Sunday, April 10, 2016

Churchgoers cheat on their spouses less than non-churchgoers do

++Addition++See Jayman's admonitions about the reliability of self-reported data, especially of the sexual variety as well as my response in the comments.

---

Heartiste reports that monotheistic religions may encourage paternal certainty. Patriarchal societies, more generally, probably require a high level of paternal certainty. Most societies throughout history have been patriarchal, so it's no surprise that cuckoldry rates among Europeans over time appear to have hung around in the 1%-2% range. In matriarchal societies, paternal certainty rates are presumably lower.

Monogamy, of course, guarantees paternal certainty. In encouraging--and to a large extent, normalizing--monogamy, monotheistic religions also encourage paternal investment. Paternal investment, in turn, is a prerequisite for advanced civilization (see sub-Saharan Africa, Haiti, or Detroit).

There is the question of causation to consider, too. Did monogamy and paternal certainty shape societal evolution or did evolution shape social norms vis-a-vis monogamy and paternal certainty? It's probably a combination of the two through the process of gene-culture coevolution, but I'm not qualified to comment on that with any certainty of my own.

Heartiste:
CH is on record (check the archives) predicting that the wholesale abandonment of religion by Western societies will lead to higher rates of cheating and cuckoldry, by both men and women.
The data to test that assertion exist. The percentage of married white men who have cheated on their wives by frequency of religious service attendance (for contemporary relevance all responses are from 2000 onward and attendance categories are exclusive, combined n = 9,205):

AttendCheat
Never25.5%
Yearly at most22.0%
Monthly at most19.1%
More than monthly16.0%
At least weekly15.5%

The same for married white women who have cheated:

AttendCheat
Never17.9%
Yearly at most18.6%
Monthly at most12.3%
More than monthly9.2%
At least weekly8.6%

Those actively involved in monotheistic religions are less likely to philander than those who are not.

Couple that with the decline in both church attendance and religious affiliation and it seems that the conditions for increasing cuckoldry are indeed present. Heartiste nailed it.

GSS variables used: EVSTRAY(1-2), SEX, YEAR(2000-2014), RACECEN1(1), ATTEND(0)(1-2)(3-4)(5-6)(7-8)

10 comments:

JayMan said...

This is self-reported cheating behavior, which is next to useless or perhaps even less than useless (a problem shared with sex research in general).

And indeed, this particular claim I don't believe for a second.

Dan said...

JayMan do you go to church? Do you have many churchgoers in your social circle? Any?

Audacious Epigone said...

Jayman,

I've heard the same thing about self-reported ethnicity, but simply looking at Census results make it clear that self-reported ethnicity isn't even close to random (they cluster predictably). No, it's not perfect, but it's not useless. The alternative is to put our hands up and say "who knows?"

It's possible that the there is no more veracity in the post's title than there is in an altered title reading "Churchgoers report cheating less on their spouses than non-churchgoers do". In other words, they could be operating entirely in "shame" mode, but as you know outbred northwestern Europeans (which is what we're mostly looking at here) are the most guilt-minded people in the world. I assume that's the explanation you suspect, because something has to be presumably driving a clear reported correlation between church attendance and fidelity.

JayMan said...

"I've heard the same thing about self-reported ethnicity, but simply looking at Census results make it clear that self-reported ethnicity isn't even close to random (they cluster predictably)."

No one made that claim. We do agree there is a difference between correlation r = 0, correlation 0 < r < 1, and correlation r = 1.0, yes?

"No, it's not perfect, but it's not useless"

Usefulness depends on what you're using it for. In the case of ethnicity, to assess broad patterns? Sure, noisy data is OK. To do fine analysis based on ethnic self-identification? Generally not as OK.

"The alternative is to put our hands up and say 'who knows?'"

Sometimes that is indeed what you have to do.

"It's possible that the there is no more veracity in the post's title than there is in an altered title reading 'Churchgoers report cheating less on their spouses than non-churchgoers do'. In other words, they could be operating entirely in 'shame' mode, but as you know outbred northwestern Europeans (which is what we're mostly looking at here) are the most guilt-minded people in the world. I assume that's the explanation you suspect, because something has to be presumably driving a clear reported correlation between church attendance and fidelity."

Churchgoers less likely to admit to cheating isn't enough?

There is some indication that religious people have higher sex drives that the non-religious (aforementioned caveats notwithstanding). That and other things is enough to make this questionable. Besides, as with sex research in general, how do you know, really? How do you know you can trust self-report? At least in the case of other self-reported features, like say personality, we have some other form of verification like observer- and peer-report that corroborate (partially, anyway) self-report, as well as correlation with outcomes. We have no such corroboration with sexual behavior (or very little, anyway). Virtually all sex research is built on self-report so the whole thing could be suspect.

In the end, I'm just advising caution, at least until we have more reliable data.


Santoculto said...

But the cuckhold via fathers can be higher because in a typical conservative society specially western, men will be relatively likely to be dishonest with their women and sleep with prostitutes as well other married women and non-official prostitute women. Bastard sons.

A speculation.

Audacious Epigone said...

Jayman,

Greg Cochran comes close here. Not random, not precise, and probably not "less than useless", either. Suggestive, but not exacting--which is all I ever assert when tapping GSS data since it is almost entirely self-reported (there are some interviewer assessments of participants included) information.

The admonition for caution is well received. To be tautological, I guess I should say "GSS data show that group X self-reports to be more likely to engage in self-reported behavior Y than group Z self-reports etc etc", but that strikes me as excessively verbose and I sort of assume it to be a given. Someone stumbling onto the post via a google search, though, might not aware.

I've linked to your comments in the body of the post, thanks.

Santoculto,

Is prostitution and extra-marital activity actually higher in patriarchal societies, though, or does it just stand out more because it goes against the grain of broader social mores?

Gilberto Carlos said...

Couldn't it be that church selects for people who cheat less by shaming and thus forcing out those who have a propensity to cheat?

Dan said...

Church attendance itself is a selector for conscientiousness.

The better someone's attendance at something they are committed to, the more conscientiousness they probably are.

Santoculto said...

''Santoculto,

Is prostitution and extra-marital activity actually higher in patriarchal societies, though, or does it just stand out more because it goes against the grain of broader social mores?''

Religion conformity forcing a lot of men who are prone to be dishonest in relationships to become apparently monogamic.

Speculation of course.

Audacious Epigone said...

Gilberto,

Yes, there's probably a boiling off effect in play.

Dan,

I wonder if there's something akin to the Flynn effect with conscientiousness (or the negative Flynn effect).

Santoculto,

Shaming works. Not perfectly, of course, but it works.