Inductivist

In response to my claim that married people are, on average, wealthier than people in other marital statuses, a commenter on Twitter suggested that this kind of finding ignores the loss of wealth due to divorce, so it is not worth it to get married. The implication is that people who never marry should be wealthier than divorced people. Of course, we would need to adjust for age since wealth tends to grow as one gets older. 

I did this using General Social Survey (GSS) data. Respondents were asked their wealth on a 15 category scale that went from less than $5,000 to more than $15 million. This first model is for men (n = 1,689):

Divorced people are the reference category: all other marital statuses are compared with them.  We can see that never-married men do not more have more wealth than divorced men. The sign indicates that the divorced have more wealth, but the relationship is not statistically significant. The only group that is significantly wealthier than divorced men is the married group. 

And for women (n = 1,769)?

Never-married women do not differ significantly from divorced women in wealth. Widowed and separated women are poorer than divorced women. Married women are the only group that are significantly wealthier. 

Bottom line: marriage is a good place to be. 

A new Italian study led by a female researcher and published by the flagship journal Intelligence reports that men, on average, have higher levels of perceptual reasoning and working memory.

Perceptual reasoning refers to mental operations used to analyze novel problems, rules, and logical relationships, and to create and test solutions. By working memory, researchers mean storing, focusing attention on, and manipulating information for a relatively short period of time.

The researchers are concerned with sex differences in subtest scores that could be due to test bias or social ways that the sexes might differ, and not due to differences in general ability (that are presumably due to biological factors).

They reasoned that if subtest scores are due to broad abilities, the size of the sex difference for the subtest should correspond to the factor loading on the broad ability. For example, if the sex difference on the arithmetic is large, the sex difference on working memory (i.e., a higher-order ability) should also be large. If the latter is small but the former is larger, the authors contend that there is bias in the arithmetic test.

While authors find evidence of some bias in the WAIS-R test, they conclude that the WAIS-IV is "quite fair." If we focus on the results from the WAIS-IV data, we find that men have a small advantage in perceptual reasoning and a moderate advantage in working memory (3 IQ points).

By contrast, there are no significant sex differences in verbal reasoning or processing speed. (Some studies have found a female advantage in processing speed). When indices are combined to created a full-scale IQ score, men have a small advantage--perhaps 1 or 2 IQ points. And, according to the authors, these differences are due to broader abilities, not various forms of bias.

UPDATE: An interesting sidenote: When analyzing WAIS-IV, the researchers look at the sex differences for 15 subtests. You might expect that if there is unintentional test bias, men would come out ahead in some tests, while women might in others. But in zero of the 15 tests do women have a statistically significant advantage.

Clearly, anyone who believes that astrology is scientific doesn't understand what science is, but how common is this belief, and what's the statistical profile of a believer?

The General Social Survey asked respondents the question with answers varying from "very scientific" to "not scientific" at all (sample size = 5,548). 36.6% of people said astrology was at least sort of scientific. That's a lot of dummies.

Here are the factors that predict belief (standardized OLS coefficients):

Predictors of thinking that astrology is scientific

Female   .09
Black   .13
Other race    .04
Age   -.07
IQ   -.17
Educational level   -.11
Church attendance   -.04
Political conservatism   -.04

All of the coefficients are statistically significant at the p < .01 level (or higher, two-tailed test).  So the profile looks like this: female, black (vs. white), other race (vs. white), young, unintelligent, uneducated, non-churchgoer, and liberal. The coefficients indicate the strength of the prediction: IQ is the best predictor of thinking astrology is not scientific.

Why have almost all of the greatest movie directors been men?  Don't women like movies just as much as men?

Feminists, of course, would blame the patriarchy: Men have mysteriously gotten control of the world and will not let the helpless women do fun things like make movies.

But if discrimination has been so pervasive in the film industry, why in the world have we seen so many top homosexual directors?  I'm not convinced that men before the 1970s were that dead set against women occupying important positions. Just the other night, I watched an old film titled, "Kansas City Confidential" (1952), and the lead female was studying to take the bar, and none of the male characters cared in the least.

