More on the demographics of mental health

In the previous post, we looked at the likelihood to have received treatment for a mental health problem by political orientation based on GSS data. It turns out that self-described political liberals are twice as likely to have been treated for mental health issues as conservatives are, with moderates falling in between*. This spurred several intriguing comments that I'll turn to the GSS again to address.

Firstly, the inevitable racial issue. An anonymous commenter writes:

This generalization is too broad, since I'm sure Caucasian Liberals will have superb mental health compared to Caucasian Conservatives. Since there's studies that shows that those in higher Academia has the least stress and hence the best health.

It turns out he was being tongue-in-cheek, and whites are more prone to experience mental health issues than non-whites are. The percentages who have been treated for mental health problems, by race (n = 1,412):

Race	Treated
White	15.1%
Black	9.3%
Hispanic	6.8%
Asian	9.4%

As blacks are only slightly more likely to identify as liberal than they are as conservative, despite consistently voting Democratic by overwhelming margins, it's conceivable that race obfuscates the association between liberalism and mental health issues. In actuality, however, the relatively good mental health of non-whites attenuates political differences that are even more stark when only whites are considered (n = 1,033):

Whites only	Treated
Liberal	23.7%
Moderate	13.3%
Conservative	10.6%

While only 1 in 10 white conservatives have been treated for mental health problems, nearly one-quarter of white liberals have. I am not qualified to explain why this is the case, but perhaps God has something to say about it. In clearing up the confusion about his first comment, the anonymous commenter later writes:

Conservatives should be mentally healthier, because they tend to be more religious, hence making it easier for regulation of mental health.

Liberals worry so much that they actually care about people they don't even know.

Whether or not liberals have more mental issues to deal with because they want to put the weight of the world on their own shoulders, God's most ardent followers tend to come from the conservative ranks, and one of the rewards they receive in return is better mental health. To avoid racial confounding, only whites are included (n = 1,064):

Worship frequency	Treated
Weekly or more	11.0%
More than once a month	13.6%
At least once a year	16.4%
Less than once a year	17.5%

Each classification is exclusive, so the second should actually read "More than once a month but less than weekly or more", etc.

What about Jews? TAE is, loosely defined, part of the alternative right blogosphere, so the question has to be asked! Jews, after all, are as a group more leftist than white Democrats are. Again, God's hand is shown! Not just any god, but our God, an awesome God who reigns from heaven above (n = 1,405):

Religion	Treated
Protestant	11.7%
Catholic	10.7%
Jewish	17.4%
Other	21.0%
None	19.8%

Unfortunately, the GSS is understandably unable to track ethnic Jewishness among survey participants. Instead, we're looking at Judaism by way of self-described religious affiliation. As Jews are considerably more likely than other Americans to be irreligious, there is presumably a sizable contingent of ethnic Jews in the "none" category.

Another trait that requires attention is intelligence. Writes Ed Tom Kowalsky:

[There is a] need to control for IQ inasmuch as I suspect mental illness tracks with intelligence.

Disappointingly, the best GSS proxy for IQ, Wordsum score, is not cross-referenced with the mental health question so controlling for it is a difficult thing to do. Educational attainment is of course correlated with intelligence, but the relationship is far from perfect. Instead of moving out several degrees by using education to estimate wordsum performance to estimate IQ scores, let's just consider differences in rates of mental health issues by educational attainment, again among whites alone to avoid problems with racial confounding (n = 1,067):

Educational attainment	Treated
Less than high school	12.8%
High school graduate	11.9%
Some college	14.7%
Bachelor's degree	20.9%
Post-secondary education	19.7%

As was the case with worship attendance, each classification is exclusive, so the second should actually read "High school graduate but no college", etc.

While conservatives are, on average, slightly more intelligent than liberals are, white liberals have an edge over white conservatives, so the association between greater intelligence and more mental health problems might go a little way in explaining why liberals have poorer mental health than conservatives do (though for what my uninformed opinion is worth, I suspect differences in personality traits other than intelligence are far more determinative).

