Friday, March 23, 2012

Sexual History Divorce Risk II

A scientific paper that may be of interest to men who frequent this corner of the blogosphere is  Divorce and Sexual Risk Among U.S. Women: Findings from the National Survey of Family Growth. Published in the Journal of Women's Health, in November 2011, the statistical findings were based on the 2002 cycle of National Survey of Family Growth (NSFG) involving 5800 women.

The authors of the study set out to fill a gap in the the literature with regard to the sexual behaviour of divorces and single adult women for the purposes of disease prevention. i.e the authors did not have some religious or social policy axe to grind. The paper is very interesting because it seems to be the first scientific paper to acknowledge a clear correlation between the risk of divorce and sexual risk.
Our study has several limitations. The data are cross-sectional and thus only correlative. We cannot conclude that divorce causes increased sexual risk any more than we can suggest that sexual risk causes divorce. The relationship exists, and further studies should focus on causal relationships between the two.
I think that the authors are being modest.

The study first cites other research which gives the rate of marital dissolution in the U.S to be 20% at 5 years,  33% at 10 years and 43% at 15 years. This "baseline" divorce rate is important in understanding subsequent findings. Secondly, they also cite research which shows that the probability of a divorce woman remarrying is 54%, this too, is important. Also important that the mean age of marriage for women is 25.3 years.

Now the authors divided women firstly into two groups. Those who had five or more partners and those who had four or less. (n=4 being the average number of partners a U.S. woman had at the time of NSFG study)


Then they divided the women according to their marital status. Separating women into four groups. Those who were never married, those who were co-habitating,  those who were currently divorced, and those who were currently married (including those who remarried).


What I've done here is plot the "natural divorce rate" against the statistical findings of the "5 or more lifetime partner group".

The first thing we see that if a woman has five or more life time partners, she is more more likely to be never married,  cohabiting or  divorced  than married. Now remember, the married group includes women who have remarried. So the actual number of women who have never been divorced is actually smaller than what is indicated on the graph.

Secondly, the NSFG obtained data from women who were between 25-44 years of age The authors of the paper did not list the average length of marriage in the group, but since the mean age of marriage is 25 years of age, it would be safe to expect (using the worst case figures)that 43% of marriages should have dissolved at the based upon the 'background" divorce rate.  Now, if promiscuity is not related in any way to divorce there should be no significant variation from the natural divorce rate, and yet, what we see  in the "five or more group" only 37% of women married (and this includes the remarriages) instead of 57%. (Based upon a rough "back of envelope" calculation, I estimate that roughly 20% of the "five or more" group have never been divorced) Something quite real is going on here.

It gets worse.

When you look at all the married women (i.e including those with four or less partners) women with five or more partners were 70% less likely to be married.


The study authors noted that the two groups of women did not appear to differ in any other meaningful way apart from partner count. Regardless of what anyone's position on social issues is there is a real phenomenon going on here. Number count does seem to matter. That's not to say that a woman who has had five or more lovers is unable to form a stable marriage, approximately 20-30% do, it's just how do you pick? Is it worth the risk? Do you want to put your kids through it? Now presumably, in a significant portion of marriages it was the men who instituted the divorce, but that does not explain  the huge variation in divorce rates based on partner count.

Finally, the authors were able to extract another interesting piece of data which I will call the "carousel index".


It's a quantitative measure in the decline of sexual market value. The near exponential decline in attractiveness is quantifiable.

14 comments:

GK Chesterton said...

Again brilliant stuff. Since this is your professional strength would love for you to dive deeper if you can.

dicipres said...

Great work man. This is important to any man who wants to get married.

TDOM said...

I'm with you right up until the end. There may be other factors than attractiveness involved in why older women have fewer partners. They may be busy putting working in order to put children through college, or just taking care of children may scare the men away, they may find men their age less attractive, or perhaps their sex drive declines, menopause, or their marriage may have put them off men. But being less attractive does sound good as an explanation.

TDOM

The Social Pathologist said...

@GKC

Thanks. I hope to delve into it a bit later.

@dicipres

It's pretty grim.

