed the general pattern. None of the estimates, whether matching or OLS, indicated a statistically significant effect of marriage on smoking-related behaviors, perhaps because smoking is not considered an abusive activity that can threaten a marriage relationship.
The economists also found similar results, using the same rigorous methodology, when quantifying the power of marriage to reduce risky behaviors relative to cohabitation, a confirmation that matrimony remains a unique institution, contrary to the imagination of many sociologists. As the researchers note, the fact that “norms of non-engagement in health compromising behaviors are less clear” and that “individuals in a cohabiting relationship behave more like singles” may explain this finding.
Encouraged by their robust findings about the protective effects of matrimony, Ali and Ajilore call for extending their analysis to measure the effects of marriage on other health-related behaviors of African Americans (such as diet, exercise, and risky sexual behaviors) as well as the role of marriage in reducing violent behavior among African-American men. Such studies would likely provide additional evidence indicating why incorporating marriage-promotion programs in social welfare policy, first broached in the 1996 welfare reform legislation, remains a good idea.
(Mir M. Ali and Olugbenga Ajilore, “Can Marriage Reduce Risky Behavior for African-Americans?” forthcoming in Journal of Family and Economic Issues.)