The ultra-citified are islands of insanity in an America that is still a sea of reasonableness.

An excellent column, Elite? Or Ultra-Citified? by Arnold Kling.  The subheading is A buzzy poll does not survey who it claims to survey.  

In recent weeks, you probably have seen articles about a Rasmussen poll (actually conducted last September) showing that elites do not want the rest of us to eat meat, cook with gas, engage in “unnecessary air travel,” and so on. Before piling onto this poll with my own commentary, I first wanted to look at the way it was conducted. Polls, like documentaries, can easily be used to construct a narrative rather than provide objective information.

I have seen numerous opinion pieces, articles and tweets stemming from this study.  And have read none of them.  The noise and chatter strike me as manufactured.  If there is something interesting, it will still be being discussed in six months.  If it is the manufactured fluff it seems, no one will remember in six months.  

Kling does the legwork that justifies my jaundiced perspective.  

The most comprehensive description I can find is a report by the Committee to Unleash Prosperity, which commissioned the poll. The first thing I noted was the criteria for selecting the “elite.”

The Elites are defined as those having a postgraduate degree, a household income of more than $150,000 annually, and living in a zip code with more than 10,000 people per square mile. Approximately 1% of the total U.S. population meets these criteria.

The requirement of having a postgraduate degree and a household income more than $150,000 strikes me as not very strict. Google tells me that fourteen percent of Americans have a postgraduate degree, and twenty percent of U.S. households meet the income threshold. Even if there were zero correlation between income and education, almost 3 percent of the population would meet the criteria. But because the correlation between income and education is pretty high, a conservative estimate is that between 5 and 10 percent of the population is “elite” by those standards.

I can infer that the requirement to be “living in a zip code with more than 10,000 people per square mile” is doing a significant amount of work here in whittling the “elite” down to one percent. This is confirmed by checking out various zip codes with which I am familiar. My oldest daughter lives in a zip code in tony Newton, Massachusetts, where many households meet the elite income and education thresholds. But the population density there is less than half of what is needed to qualify for the Rasmussen survey. The same is true for many of the affluent zip codes near me.

In practice, a population density of more than 10,000 per square mile almost assures that high-rise buildings will supply a big share of housing. You are pretty much selecting for large northern cities, like New York, Chicago, and Boston. And you are selecting against some of the most affluent suburbs near those cities. Wealthy residents of large homes in Beverly Hills, California or Potomac, Maryland are going to be excluded from the sample.

In the United States, the typical person living in a zip code with at least 10,000 people per square mile is poor. I would bet that a large fraction of the zip codes with that much density are slums.

When you think of the elite, you think of the most affluent, highly-educated Americans. You think of what Charles Murray says is represented by Belmont. Residents of Belmont are not in the Rasmussen poll because Belmont fails to meet the density requirement.

Instead, what this sample is picking is people who are indeed well above average in terms income and education, but whose most distinguishing characteristic is choosing to live in crowded sections of older cities. The best label I can come up for this demographic is Ultra-Citified Upper Middle Class, or ultra-citified for short.

You know who else this looks like?  The natural environment of the remnant journalists from the legacy mainstream media.  My longstanding plaint has been part of the disconnect of the Mainstream Media has been that they are bunkered in half a dozen unrepresentative cities sharing few of the conditions and concerns of the average American.

If you live and breathe center city New York, Chicago, LA, Washington, D.C. and San Francisco, of course you are concerned about race relations, inequality, urban decline, immigration, social justice, poverty, homelessness, etc.  Those are important issues in those cities.  And mostly, primarily in those cities.  

Elsewhere, not so much.

Kling is indirectly pointing out that the survey design, intentionally or not, has been designed to pick up not the actual movers and shakers of America but rather the wealthy and educated of a few zip codes in a few cities.  

This an example of the Millionaire Next Door problem.  Who you consider a mover and shaker is a matter of definition.  Most definitions are colloquial and loose.  Once you get specific and tight, the movers and shakers are quite different than you expected.

Kling is point out that the survey is of little value because it has effectively chosen, through its density requirement, to only survey rich educated people in a handful of urban zip codes who are already known to be disproportionately left leaning.  

Had they surveyed wealth and educated alone or any other proxy for power and influence (wealth, income, education, social status, etc.) they would have gotten dramatically different answers.  And probably much less remarkable answers and therefore a much less noteworthy survey.  And who wants to conduct an expensive survey that merely affirms that which is already known or believed?

I agree with Kling's conclusion.

But if you were to study more broadly the affluent and highly-educated portion of the American public, you would find a diverse set of values and beliefs. It is not just “the deplorables” living in small towns that are alienated by the ultra-citified. It is not just the working-class urban neighborhoods that Murray symbolizes with Fishtown. The ultra-citified also alienate much of Belmont.

The ultra-citified are islands of insanity in an America that is still a sea of reasonableness. At a national level, the ultra-citified are destined to lose most of the political battles.

As I have pointed out in the past, our legacy mainstream media beacons shine from ultra-citified zip codes.  What undermines this survey is what also undermines legacy mainstream media reporting.  It is unrepresentative and substantially irrelevant.