Now, imagine the same story, but the romantic interest is a gay man preparing for the bar.  Do we see movies like that for most of the 20th century?  Hell no. Off the top of my head, I know that Clark Gable did not want George Cukor--known to be a homosexual--to direct "Gone With the Wind" and was influential in having him replaced by Victor Fleming. While there were plenty of homosexuals in Hollywood, people loathed it.  No matter--gay men thrived in Hollywood and Europe as well. Perhaps you doubt this.

I went to the website "They Shoot Movies, Don't They? and looked at their list of the top 250 directors of all time. The ranking is based on such factors as voting by directors and critics. I categorized a director as gay or bisexual if Wikipedia indicated they were. I put together the following list:

Gay Directors (from top 250)

Pedro Almodóvar
Lindsay Anderson
Kenneth Anger
Marcel Carné
Jean Cocteau
George Cukor
Terence Davies
Jacques Demy
Rainer Werner Fassbinder (bisexual)
Robert Hamer
Todd Haynes
Vincente Minnelli (bisexual)
F. W. Murnau
Pier Paolo Pasolini
John Schlesinger
Gus Van Sant
Luchino Visconti
Lana Wachowski (male-to-female transgender)
Lilly Wachowski (male-to-female transgender)
James Whale

I included the transgender Wachowski brothers since transgenders should face discrimination, if it is indeed such a profound problem.

That's 20 gay, bisexual, or transgender directors or 8% of the total. Keep in mind that sexual minorities are less than 8% of men, so they are over-represented among the greatest directors.

How about women?

Female Directors (from top 250)

Kathryn Bigelow
Jane Campion
Claire Denis
Danièle Huillet (co-director with her husband Jean-Marie Straub)
Leni Riefenstahl
Agnes Varda

That's 6 or 2.4% of the total, and let's not forget that if women we're punching at their weight, they would be half of the best directors. Their numbers are abysmal.

You might counter that discrimination was intense through 1970, but things have changed and that is why we see that all top women are from the past few decades. Again, I would argue that if bias was intense prior to 1970 for women, it should be even more so for gay men--a hated minority if I've ever seen one--yet easily half of them worked prior to 1970.

The facts suggest that men are simply better at making movies. If a studio exec wanted to make the best "Little Women" possible in 1933, he needed to hire George Cukor and not give a damn about his "quirks." If a woman would have done it better, I submit that the studio execs would have swallowed hard and given her the job, even in 1933. A lot of money was on the line.

What qualities do men possess that give them such an advantage? Well, I'm no expert on directing, but I know that these are incredibly talented people at the highest percentiles of all relevant traits. These would include: intelligence, leadership, charisma, confidence, decisiveness, technical mastery, visual skills, writing skills (plot, character, dialogue, mood, humor), effective criticism, and ability to deal calmly through all the drama that comes with managing creative types. There is evidence that at the highest levels, men surpass women on these traits. And, by the way, the traits are all rooted in biology.

While I am convinced that genes are a strong influence over people's lives and that the impact of the parenting is wildly exaggerated, I am open to data on these questions--I call myself Inductivist after all.

This meta-analysis of 15 heritability studies conducted in a variety of countries and decades found that shared environment explained a sizable portion of the variation in educational attainment; to be specific, almost 40%. That's a much higher average than typically seen in heritability studies.  The authors also found that shared environment was stronger for women and for people studied prior to 1950, suggesting that factors like family financial support have mattered more for women and for people in the past.

We can use General Social Survey data to answer a related question: Is educational attainment due to IQ or dad's socioeconomic status (SES)?  First, let's see how strongly each predicts years of education completed (I limited the analysis to data from 2010-2018, sample size = 947):

Standardized OLS regression coefficients

Model with Father's SES only
Father's SES  .37***

Model with Child's IQ only
Child's IQ   .45***

IQ is the stronger of the two predictors, but nurturists might argue that father's SES causes child's IQ which, in turn, determines educational attainment. We can address this question by entering both into the model as predictors. By doing so, we can see if the link between IQ and education shrinks to nothing once we've accounted for the influence of dad's SES.

Model with Father's SES and Child's IQ
Father's SES   .24***
Child's IQ  .44***

When both predictors are entered into the same equation, the father's SES/child education correlation is reduced, but the impact of IQ on schooling is basically unaffected. We can interpret these findings this way: How far you go in school is influenced by your dad's social class (consistent with the meta-analysis), but your own IQ is much more important. The strong correlation between IQ and schooling is not at all due to the tendency of high-status men to both have smart kids and to help them continue in school.

