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
If the study above, based upon the National Survey of Family Growth, does not deliver enough punches, a meta-analysis published the same month by psychologists at the State University of New York (Stony Brook) carries an even greater blow to the notion that shacking up before marriage is a good idea. Pulling together twenty-six studies published between 1980 and 2006, this extensive review confirms that the greater incidence of cohabitation over time has not tempered its consistent negative impact on marital success. More importantly, the analysis offers rigorous evidence that premarital cohabitation—even when it might lead to marriage—inversely correlates with marital stability and marital quality.
The researchers first measured the impact of premarital cohabitation on subsequent marital stability, using data from sixteen studies and collected from multiple countries between 1981 and 1999. To compare the treatment group of premarital cohabitants to the control group of non-cohabitants, the study calculated “effect sizes” using odd ratios (OR). Here, the overall OR effect size for all sixteen studies was not only large (.81) but statistically significant (p<.01). This “relatively robust” effect means, according to the authors, that premarital cohabitants had significantly lower odds of staying married than did non-cohabitants. Effect sizes of a subset of the total sample (representing just six studies where cohabitation was limited to the eventual spouse) were also tabulated. Here, the OR effect size remained large (.87) but it did not achieve statistical significance, meaning that while cohabitants that married each other, relative to married couples that did not cohabit, had lower odds of staying married—but not significantly lower odds.
To assess martial quality as an outcome, the researchers used twelve studies that analyzed data collected in North America between 1974 and 2000. They calculated effect sizes using the “standard mean difference,” or the amount of nonoverlap between the two groups. Here, the overall effect size was small (–.19) but reached statistical significance (p<.01), meaning that premarital cohabitation is modestly negatively associated with marital quality. Even when limited to studies in which participants had married their cohabiting lover, the effect size remained small (–.19) and statistically significant (p<.05).
But rather than encouraging couples to marry and forgo cohabitation, the researchers think a single finding—that cohabiting with one’s future spouse tempers some of its negative effects—warrants an admonition to couples “to consider their own commitment to the relationship as well as their own expectations about what cohabitation will (or not) lead to.” With this gobbledygook coming from authorities that should know better, no wonder young people find growing up hard to do.