I’ve seen this idea before, but I wonder if we actually have any empirical evidence that it is true that short-term mating reduces the likelihood of pair bonding for women later in life.
It says they controlled for a variety of variables but the obvious question is whether families / cultures that discourage pre-marital sex also discourage divorce and whether this was controlled for. I don’t have time to read the paper now so if anyone knows the answer to that I’d be interested.
I read Teachman’s paper in the meantime; overall, it looks like a solid piece of work. The controls are definitely not broad enough to rule out the above hypothesis directly. This however should not be held against him, since this question is outside the scope of the study, and it would be a very difficult task to come up with controls that cover all such possible familial and cultural influences reliably.
Nevertheless, it seems to me that the basic finding of the paper is inconsistent with the above hypothesis. One would expect that conservative families and subcultures discourage premarital sex and (especially!) cohabitation even between future spouses to a significant extent. Therefore, if the effect of premarital sex on divorce risk is entirely due to such influences, we would expect to see a difference between women who practiced cohabitation or premarital sex only with their future husbands and those who didn’t practice it at all. Yet as Slumlord points out, the striking result is exactly that there is no such difference.
That said, this finding is inconsistent with a previous study that looked into the question of why exactly virginity at marriage predicts lower probabilities of divorce: J.R. Kahn & K.A. London, Premarital Sex and the Risk of Divorce. (Unfortunately, I haven’t found an ungated version.) Kahn & London’s conclusion was that this is because lack of premarital sex correlates with traditionalist attitudes, i.e. basically the above hypothesis. However, their conclusion is based on a complex statistical model that makes it look quite far-fetched to me. Frankly, I lack the statistical knowledge to judge it reliably and authoritatively, but in any case, Teachman’s work looks like a much stronger and more straightforward piece of evidence that points in the contrary direction.
[Update: after rummaging through the literature a bit more, I found a letter to the editor by one Tim Heaton, published in the same journal (vol. 55, p. 240), which harshly criticizes the soundness of Kahn & London’s statistics. This was followed by an unconvincing response by K&L, presenting the usual cop-out of the sort, yes, our methodology is lousy, but we couldn’t do any better. On the whole, my above conclusions are further reinforced by this finding.]
HughRistik:
I just ran into an interesting link that’s highly pertinent for this topic. Slumlord discusses a paper that provides for a very strong case that the answer is yes:
http://socialpathology.blogspot.com/2010/09/sexual-partner-divorce-risk.html
(H/t Thursday via OB.) I haven’t had the time to read the paper in detail, but on a casual look, it seems quite convincing.
It says they controlled for a variety of variables but the obvious question is whether families / cultures that discourage pre-marital sex also discourage divorce and whether this was controlled for. I don’t have time to read the paper now so if anyone knows the answer to that I’d be interested.
I read Teachman’s paper in the meantime; overall, it looks like a solid piece of work. The controls are definitely not broad enough to rule out the above hypothesis directly. This however should not be held against him, since this question is outside the scope of the study, and it would be a very difficult task to come up with controls that cover all such possible familial and cultural influences reliably.
Nevertheless, it seems to me that the basic finding of the paper is inconsistent with the above hypothesis. One would expect that conservative families and subcultures discourage premarital sex and (especially!) cohabitation even between future spouses to a significant extent. Therefore, if the effect of premarital sex on divorce risk is entirely due to such influences, we would expect to see a difference between women who practiced cohabitation or premarital sex only with their future husbands and those who didn’t practice it at all. Yet as Slumlord points out, the striking result is exactly that there is no such difference.
That said, this finding is inconsistent with a previous study that looked into the question of why exactly virginity at marriage predicts lower probabilities of divorce: J.R. Kahn & K.A. London, Premarital Sex and the Risk of Divorce. (Unfortunately, I haven’t found an ungated version.) Kahn & London’s conclusion was that this is because lack of premarital sex correlates with traditionalist attitudes, i.e. basically the above hypothesis. However, their conclusion is based on a complex statistical model that makes it look quite far-fetched to me. Frankly, I lack the statistical knowledge to judge it reliably and authoritatively, but in any case, Teachman’s work looks like a much stronger and more straightforward piece of evidence that points in the contrary direction.
[Update: after rummaging through the literature a bit more, I found a letter to the editor by one Tim Heaton, published in the same journal (vol. 55, p. 240), which harshly criticizes the soundness of Kahn & London’s statistics. This was followed by an unconvincing response by K&L, presenting the usual cop-out of the sort, yes, our methodology is lousy, but we couldn’t do any better. On the whole, my above conclusions are further reinforced by this finding.]