By contrast, part of the reason why father's SES is linked to child's educational level is because high status men have smart kids, and smart people naturally go further educationally. Once you take into account the pathway from dad's status through offspring IQ to completed education, the link between dad's class and child's educational attainment is weakened substantially. In other words, factors beyond the kid's IQ, like family financial support, are not as strong important as they look.

I looked at females only and got the standardized coefficients of .23 (dad's SES) and .43 (child's IQ), so the process works the same for girls as well as boys.

I also looked at mom's SES, and I found very similar results.

***p < .001, two-tailed test

UPDATE: The strong correlation between IQ and years of education reminds me of Taleb's anti-IQ argument: IQ-type tests get you into college, so there is a built-in correlation. There might be a link between test score and which college you get into, but there is no such circularity with how many years of school you complete. Regardless of your test score, you can get admitted to some college.

A reader at Reddit pointed me to this table from GenForward, an online survey of young Americans (ages 18-34). It's associated with the University of Chicago. You can see the question they asked about identity and the results by race:
















The results are consistent with my analysis GSS data of adults of all ages, but the identity politics is seen more sharply here among young Americans.

Race is most important for nonwhites. Race (probably some of it ethnicity) is 3rd most important for whites. Religion is not important in any group except for Hispanics, perhaps. It's 5th for whites. Class and gender are important. Sexuality ranks highest among blacks which supports the stereotype of blacks being more obsessed with sex than the rest of us (which is saying a lot).

The one thing that could unite us--American nationality--is one of the least important identities. It reaches its peak--4th--among whites. The future looks like identity politics.

The one type of identity that will probably be considered illegitimate by elites for the foreseeable future is whiteness. Every other interest group will push, but when whites qua whites push, they will likely be crushed by the powers that be. I could be wrong, but I don't see a time when elites will ever see whites as simply another normal interest group.

UPDATE: One problem with the question is that it seems to be designed to emphasize the kinds of identity that are closely linked with politics. While the sample is of young adults, some people will be married and have kids by their early thirties, but 'marriage' or 'parenthood' (or something like 'family' which can be important at any age) are not included. These identities are important to many people, and I suspect the popularity of gender might be linked in people's minds to family roles.

Also--since the question asks about identities that "have the most impact on your life." some liberals whites might be want to choose race since they feel their white privilege is so consequential. I'm thinking of a new Inductivist slogan: "The definition of white privilege (or Jewish or Asian privilege) is the accident of being part of a social network that has more people who have their shit together."

Everyone who does social research knows that correlations about people are typically weak. You suspect that IQ predicts criminality, but then the data tell you the association is a mere .2. This is the rule rather than the exception. People are complicated. You can't reduce their behavior to a single factor. Plus, measurements are far from precise.

But on rare occasion, connections can be strong. The correlation between your education and your spouse's is .6 or .7.  The link between number of delinquent friends and one's own involvement in delinquency--about .6. Pretty good.

The largest individual-level correlate I've ever calculated (macrolevel correlations tend to be bigger) deals with sexual attraction: Using data I collected myself on 330 people, the point biserial correlation between being male and level of attraction to females is .82. For women being attracted to men, it's .84. Those are huge numbers.

Another way of describing it is in terms of standard deviations: the gap between male and female attraction toward females is 3.6 standard deviations. The difference between the two sexes on liking males is just as big--3.6 sds. You've probably heard that the black-white IQ gap is big. It is, at ONE standard deviation. The attraction gap is enormous.

Now you might be saying, duh, we would expect men to like women and women to like men.

Well, you might expect that, but then again you don't have a PhD in Gender Studies. Many years after the idiot Kinsey claimed that sexual attraction is a continuum, not categorical, researchers are still making similar claims.

With a continuum, one assumes that males will range from being strongly attracted to females to being strongly attracted to males, but most men will be somewhere in the middle with various levels of attraction for both sexes. The same for women. Most of them will have a mix of attractions. If this were the case, being male would not be a strong predictor of level of attraction to females. But common sense and my data predict the outcome very well: what do you know, men like women! And women like men!

I'm afraid a great deal of social research makes us dumber, not smarter. I'd love to see a great deal of it scrapped.

