Home / General / Non-Correlation Isn’t Causation

Non-Correlation Isn’t Causation


Let’s play spot the egregious howler with Ross Douthat:

Yet for an argument that has persuaded so few, the conservative view has actually had decent predictive power. As the cause of gay marriage has pressed forward, the social link between marriage and childbearing has indeed weakened faster than before. As the public’s shift on the issue has accelerated, so has marriage’s overall decline.

Since Frum warned that gay marriage could advance only at traditional wedlock’s expense, the marriage rate has been falling faster, the out-of-wedlock birthrate has been rising faster, and the substitution of cohabitation for marriage has markedly increased. Underlying these trends is a steady shift in values: Americans are less likely to see children as important to marriage and less likely to see marriage as important to childbearing (the generation gap on gay marriage shows up on unwed parenting as well) than even in the very recent past.

Correlations do not, of course, establish causation.

See the problem with the argument that “marriage is changing now that the United States has same-sex marriage”? I assume you do, but I’ll turn things over to Pareene anyway:

Douthat’s next paragraph begins: “Correlations do not, of course, establish causation,” but in this case there isn’t even correlation. Gay marriage is still banned in most of the United States. Exclusivity to heterosexual couples did nothing to prevent the severing of sex and procreation from “weakening” the institution of marriage. I also fail to begin to comprehend how growing public acceptance of gay marriage convinced more women to have children out of wedlock, or how national attention to a class of people who wish to get married convinced more straights to cohabitate outside of marriage.

If the tiny handful of states where same-sex marriage has been legal for any significant length of time were particularly acute examples of this trend, then Douthat might have something, but of course the opposite is true. In Massachusetts, which has had SSM the longest, marriage rates have held steady while they’ve decreased substantially nationwide, while divorce rates have also head steady at a level well below the national average.

Even if the Douthat/Frum circa 1997 argument was true it wouldn’t be a good reason to oppose same-sex marriage. But there’s no reason whatsoever to believe that it is.

  • Facebook
  • Twitter
  • Google+
  • Linkedin
  • Pinterest
It is main inner container footer text