Silly girl contemplates the relationship between criminality, political orientation, and mental health:

I wonder whether folks who have ever been arrested or incarcerated are more likely to be liberal.

I would say that being anti social enough to end up in prison is a pretty good indicator of mental issues.

As Jokah points out, small sample size is a huge (heh) issue here, as the mental health question was only posed in one year of the survey, and the number of people who spend time in prison is in the middle single-digits range. This GSS well is dry.

For what it's worth, in Freedonomics (p182-184), John Lott--who it should be noted is clearly unsympathetic to the Democratic party--reviews multiple studies showing that to the extent that they express political preferences, felons tend to vote even more heavily Democratic than their demographic statistics (which are very Democratic) would predict.

Since we're running the conventional demographic gamut, let's look at marriage and children. To give respondents a chance to have tied the knot if their plans include as much, those under the age of 30 are excluded, while again only whites are considered (n = 926):

Marital status	Treated
Married	12.8%
Unmarried	18.5%

And finally, kids. We're looking at the number white adults aged 30 and older have had (n = 927):

Offspring	Treated
0	15.2%
1	20.2%
2+	13.2%

So, mental health issues are most commonly experienced among well-educated, unmarried, irreligious white liberals who don't procreate (or do the SWPL thing and have one kid after both parents are firmly established in their professional careers). Mental health problems are progressive, baby!

GSS variables used: MHTRTSLF, RACECEN1(1)(2)(4-10)(15-16), MARITAL(1)(2-5), EDUC(0-11)(12)(13-15)(16-17)(18-20), CHILDS(0)(1)(2-8), POLVIEWS(1-3)(4)(5-7), ATTEND(0-1)(2-4)(5-6)(7-8), RELIGION

* As previously noted, the GSS item being utilized here asked respondents if they had ever received treatment for a mental health problem. That's not exactly the same as asking whether or not they had ever suffered from mental health issues, but it's pretty close.

White, black, and Hispanic birth rates by state

In the comments of a previous post, I wondered to what extent intra-state birth rates correlated across racial groups. That is, in states where white fertility is high, are black and Hispanic fertility rates also high?

I am unable to find birth rates broken down by race at the state level, but the Kaiser Family Foundation (which maintains a fantastic, data-rich website) does provide live births by race at the state level with data as recent as 2008 and accompanying data from 2009 on population by race at the state level. From that, I was able to compute birth rates per 1,000 people for whites, blacks, and Hispanics in each state*. For the handful of states with insufficient data on black population sizes, the US Census quick facts page is used as a supplemental.

Turns out the correlations are pretty weak: White and black birth rates correlate at a statistically insignificant .14, white and Hispanic birth rates at .27, and black and Hispanic birth rates at a statistically insignificant (.03).

White birth rates (map here) are highest in the Rocky Mountain states and the Midwest, middling in the South, and lowest on the coasts. The one seeming outlier is Hawaii, which shows the highest white fertility rate of any state in the country. Hawaii's population is not young (although I'm not able to find age broken down by race, and a majority of Hawaii's population is Asian or Pacific Islander), so I'm not sure if this is the result of a data transcription error, or if the relatively large percentage of Hawaii's whites who are in the military has something to do with it. Hawaii's overall birth rate is tied for ninth highest in the nation, and its black and Hispanic birth rates are also correspondingly high, so it might be accurate. Far less surprisingly, Mormon Utah comes in at #2.

Black birth rates (map here) are highest in Upper Midwestern states where blacks receive large welfare distributions relative to whites compared to black/white welfare recipient rates in other states. I'm bemused by Maine's high rate, only being able to point out that numbering fewer than 14,000, the black population there is extremely small. The South, where the largest share of the country's black population is concentrated, in contrast, has relatively lower and more middling rates, while birth rates for blacks are lowest in Rocky Mountain states and the Southwest.