UPDATE: The latest sex theory I've come across claims that our ancestors were indiscriminate copulators: They would hump anything that moves. So males and females being attracted to each other was not the original system. With its de-privileging heterosexual sex, I predict the theory will be wildly popular.

In the last post, it was mentioned that high IQ people are VERY diverse in terms of income. Many are not particularly interested in pursuing lots of money. But let's look at the other end: Few people would want to be poor, so does IQ keep one out of poverty?

Using General Social Survey data, I looked to see how many people in the highest IQ category (125+) are found in the lowest 10% of income earners (sample size = 16,626). For men, it's 3.4%. So smart guys have a low rate of poverty, but not all escape it. I imagine these men have serious physical or mental health issues.

For smart women, it's 8.0%. Their rate is lower than average, but not by that much. In addition to the health issues that men might face, some intelligent woman are likely to be stay-at-home moms who don't earn much.

Stefan Molyneux retweeted today the claim that young single women now make more than young single men.  Is that true?

Using General Social Survey (GSS) data, I calculated median incomes for never-married men and women ages 18 to 29 without children who are working full-time.  The numbers shown below are the sex difference between medians, and I include all decades since the 1970s for comparison (sample size = 2,539).  The figures are in constant dollars.

Difference between median male and female incomes 
1970s    1,366
1980s    5,484
1990s       538
2000s    2,625
2010s    6,309

Young men have made more than young women in every decade for five decades, and the biggest gap has been this decade.

The gap is not due to some patriarchal conspiracy (Remind me guys, when and where is the next meeting?) to keep women down. Young men take jobs that make higher wages like construction, while young women are more likely to do something like childcare which doesn't pay crap.

More women are in college, too, so that will make them earn less in the short-term.

A recent, amazing study looked at how strongly the morphology of the brain (as measured by structural brain MRI scans) predicts IQ and the sex of the research subject.  The graph below shows the results from two different data sets.  "Morphometricity" on the y-axis is defined as the proportion of the variation of the trait that is explained by variation in brain morphology.














The morphology of the brain explains NINETY-FIVE percent of the variation in IQ.  In other words, intelligence is very strongly predicted by the "architecture" of the brain.  And gender?  Morphology explains NINETY-THREE percent of the variation in sex, or whether the subject is male or female.

In a table not show here, brain morphology only explains 55% of ADHD, 50% of schizophrenia, 38% of autism, and 20% of Parkinson's. 

Now, if gender and IQ are merely social constructions and are not biologically meaning, why are they so closely related to brain morphology, even more closely related than several brain diseases?  People who push the biology-doesn't-really-matter-for-IQ-and-gender view are idiots and liars. 

Anyone who has thought about it for more than a minute knows that policing is a tough job.  You have to deal with screwed-up people and situations all the time, and there is bureaucratic BS up the ying-yang.  Like a high-paid ER doctor, you have to make life-and-death decisions in a split second, and let's not forget that many people hate you for your efforts, and even want to kill you. Heck, you might go to prison if authorities don't like how you behave on the job.

Good law enforcement requires real ability, but what is the average intelligence of an American police officer?

Using General Social Survey data, I identified 90 law enforcement officers. Their average IQ is 100.2.  If we look at blacks and whites separately, the means are 99.4 and 100.3, respectively.  Female officers averaged 99.0, while the average man was 100.5.

I work with many aspiring police officers, and they seem to be motivated by the excitement and the desire to be the hero rather than the security of a government job and the early retirement. Money does not seem to be a priority (although it never hurts).

To inspire a talented young man to choose policing as a career, society needs to portray officers as something like superheroes, not as the executioners of innocent black men. Ironically, demonizing law enforcement might push away the talented and attract the leftovers who are more likely to use excessive force against minorities.

On Twitter, Steve Sailer writes, "The pattern that gay black males tend to assimilate into white corporate culture better than do straight black males deserves more social science attention than it has received.  It's hard to notice if you are Woke and believe in the conventional wisdom about Patriarchy, etc."