@TDOM

Fair comment. I'll put up some more data later in the day that backs up my view a bit.

Vladimir said...

It looks like you've found some very interesting stuff, but I can't find an ungated copy of the paper, and I don't understand the graph. What exactly is the meaning of the colored bars?

The Social Pathologist said...

@Vlad,

Agree it's not a great graphic but let me try and help.

1) The purple coloured curve is a graphical representation of the number of surviving (non-divorced) marriages by length of time.

If for instance, we were to take a random group of just-married U.S. couples we would find that after 5 years 80% of them will still be married and divorced. The curve basically graphically represents the expected attrition rate of marriages in the U.S.

The coloured bars represent data from the papaer demonstrating the couple status of women who had five or more lifetime partners at the time of the survey.

My point with this graph was to demonstrate that the rate of marriage in women with 5 or more lifetime partners is far less than what we would expect according to the expected marital attrition rate.

The study didn't list the mean length of marriage in the five or more group, but if promiscuity was an independent variable, we should expect to see the number of marriages in that group to be somewhere on that curve. The fact that it is so far below it means that promiscuity is exerting some form of influence.

Does that explain things?

2) As far as I'm aware there are no unlocked versions of the study.

Vladimir said...

@SP,

The coloured bars represent data from the papaer demonstrating the couple status of women who had five or more lifetime partners at the time of the survey.

What's confusing me is that I can't make the numbers add up in any reasonable way. The "never married" (black) bar shows a percentage of around 57%, and the "divorced/separated" (red) bar is around 67%. How can both numbers apply to the same group of women?

Since I wasn't able to understand the meaning of these numbers, I was also left puzzled by the discussion that follows.

The Social Pathologist said...

What's confusing me is that I can't make the numbers add up in any reasonable way. The "never married" (black) bar shows a percentage of around 57%, and the "divorced/separated" (red) bar is around 67%. How can both numbers apply to the same group of women?


I see what the problem is.

The way the authors did the study was as follows.

Firstly, they determined the marital status of the subjects under study. i.e women between the ages of 25-44 were divided into their respective categories; married, never married, divorced, cohabiting.

Then each one of these groups was divided into either five-or-more or four-or-less partners.

Therefore the study shows 67% of divorced women had five or more partners which implies that of the divorced women, 33% of them had four or less partners.

The same goes for the other groups. 57% of the never married had five or more whiles 43% of the never married had four or less.

Does that clear things up?

Vladimir said...

I see. That makes sense, but it looks like the problem is with the following paragraphs, which are written as if the bars show the percentage of women who are married etc. among the "five or more" group, rather than vice versa.

For example, you say "what we see in the "five or more group" only 37% of women married," as if the light blue bar says that 37% of the 5+ group are married, not that 37% of the married group are in the 5+ category. You also have a few typos that make it hard to parse some of the sentences (e.g. the part with "...dissolved at the...").

I'm pointing this out because I think it would be great if you could improve the text to make the argument clearer. It would provide another excellent reference for discussions of these topics (along with the old post where you discuss Teachman's paper).

The Social Pathologist said...

Is it clearer now?

Vladimir said...

@SP,

It's better, but I think the article could be further improved, and I'd be interested in helping if I can get hold of a copy of the article. I think your series of posts is the best online reference to which people can be directed in discussions of this vastly important topic.

Do you have an email address where you accept correspondence from readers? (Maybe I'm just being clumsy but I can't find one on the blog web page.)

The Social Pathologist said...

@Vladimir

I can be contacted at fluvox@gmail.com

Anonymous said...

Many gaming portals, forums, magazines and publications give away free Xbox live codes for their readers.
Credit cards are readily available in both Sixteen hundred or 1000 stage sects.

Another way to earn free Xbox codes is by competing in
the competitions that are held up by Microsoft.


my blog :: msp codes

Anonymous said...

Wow that was unusual. I just wrote an very long comment but
after I clicked submit my comment didn't show up. Grrrr... well I'm not
writing all that over again. Anyway, just wanted to say fantastic blog!


Also visit my web blog - hostgator coupon code