New research on testosterone and behavior

A retired researcher friend asked for my help. He has been linking male sex hormones to behavior for more than three decades. Since he is retired, he has no access to students to volunteer as research participants. I can't help him because there is no way the IRB committee at my university will approve this study.  He needs volunteers to complete this 5 minute survey. Please help if you can, and invite others. We need the broadest sample we can get. Warning: you will need to measure a body part!  https://www.surveymonkey.com/r/NKBRYYR

UPDATE: Of course, this study is very exploratory. He wants to see if it looks promising. 

Addiction among American Indians

I ran across a new study reviewing the research on race and addiction. The authors find that: 

In these studies, both past year and lifetime rates of alcohol, nicotine, and drug use disorders were consistently higher in Native Americans and lower in Asians,with Blacks, Hispanics, and Whites in between. 

It looks like any theory of American Indians (AIs) that limits itself to alcohol (e.g., the notion that AIs have not been exposed to it long enough to adapt to it) is missing part of the story. Addiction to drugs in general tends to be more common in this population.

Also--the stark differences between Northest Asians and Natives--addiction differences being only one illustrative trait--show how quickly significant genetic change can occur (the differences surely can't be 100% due to environment). It's not like these groups split 100k years ago. 

Liberal men, sexist pigs?

I analyzed General Social Survey data a few years and showed that liberal women are more likely than conservatives to report being victims of sexual harassment. My take at the time was that liberal ideology makes women define normal sexual advances as inappropriate and abusive. But now I'm wondering if liberal women simply work around many liberal men, and liberal men simply lack the old-fashioned chivalry found more often among conservatives.


Here's a replication:

Percent reporting sexual harassment over past 5 years

Extremely liberal  23.7
Liberal  18.9
Slightly liberal  19.5 
Moderate  16.7  
Slightly conservative  19.3
Conservative  15.9
Extremely conservative  10.3

99 genetic loci related to IQ in a sample of 280k

Recent studies of the genetics of IQ have been AMAZING. The latest--posted just today--is a sample of 280,000 subjects, and it found 99 loci (and 9,714 SNPs) significant related to IQ. Many of these loci also predict information processing speed, general cognitive function and health variables. The picture emerging is that a well-functioning, healthy brain is a smart brain, and genes are responsible for much of it.  Intelligent people, for example, are less likely to suffer from Alzheimer's.

The importance of self-discipline

The Alt-Right makes a big deal about IQ, and rightly so, but I just ran across this meta-analysis of dozens of studies that shows that the supertrait of conscientiousness--basically self-discipline--is just about as predictive of academic performance as intelligence.

Race and gang membership

I ran across this self-report study of youths in seven cities. According to my calculations, Hispanics are 3.3 times as likely to be gang members as whites. For blacks, it's 3.1 times more likely. And these estimates include a lot of wannabe gangsters, so the racial/ethnic gaps are likely to be larger among serious gang members. Not surprisingly, the researchers find that members of gangs commit a lot more crime than people not in gangs. Gangs are not just social clubs. 

Another look at ancestry and IQ

It's been a few years since I looked at verbal IQ data listed by ethnicity. It's important to look at recent data since so many immigrant groups have such small samples sizes, estimates are not very reliable. Using GSS data, here is the ranking of scores to a 10-point vocabulary quiz--a quick measure of verbal IQ. I omitted people not born in the US since the quiz is biased against non-native speakers:

Vocabulary Ranking by Ancestry

Austria   7.27
Russia    7.12 
Lithuania   7.08
Romania   6.95
Denmark   6.85 
England/Wales   6.84
Hungary    6.78
Scotland   6.75
Yugoslavia   6.73
Sweden   6.71
China   6.70
Norway   6.68 
Switzerland   6.65
Arab   6.61
Finland   6.58 
Greece   6.53
Czech   6.56
Ireland   6.46 
Poland   6.46
Japan   6.47
Italy   6.42
France   6.34
French Canadian   6.30
Germany   6.29
Belgium   6.28
West Indies   6.19
Portugal   6.18
Netherlands   5.99
Philippines   5.84
Spain   5.83
India   5.53
American Indian   5.31
Africa   5.16
Mexico   5.00
Puerto Rico 4.99

The pattern hasn't changed much since the last time I looked. What jumps out at me: 1) Mex-Ams and Puerto Ricans are still at the bottom, even lowers than blacks; 2) recent data indicates that American Muslims might assimilate better than European Muslims--the data here indicates that US Arabs are pretty smart; 3) Asian Indians (n = 40) still have a low average despite their reputation for intelligence; 4) it looks like southern Europeans don't do as well as northern Europeans (but Romania and Yugoslavia are high); and 5) central and Eastern Europeans might get a boost toward the top of the list by the large numbers of Jews (with a mean of 7.56) among their ranks.