Let's test this kind of idea with General Social Survey (GSS) data.  Based on intersectionality theory--experiencing more discrimination with each additional minority status--job prestige rankings should look something like this:

Top Tier--Zero Strikes Against
Straight white male

Second Tier--One Strike Against
Straight white female
Gay white male
Straight black male

Third Tier--Two Strikes Against
White lesbian
Straight black female
Gay black male

Fourth Tier--Three Strikes Against
Black lesbian

By contrast, biological theory would focus on traits that are important for high-status jobs, and the obvious candidate is intelligence and the glaring racial gap in IQ. So this view would predict that the big divide will be race, with whites above average in job prestige and blacks below average. The other demographics shouldn't matter much. 

And what are the facts?  According to GSS data for 19,901 respondents: 

Mean occupational prestige 

Gay white male  47.1
White lesbian  46.2
Straight white male  45.2
Straight white female  44.5

Total Sample  44.2

Black straight female  40.9
Black gay male  40.7
Black lesbian  39.8
Black straight male  39.2

As predicted, the major divide is race.  All whites are above average, and all blacks are below.  If intersectional theory were true, black lesbians would be on the bottom, and straight white males would occupy the top position.  Neither is the case.  Biological theory would predict no disadvantage for homosexuals--in fact, I reported in a previous analysis that they have above-average IQs--and indeed they do well.

And relevant to Steve's point about black males, homosexuals do better than straights, which contradicts the claim that it's all about intersectional discrimination.

Biology explains more than sociology.

I've always assumed that smart people are fascinated by politics. One complaint I have of social scientists is that they're so obsessed with politics, they can't conduct fair research. (But I can hear you saying, "You said SMART people.")  Anyway, is my assumption correct? Is it different for men and women?

The General Social Survey asked respondents: "How interested would you say you personally are in politics?" Answers ranged from 'very interested' to 'not at all interested.'  I measure IQ with a 10 question vocabulary quiz (sample size = 1,141):
















Look at how those scoring a perfect score on the test are much more interested in politics than any other group. To get precise, the correlation between the two variables is .20 -- a moderate relationship.

Now for the women (sample size = 1,507):



I separated by sex because I suspected that the rise in interest as one moves up the IQ scale might be sharper for men. My experience is that smart women are more interested in politics than less intelligent women, but I see more intensity among men.

The female pattern is similar to that for men. The correlation is slightly weaker -- .18. The gender gap is pretty visible at the highest IQ level: 16.8% of these women have little or no interest, compared to only 5.4% of the men.

In the last post, we looked at ethnic differences in whether the environment or genes are thought to be more important for a variety of traits. Now let's see which factors predict choosing nurture over nature.

Using GSS data, I estimated linear regression models (OLS) with each of the four questions as dependent variables, plus a scale of all four of them summed.  I included all demographic predictors I could think of, including: sex, age, race, southern region, immigrant vs. native-born,  educational level, income, church attendance, number of children, and political orientation. I list below the significant effects (beta weights are shown):

Obesity
Black  -.15
Education  .10

So blacks, compared to whites, and less educated people think environment is less important. Race is the more powerful predictor.

Alcohol Abuse
Female  -.08
Education  .07

Females and the less educated think genes are more important for alcohol abuse.

Altruism
Female  -.06
Black  -.06

Women and blacks are shifted toward seeing genes as important for altruism.

Athleticism
Education  .06
South  -.07

For athletic ability, Southerners and the less educated tend to see genes as being more important.

Nurture over nature scale
Black  -.11
Education  .11

When the scores for all four questions are added together to make a scale, blacks and the less education are shifted toward genes having the most impact.

Not surprisingly, people exposed to more education tend to believe in the power of the environment. After years of getting the same message from liberal teachers, what do you expect?  It is a surprise, though, that blacks, after adjusting for education, give higher estimates to the power of genes.

UPDATE: It might surprise you that political orientation (liberal vs. conservative) is unrelated to one's view of the importance of genes. 

UPDATE II: I wonder if the race difference comes from the fact that blacks are more fatalistic than whites, and people tend to assume (wrongly in my view) that genes imply determinism but environment does not. Whites might embrace nurturism because it sounds compatible with the idea that we can take control of our lives and improve things.  