Hispanic birth rates (map here) are probably the least reliable and also the least geographically consistent due to states with small Hispanic populations not accurately accounting for the sizes of their respective illegal residents, which may partially explain why rates show up as being astronomical in South Carolina. In contrast, states with large Hispanic populations--most notably the border states--have relatively low fertility rates. It should be noted, though, that Hispanic birth rates are generally considerably higher than black and especially white birth rates across the board. Nationwide, the Hispanic birth rate is nearly double the white birth rate.

* The rates are a bit lower than I'd expect based on national figures across the board (about 80% of states come in with a birth rate among whites that is lower than the national average), for reasons I'm not exactly sure of. Take the maps in more for comparative purposes than for exacting fertility rate measurements by race and state.

Percentage of ancestral Mexicans racially self-identifying as white by state

Razib Khan recently posted a table showing racial self-identification among those of Mexican ancestry in the US as well as the racial self-identification of those of Mexican ancestry in a selection of states using data from the 2006-2008 American Community Survey. A couple of commenters, including Steve Sailer, noticed an apparent relationship between Republican-leaning states and the percentage of ancestral Mexicans in those states who consider themselves to be white. With Idaho at the top of the list and New York at the bottom, that seems reasonable.

Presumably, the relationship would work via the more conservative, traditional ethos and culture of red states encouraging ancestral Mexicans in those states to identify more strongly with the majority rather than distinct from it, as they would tend to in places like California and Illinois.

To flesh out the accuracy of the observation, I looked at all 50 states (to see an accompanying map, click here*). There looks to be something to it, anyway. The correlation between the percentage of ancestral Mexicans who identify as white and McCain's share of the vote in 2008 is a positive but modest .25 (p = .08).

One issue in attempting to discover how voting patterns correlate with various other behaviors or attributes is the fact that blacks are so politically monolithic. The common perception of Mississippi is that it is a patriotic place, but over one-third of its population is black, so despite having the most conservative white population in the country, there are other more moderate states like Alaska and Utah that show up on electoral maps in a deeper shade of red than Mississippi does. The relationship between the percentage of ancestral Mexicans who identify as white and McCain's share of the non-Hispanic white vote, however, is even weaker, at .18 (p = .21).

What about ancestral Mexicans themselves? Unfortunately, state-level data on Hispanic voting patterns are only available for 13 states where the Hispanic population is substantial enough to report on. Despite that limited sample size, though, the relationship is strongest here--the correlation between the percentage of ancestral Mexicans who identify as white and McCain's share of the Hispanic vote is .51 (p = .07). This appears to mesh well with the narrative presented above--the more Hispanics identify with traditional American values, the more likely they are to identify with the majority white population.

By way of the GSS, consider how well this seems to hold up on the individual level. The following table shows, among those of Mexican ancestry living in the US, political orientation by racial self-identification, with blacks excluded and extending back only as far as the year 2000 for contemporary relevance (n = 713):

Race	Conservative	Moderate	Liberal
White	29.4%	43.9%	26.7%
Other	30.9%	40.0%	29.2%

White ancestral Mexicans are only marginally less liberal than ancestral Mexicans who identify themselves as members of another non-black race are. To the extent that the effect exists, it's pretty weak.

Parenthetically, recall that among the broader American public, people are nearly twice as likely to call themselves conservatives as they are to call themselves liberals. The preceding table shows that among ancestral Mexicans, the conservative and liberal numbers are at parity. Excluding a Cuban population that is decreasing in relevance, Hispanics, irrespective of racial self-conception, are considerably more politically liberal than whites are.

ACS variables used: RACE(100), STATEICP, ANCESTR1(2101, 2102, 2110, 2111, 2130, 2183)

GSS variables used: ETHNIC(17), RACE(1, 3), POLVIEWS(1-3)(4)(5-7)

* North Carolina really stands out as having a low percentage of ancestral Mexicans who racially identify as white. Is there an obvious explanation for why this is the case that I'm unaware of?

Peace is so SWPL

The Institute for Economics and Peace compiled a peace index ranking the 50 states (see the map here) earlier this year. The index uses five indicators in making the peace score calculations--homicide rates, violent crime rates, incarceration rates, per capita police officers, and the availability of small arms.

