Religiosity and illegal drug use

A typical liberal refrain is that religious folks are just as bad a irreligious people; they're just hypocrites. Even religious types will sometimes say that many of their own behave badly because the church attracts people with problems.

GSS respondents were asked if they had used illegal drugs in the past year. Here are the percentages who answered yes by church attendance:

Percent who used illegal drugs in the past year (sample size = 2,323)

Never attends 8.9
Less than once a year 6.1
Once a year 8.7
Several times a year 5.3
Once a month 1.8*
Two or three times a month 3.3*
Nearly every week 1.5*
Every week 1.0*
More than once a week 1.8*

*significantly less than the "never attends" group


People who never go to church are five times as likely to use illegal drugs as people who attend more than once a week.

Sexual orientation and sexual satisfaction

MIDUS study participants were asked to rate their sex lives from 0 to 10. Here are the means (sample size = 3,729):

Mean sex life rating:

Straight men 5.35
Homosexual men 5.09
Straight women 4.91*
Female bisexuals 4.44
Male bisexuals 4.12
Lesbians 3.50*

*significantly lower than straight males

Straight men are the most satisfied; lesbians are the least. The gap between the highest and lowest groups is sixth-tenths of a standard deviation--a large difference. Women tend to give lower ratings, perhaps because they get less out of sex than men.

Lesbians and the Big Five

Following the method I used in the last post, I estimated the relationship between sexual orientation and the Big Five personality traits for women (sample size = 2,094):

Standardized OLS regression coefficients

Extraversion .01
Negative emotionality .04*
Conscientiousness -.05*
Agreeableness -.03
Openness to experience .01

Compared to straight women, lesbians are more negatively emotional and less conscientious. Lower agreeableness falls just short of statistical significance.

Gay men and the Big 5

I looked at the relationship between sexual orientation and Big 5 personality traits. I included self-esteem as a control--results are not shown (sample size = 1,724):

Standardized OLS regression coefficients

Extraversion .02
Negative emotionality .01
Conscientiousness -.01
Agreeableness .05*
Openness to experience .13*

Compared to straight men, gays are more agreeable and open to experience. Lesbians are next...

Predictors of getting cheated on

The MIDUS Study asked respondents if their spouse had ever been unfaithful. I wanted to identify respondent characteristics that predict this.

Logistic regression coefficients

Men (sample size = 717)

Extraversion .12
Negative emotionality -.02
Conscientiousness -.15
Agreeableness .40*
Openness to experience .01
Age -.03*
Social class .00
Religiosity -.27*
BMI .00

*statistically significant

Of the Big 5 personality traits, only being agreeable predicts being cheated on for men. Older men are less likely to have this happen, as are religious men. Body mass index has no predictive power.

Women (sample size = 850)

Extraversion -.11
Negative emotionality .01
Conscientiousness -.44*
Agreeableness -.03
Openness to experience .43*
Age .00
Social class .00
Religiosity -.14*
BMI .01

*statistically significant

For women, conscientiousness lowers the risk of being cheated on, while being open raises one's chances. The only other predictor that matters is religion: religious women are less likely to get cheated on.
 
By the way, my experience with the data is the same as that of Rodney Stark as described in What Americans Really Believe. Compared to each of the Big Five traits, religiosity is a more potent variable. It frequently predicts outcome measures more strongly. Add this to the finding that the heritability of religiosity is .5, and perhaps Stark is right that religiosity should be considered a major personality trait.

The wussification of America II

This graph shows trends in American attitudes toward spanking (GSS data). From 1986 to 2010, the percent disapproving or strongly disapproving (green and yellow) almost doubled from 17 percent to 31 percent. At this rate, we'll soon be joining the other countries that have banned corporal punishment.

The wussification of America

This graph shows the percent of U.S. households that have at least one firearm (GSS data). From 1973 to 2010, the figure has dropped from 48 to 32 percent. Over the period, ownership has fallen by one-third. This is another indicator of the country's move in a liberal direction.

Atheists who believe God created nature?

GSS respondents were asked whether they believed that: 1) nature is sacred because it was created by God; 2) nature is sacred in itself; or 3) nature is important, but not sacred (sample size = 3,505). Of the people who "know that God exists", 54 percent gave answer 1; 18 percent gave answer 2; and 28 percent answered number 3. In other words, most believers think nature is sacred because it is God's creation.

But here's the weird part: 13 percent of the 92 surveyed atheists said that nature is sacred because it was created by God. (Twenty-seven percent answered 2, and 60 percent answered 3). "Atheist" is defined as "not believing in God." Unless the 12 atheists who gave 1 for an answer are retarded (or perhaps there was some other type of miscommunication) they mean something else by "doesn't believe in God." I can imagine people interpreting the statement to mean "I don't follow or agree with God."

According to the GSS, only 2.6  percent of Americans say they do not believe in God. It might be the case that fewer actually do not believe in the existence of God.  

It's interesting too that atheists are less likely to give nature an elevated status. Sixty percent of them say that nature is important but not sacred, compared to 28 percent of believers. Perhaps a more secular term like "cherished" would have elicited more affirmative responses; perhaps atheists are less likely to give anything a really elevated value--I don't know.  

Jews and Big 5 traits

I estimated OLS models for Big Five personality traits with self-esteem and whether or not you are Jewish as predictors (sample size = 3,915--94 Jews):

Standardized OLS regression coefficients

Extraversion
Jewish .03*
Self-esteem .40*

Negative emotionality
Jewish .04*
Self-esteem -.51*

Agreeableness
Jewish .02
Self-esteem .17*

Conscientiousness
Jewish .01
Self-esteem .37*

Openness to experience
Jewish .03*
Self-esteem .39*

Controlling for self-esteem, Jews are significantly more extraverted, negatively emotional, and open to experience. They are not more conscientious or disagreeable (as I thought they would be).