And if you're thinking, this is just all environment, you're wrong: According to this meta-analysis, vocabulary is 62% genetic among children, and it is certainly higher than that among adults.

Oh, and if you think the differences are no biggy, again you're wrong: The gap between the top and the bottom is huge: well over 1 standard deviation. 

Does ignorance or familiarity breed contempt?

Does ignorance or familiarity breed racism against blacks?

To answer this question, I correlated how cool a white person feels towards blacks with the percent of the population in their region that is black (General Social Survey data). The correlation across nine census divisions is .48--a strongly positive correlation.

In other words, whites who live in areas with more blacks--an indication that one has had more actual experience with blacks--tend to feel cooler toward them.

The data suggests that familiarity breeds contempt, not affection. 

 

Conservatism does NOT predict hatred for blacks, but liberalism STRONGLY predicts hatred for conservatives

Using General Social Survey data, I calculated the correlation between conservatism and feeling cool toward blacks. It's practically non-existent--a trivial .07. Next, I correlated a correlation between liberalism and coolness toward conservatives. It's large--.45.

To put this in plain English, conservatism does NOT predict hatred for blacks, but liberalism STRONGLY predicts hatred for conservatives.

The polls suggest that most of the undecideds will go for Trump

Let's take another look at the Trump v. Hillary polls.

It looked to me like the polls that nudge people toward choosing one candidate or another--as opposed to giving some other answer like undecided, don't know, etc--tend to yield a smaller gap.

I calculated a correlation between the size of the gap and the percentage of people NOT saying Trump or Hillary for the 13 recent surveys listed on either RealClearPolitics or the Huffington Post. I threw them all in there, including the ones I considered to be not-so-great surveys.

The correlation is .38 which means of moderate magnitude. In other words, when techniques are used to get people to take a side, Hillary's lead decreases. When nudged, the uncommitted tend to go for Trump. To illustrate, 23% of Reuters participants fail to give either candidate as their answer, and this poll shows Trump behind by 7 points. With the UPI/CVOTER poll, which managed to get all but 4% to make a choice, Trump came out on top by one point.

So this seems like evidence that the undecideds are leaning Trump, and in the end, most will vote for him. 

These polls are closer to the truth: Gravis, PPP, Rasmussen, YouGov, UPI/CVOTER, LA Times/USC, and Zogby

The Huffington Post has a useful interactive website where you can compare poll results by several factors. I wanted to look a little closer to see if polls tend to be biased against Trump because some supporters are shy about admitting that they plan to vote for him, and that the methods fail to fully capture the enthusiasm for him.

What you see below is the average Hillary lead by poll characteristics:

Partisan/Registered Voters/Live Phone  11.0

Non-Partisan/Registered Voters/Live Phone  6.2
Non-Partisan/Likely Voters/Live Phone  7.3
Non-Partisan/Registered Voters/Non-Live Methods  5.9
Non-Partisan/Likely Voters/Non-Live Methods  4.8

You can see that partisan polls REALLY favor Hillary. Non-partisan polls that are otherwise the same produce a gap that is 4.8 points smaller (11.0 vs. 6.2 points).

Likely voter polls do not necessarily yield smaller gaps than registered polls, so there is no clear evidence here that registered voter polls are failing to capture the greater support for Trump. On the other hand, it is possible that there are more Hillary supporters who say they are likely to vote but won't actually do it because they can't get excited about her.


Finally, there is clear evidence that the gap shrinks when private methods are used so participants don't have to tell real humans their preferences. The gap shrinks .3 points for registered voter polls and 2.5 points for likely voter polls.

So, the most reliable polls use non-live methods. On the other hand, registered voters polls are not necessarily inferior at this point in the race. Still, it makes sense to give preference to likely voter polls since many registered voters will not actually vote. Based on this, the surveys to place the most trust in (at least the ones that come out frequently) are: Gravis, PPP, Rasmussen, YouGov, UPI/CVOTER, LA Times/USC, and Zogby.  Reuters should also be on the list, but my confidence in them has been shaken. The results are sometimes really weird, and of course, we've seen reporting that suggests they monkey with the data. The good polls indicate Trump is down 4 or 5 points.

The best polls indicate a small gap between Trump and Hillary

All the forces that hate Trump--the Dems, the Media, and the NeverTrumpers--are all chanting the phrase that Trump is toast, but the polls have seemed a little off to me for a long time. Audacious has addressed this kind of thing before. The Clinton lead seems to be larger in registered voter polls than likely voter polls, and the gap seems to be smaller with online polls. The first suggests more enthusiasm among registered voters for Trump, and the second suggests that some people might hide their preference for Trump from telephone interviewers.