It's always exciting for nerds like me when new GSS data arrives. 2018 is now available. They asked respondents whether they voted for Clinton or Trump. I used logistic regression to determine what predicts voting for Trump (sample size = 904):

Voted for Trump 
Male  .57
Southern region  .33
Black  -3.78
Other race  -1.30
Education  -.11
Income  .01
Church attendance  .16

The coefficients are not standardized, so they can only tell you the direction of the relationships, not the magnitudes. I included age and IQ, but neither one was significantly related to vote choice.  Keep in mind that the relationships are net of the influences of the other factors included in the model, so, for example, age might predict voting for Trump because older people are whiter, wealthier, less educated and more religious, not because they grew up in an earlier era.

So here are the characteristics of people voting for Trump: male, Southern, white, less educated, higher income, with more church attendance. No surprises there. And what mattered the most in order from most to least predictive: 1) white, 2) church attendance, 3) male, 4) education, 5) income, and 6) southern residence.  Trump owes the election to religious, white males, like myself.  You're welcome.

I didn't have data on the impact of Russian colluders.

We saw in the last post that religious women are having more babies than women who never go to church. What about men?  Here a graph for number of children for men ages 45-60 (GSS, N = 1,251):
















The pattern for men is even clearer than for women: guys who attend more than once a week have a much higher mean (2.38) than those who never go (1.61). That's roughly half of a standard deviation difference.

Is this religiosity-fertility link explained at all by differences in attitudes?  Do religious people want more kids?  Look at the graph for the same age ranges (GSS, N = 2,173):
















Males are green, and females are purple (I'm a dude, so I won't try to give more precise colors).

There is a small tendency for more religious people to idealize larger families. It seems too small to explain much of the reason why the religious have bigger families. It's probably due to getting married earlier and less willingness to get an abortion if you get a surprise.

NOTE: You have read an updated version.  The earlier version had the error of leaving in cases of those who answered "whatever number people want," cases that were scored an 8.  The results you see above omit these cases (about 6% of the sample). 

This new study of 3.9 million American children over 27 years finds that girls are better than boys and reading and writing, and that this gap emerges early.

The authors say that, following masculine norms, male peers might discourage reading and writing. Reading and writing is sissy stuff, but every he-man-womanhaters club knows that math is macho. How long will we have to endure such crap? Males and females are biologically different. What's the big deal? Why all the silly lies?

Feminists instruct us that men are organized to oppress women everywhere.  Women have incredible talents and ambitions, but these are crushed by male rule.

If this were true, we should see high levels of female dissatisfaction. Gifted individuals who are blocked from success surely cannot be happy about it.

A new meta-analysis of hundred of studies and more than one million people from many countries -- some of them with low levels of gender equality -- reports that there are no significant sex differences in being satisfied with life or one's job.

How can this be?  The truth is that women have it pretty good.  Surveys are not going to detect the handful of women who fantasize about Evil Males and push the lie on others, usually in a classroom somewhere.

This meta-analysis synthesized the results of 126 studies that used various measures to compare the brains of males and females. The results are amazing. Males not only have larger brains than females, on average: the difference is huge! Depending on how the brain is measured, the difference is between 1 and more than 3 standard deviations. The average guy has a brain that is more than 50 milliliters larger than the average female.

Male and female brain volumes are different in almost 30 areas of the brain. And it's not just size: male grey matter is denser in 7 areas.

The most important region for the difference is the limbic system, which is an emotional center. The big male advantage is on the left side of the limbic. Females are larger in the language areas of the right hemisphere.

I was surprised to read that sex differences were not located in areas of the brain in charge of spatial tasks since men have an advantage here, nor did the researchers find a difference in the corpus callosum, the region that connects the two sides of the brain. Women are said to have better integrated hemispheres, but this region does not appear to be larger in females.

The authors did not go into sex differences in behavior that might stem from brain differences, although they did mention the language difference. Women might get their advantage over men in verbal fluency from larger language areas in the brain.

The large limbic system in men suggests males and females will differ emotionally. Perhaps the larger, more dense limbic system enables greater integration with the frontal lobes which are in charge of rational thought. Men might be better at disengaging the feeling-thinking connection, giving them an advantage at a detached, systematizing style of thinking. Women are much better than men at empathizing--a more personal, less abstract style of thinking. These differences produce many men who are good at, say, engineering, and many women who have excellent interpersonal skills.

The brain differences emerge in young children, although there aren't enough studies to show this definitively. So how boys and girls are brought up probably has little to do with the gender gap. Vive la difference!