By the way, Jews do not differ from others in self-esteem.

Episcopalians, self-esteem, and Big Five personality traits

Using MIDUS data, I estimated OLS regression models with Big 5 personality traits as dependent variables and self-esteem and whether or not the respondent is an Episcopalian as predictors (sample = 3,915--93 Episcopalians):

Standardized OLS regression coefficients

Extraversion
Episcopalian .00
Self-esteem .41*

Negative emotionality
Episcopalian .00
Self-esteem -.51*

Conscientiousness
Episcopalian .00 
Self-esteem .37*

Agreeableness
Episcopalian -.02
Self-esteem .17*

Open to experience
Episcopalian .04*
Self-esteem .39*

*statistically significant

In spite of being an elite religious group, Episcopalians do not differ from others except that they are a little bit more open to experience.

The more interesting finding, perhaps, is that self-esteem is strongly related to all five traits. It is positively associated with desirable traits and inversely related to negative emotionality. People who have more esteem for themselves rate their traits more highly across the board. The correlation between agreeableness and self-esteem is weaker than the others perhaps because, while agreeableness is considered to be a good trait, it also suggests the person is a "yes-man."

Mormons, self-esteem, and conscientiousness

I looked into Dr. Charlton's suggestion in the comments of the last post that self-esteem might boost self-assessment of conscientiousness. Using MIDUS data, I estimated an OLS model with conscientiousness as the dependent variable and self-esteem as a predictor. I also added a Mormon-versus-others dummy variable as a predictor in order to see if greater conscientiousness might emerge for Mormons once the influence of self-esteem is controlled. 

OLS Standardized Regression Coefficients

Mormon -.03
Self-Esteem .37*

*statistically significant

Self-esteem is strongly related to higher conscientiousness scores. Personality researchers may do well to examine if self-assessments are distorted by the level of self-esteem. On the other hand, the result for Mormons remains unchanged: they do not differ from others in conscientiousness even when the influence of self-esteem is controlled.   

Mormons and personality

Personality differences across religious affiliation is interesting. Let's use MIDUS data to look at Mormons (sample size = 3,960, 76 Mormons).

Mean scores

Extraversion
Mormons 2.98
Others 3.10

Negative emotionality
Mormons 2.06
Others 2.09

Agreeableness
Mormons 3.42 
Others 3.45

Conscientiousness
Mormons 3.40
Others 3.49

Open to experience
Mormons 2.86
Others 2.90

There are no significant differences. It turns out that, in terms of personality, Mormons are boringly the same as everyone else. (I hypothesized that they would be more agreeable, more conscientious, and less open to experience. I was wrong on all counts.)

Agnostics and Big 5 personality traits

Readers SFG and HBD Chick suggested that I compare the personality traits of agnostics with non-agnostics (MIDUS data, sample size = 3,960).


Mean scores

Extraversion
Agnostics 2.82*
Others 3.10
Cohen's d .49

Negative emotionality
Agnostics 2.15
Others 2.09

Agreeableness
Agnostics 3.23*
Others 3.45
Cohen's d .44

Conscientiousness
Agnostics 3.36*
Others 3.49
Cohen's d .29

Openness to experience
Agnostics 3.17*
Others 2.90
Cohen's d .50

*Agnostics and others are significantly different

Compared to the general population, agnostics are more introverted, less agreeable, less conscientious, and more open. They are like atheists, only shy and neurotic.

Atheists and the Big 5 personality traits

Using MIDUS Study data, I calculated mean Big 5 personality scores for atheists and others (sample size = 3,960).

Mean scores
Extraversion
Atheists 2.99
Others 3.10

Negative emotionality
Atheists 1.81*
Others 2.10
Cohen's d .46

Agreeableness
Atheists 3.15*
Others 3.45
Cohen's d .60

Conscientiousness
Atheists 3.37
Others 3.49

Openness to experience
Atheists 3.36*
Others 2.90
Cohen's d .85

*atheists and others are significantly different

Cohen's d is a measure of the gap between means. Compared to others, atheists are significantly less neurotic, less agreeable, and more open to experience. The differences for the latter two are large. It is not surprising to observe that atheists are disagreeable and open to new things.

Interest in Mex-Ams

Some readers, I'm sure, think I have some odd interest in Mexican Americans. Well, it's not odd--it's simple math. New from Pew:

31.8 million people is roughly TEN million more people than all Americans who say they are of English ancestry. TEN million more!! (2006-2008 American Community Sample)

Peruvians are fascinating people, but who cares?

Men, women, and childlessness

In the comments of the last post, Dr. Bruce Charlton reminds us of the issue of growing female childless ness. Let's compare men and women on this. I list below the percent of people without any children. I limit the women to those ages 45 to 64, and ages 50 to 69 for men.

Percent childless

1970s
Men 17.3
Women 13.8

1980s
Men 13.4
Women 10.2

1990s
Men 14.3
Women 15.8

2000s
Men 19.0
Women 15.8

2010
Men 16.6
Women 18.8

The percent of women with no kids almost doubled from 10.2 percent in the 1980s to 18.8 percent in 2010. Having no kids has also grown a bit among men. The childless numbers for men and women are pretty close to each other which fails to support the claim that men are much more variable in their fruitfulness than women.