I created a list of polling companies that were not off by more than an average of 1.5 points in the 2012 election that have also conducted polls in the last two weeks. I predict that the nationwide candidate gap will be largest for telephone surveys of registered voters, and smallest for online surveys of likely voters.

Three polls are live interviews of registered voters--IBD, CNN, and Marist. The mean is a lead of 10.3 points for Hillary.

One poll is an online poll of registered voters--YouGov. It gives Clinton a 7 point lead.

Reuters/Ipsos is the only online poll of likely voters that made my list.  Now Audacious and others have rightly criticized this survey, but I'm assuming they've gotten things straightened out over there. They give Hillary a 4 point lead.

Long story short, Hillary definitely seems to be ahead, but the best methodology produces a small gap between the two candidates, not a large one. And keep in mind that most media polls are of registered voters interviewed by telephone, so the gap is repeatedly exaggerated in the news.

UPDATE:  I checked Reuters latest poll, and it looks like their sample is 44% Democrats and 34% Republicans. That sounds biased to Republicans, but these kinds of numbers are used by polling companies that have been accurate in the past. Evidently, a big chunk of the country has screwy political views, and it's an uphill battle for us to win on the national level.  

Part of the reason Hispanics are so liberal is because they are young

I showed recently that some of the reason that Asians tend to vote Democrat is that they are, on average, younger than whites. Let's now see if this is true of Hispanics. Using GSS data, here are the percentage who voted for Romney in 2012:

Percent Voting for Romney

Hispanics
Ages 18-44  15.5
Ages 45+  24.1

Whites
Ages 18-44  47.8
Ages 45+  49.1

You can see than young Hispanics are very liberal. Older Hispanics are liberal, too, just a little less so.  The mean age of whites in 2012 was 49.4. It was only 40.9 for Hispanics, so they are a much younger group. Conclusion: Part of the reason why Hispanics are so much more liberal than whites is because they are younger, but even if their average age was that of whites, they would still be a liberal group.

Elites vote Democrat, not Republican

I'm tired of the old Democrat myth that the Republican Party is the party of elites. Looking at General Social Survey data, it's not surprising to see that only 24% of high school dropouts voted for Romney in 2012, but how many people with advanced degrees voted for him? A whopping 32%. Most highly successful people vote Democrat because their competitors aren't the poor. Their enemies are America's Middle. They seek an alliance with the poor so they have the numbers to subjugate ordinary Americans.

Support for Trump skyrocketed among religious people post-Orlando

The graph is taken from Reuters Polling. It shows trends among religious people (i.e., those who attend church nearly every week, every week, or more than once a week).

You can see the support for Trump (red line) has bounced around since the beginning of April, but it hasn't been impressive since you would expect religious people to support the Republican nominee. As I wrote before, highly religious people are not enthusiastic about Trump, and I suspect it's because they want their candidate to be straightlaced. (You don't have to be religious to not like Trump -- just stuffy and upright -- atheist George Will doesn't like him either.)

But notice the 13 point jump in support since Orlando. That's a big, fast increase. My guess is that moments like Orlando force religious people to face the fact that there are only two choices: either the next Prez will be Trump or Hillary. And who wants Hillary when ISIS wants to kill you? Trump support among the religious has fallen the last couple days (not shown) so it might take several clarifying moments to get these righteous folks to pull the level for the Donald.

Gay men, but not lesbians, are more likely to favor gun control

With the Orlando shooting, I wondered about gun attitudes among gay people. The General Social Survey asked people if they favored a law requiring people to get a police permit before buying a gun. Here are the percentages in favor:

Men
Gay  82.9
Bisexual  64.3
Straight  65.6

Women
Gay  66.7
Bisexual 60.0
Straight 77.6

Among men, gays are more likely to favor a gun permit law. In fact, the vast of majority of them want it. In contrast, bisexual men are like straight men.

The pattern is different for lesbians: their numbers are similar to those of straight men. Bisexual women shows the LEAST support for gun permits of any group.

The overall pattern is consistent with the view that gay men tend to be psychologically like women, while lesbians are like men. Some of this could be due to the level of exposure to prenatal testosterone. Bisexuals in some ways seem hyper-masculine; turned on by anything that moves.