At first blush, the last indicator seems to reveal an unsatisfying leftist outlook in the report as less densely populated, politically conservative flyover states tend to have less restrictive legal gun ownership requirements than Northeastern and West Coast states do. That's probably a more-or-less accurate read into the authors of the report. However, gun ownership rates are estimated rather bizarrely by the percentage of suicides in each state committed by the use of a firearm, so while this is surely influenced in some degree by the availability of guns, it's difficult to dispute that a relatively high percentage of people ending their lives by blowing their brains out instead of overdosing on sleeping pills proxies for the violent tendencies of the residents in question reasonably well. Anyway, this factor was assigned the least weight of the five, representing only 1/15th of the peace index total.

Using incarceration rates and the number of police officers per capita is a bit problematic as well. Throwing increasing numbers of people who commit crimes in prison and holding them there is arguably one of the primary reasons criminal activity has decreased in the US over the last couple of decades. Similarly, relatively large police forces could be seen as a necessary response to a violent population, but could also be viewed as an expression--not necessarily related to the actual prevalence of violence--of how high a priority that population makes deterring crime and capturing criminals.

A few years ago, I computed a strictness index that ranked states by how many people they incarcerated relative to the amount of crime that occurred. South Dakota comes out as the most 'draconian'. This is notable because the report contains a "case study" on New York and South Dakota, the former being the state where the peace index has increased the most over the last 20 years and the latter being the state where it has decreased the most over the same period of time. Singling out South Dakota as a state that has slid backwards since 1991 highlights why incarceration rates and police presence are suboptimal measures of peacefulness--homicide and other violent crime rates are almost exactly where were 20 years ago, but because the incarceration rate and the per capita police presence have both increased markedly over that period of time, South Dakota looks bad. As a law-abiding citizen, if more thugs are in jail and more patrol cars are out than they were when I was a kid, but the frequency of crime has been held in check, I'm not complaining too much. These changes are certainly preferable to even a modest uptick in the crime rate.

These qualifications notwithstanding, state-level correlations with indices like these are rarely uninteresting. So without further ado, here are some of them. Positive correlations indicate the variable in question is associated with more peace, negative correlations with less of it:

Mean IQ -- +.62
McCain's share of the '08 vote -- (.31)
McCain's share of the '08 white vote -- (.58)
Median age -- +.28*
White population % -- +.48
Black population % -- (.56)
Hispanic population % -- (.34)
NAM population % -- (.67)
Male:female sex ratio -- +.26*

* Not statistically significant at 95% confidence

The only surprise is in the sex ratio, with relatively more men being modestly correlated with more, not less, tranquility. Crime is a young person's game, so it's hardly shocking that states like Maine and Vermont, where the median age is over 40, are among the nation's least violent.

As always, higher IQ is positively associated with a higher quality of life. And also as always, ugly a fact as it is, the more black and Hispanic a place is, the lower that place's quality of life tends to be.

Although the report, at 52 pages in length, delves deep in discussing the results and implications of the index, and the racial composition of a state is apparently the single most determinative factor in how that state fares, there is not a single mention of race in the entire thing. Those of us interested in human biodiversity are often accused of being obsessed with racial differences. Well, race matters. It matter a lot. Reports like this, in which the most influential variable is inexplicably ignored, illustrate that rather than HBDers being excessively focused on race, the majority of people are tragically ignorant of it (or at least profess to be so) whenever paying attention risks casting non-whites in a bad light.

It's plausible to think that in places where crime is an acute problem, whites react by tending towards political candidates, generally Republican, who take a tough-on-crime approach, and that is reflected in the strong inverse correlation between peace and the percentage of whites who voted for McCain. Whites in the South vote heavily Republican--at rates in Alabama and Mississippi that nearly mirror black support everywhere else for whoever the Democratic candidate is--and the South, unsurprisingly, is rated the least peaceful in the country.