Asians Americans vote Democrat because they're young (and liberal)

An interesting claim was made by Tantum Malorum on Twitter (where I spend all my time these days).  He said that Asians tend to vote Democrat because they are disproportionately young.  I looked into this with GSS data. The mean age of all voters in 2012 was 52.2. Here are the means for Asian groups listed in GSS:

Mean age
Chinese  51.1
Filipino  42.6
Japanese  45.5 
Indian  48.6
Other Asian  45.0

Mean age for all groups was lower than for the total sample, although it's not much lower for Chinese Americans. Since young people tend to vote Democrat, it could be that Asians are not more inclined to be liberal, they're just younger.

Next, let's look at voting in 2012 for: 1) young Asians, and 2) older Asians. I'm going to add all Asian groups together since sample sizes are so low.

Percent voting for Romney

Ages 18-44  
All voters  32%
Asian voters  25%

Ages 45 and up
All voters  40%
Asian voters  38%

First of all, it looks like the numbers are lower than they should be. GSS oversamples women, so this might be a problem.

Anyway, 25% of young Asians voted for Romney, compared to 32% of all young people. For older Asians, there is only a two point gap with all older voters.

So, it looks like the Asian vote is tilted Democrat because they are a younger population, but they are a little more liberal as well.

The numbers are too small to make much of it, but of the Asian groups, Indians were noticeably more likely to vote for Obama.

Data shows that college liberalizes students

I was curious if college liberalizes students. Common sense tells me it does, but I have seen research that throws doubt on the idea.

Listed below are the mean conservatism scores by year of college for those attending college since 2000 (General Social Survey data).

Mean conservatism

Freshmen  4.04
Sophomores  3.71

Juniors  3.62
Seniors  3.69

Students tend to get more liberal as they move through college. The shift from the freshman to junior year is almost half a standard deviation, which means it is a fairly big change.  So it does look like academics are somewhat successful at their goal of turning our youth into progressives. They seem to have the most luck in the first three years--the seniors are no more liberal than juniors.

Which American groups are the most race-loyal?

In a recent post, I found that race predicts political orientation more strongly than does income for many minority groups. It seems to me that the race/politics link, adjusted for the influence of income, is an indication of the importance of race to that particular group.  If members always look at issues through the lens of their racial group, they will tend to vote in the same way and will vote Democrat, regardless of how wealthy they are.

Based on this line of reasoning, here is a ranking of the most race-loyal, race-conscious (if they were white, the term would be race-ist) minority groups.

Race-consciousness ranking
1. Blacks
2. Mexicans
3. Jews
3. Asian Indians
5. Puerto Ricans
6. American Indians
7. Chinese
7. West Indians
9. Japanese
10. Arabs

Not surprisingly, blacks are the most race-conscious group, but it's interesting that West Indians are much less so. Since Jews are white, you might not expect them them to be ethnocentric at all, but they outrank many nonwhite groups.

Asian Indians are quite race-conscious, as are Mexicans and Puerto Ricans (to their own group, not to the fake category of "Hispanic"). Despite being non-white, Chinese and Japanese Americans seem to be less uptight about race. Arabs are even less ethnocentric. (Keep in mind that many of these Arabs are Christians.)

Hispanic IQ in the United States

Average IQ is a good measure of how well a group will do in the United States. A low IQ accurately predicts more problems in a community. Let's see where various Hispanic groups are at. To set white Americans at a mean IQ of 100, I had to set the mean IQ for all Americans at 96.5. This indicates that US IQ has dropped to a level lower than many European and East Asian countries. The results shown below are limited to Hispanics born in the US, since the measure of IQ is a vocabulary test in English, and foreigners are at a disadvantage.

Mean IQ

White Spanish 100.3
White Cuban 99.2
White Mexican 91.5
Mexican--other race 90.5
White Puerto Rican 90.3
Black Puerto Rican 88.4
Puerto Rican--other race 84.0

The only groups with IQs similar to white Americans as a whole are people who describe themselves as White Spanish or White Cuban. We would expect these folks to assimilate well into mainstream America. Marco Rubio and Ted Cruz, for example, are white Cubans.

For the remaining groups, the average IQs are much lower, ranging from 84 for non-white Puerto Ricans  to 91.5 for self-described White Mexicans. Many Hispanics consider themselves to be white which ignores the fact that most have some American Indian ancestry. As you can see, the IQ difference between white and non-white Hispanics is small. These groups are much more similar to each other than they are to non-Hispanic whites. At least in terms of IQ, it makes sense to lump white and non-white Hispanics into one group, but white Cuban Americans and those who call themselves Spanish should be lumped in with whites.

Based on average IQ, we expect these Hispanic groups with low numbers to fare poorly in the United States, and since IQ is a stable, highly heritable trait, the poor performance is likely to continue indefinitely.