Monthly Archives: July 2011

Obama 2008 vs. Kerry 2004 in Kentucky

This is a quick follow-up to my previous post on “unhyphenated Americans” and support for President Obama in the 2008 election. Here’s a map of Kentucky’s counties, comparing the 2008 and 2004 presidential results. The blue counties are where Obama did better in 2008 than Kerry did in 2004. The red counties are where he did worse. (Data from The New York Times.)

The most dramatic swings were in eastern Kentucky, where Obama did between 10%-20% worse in 2008 than Kerry did in 2004. Things didn’t change much here in Boyle county. Kerry got 37% of the vote in 2004 and Obama got 38% in 2008.

The political puzzle of “unhyphenated Americans”

This is a map constructed from U.S. Census data on the distribution of people who claim “American” ethnic heritage. The census asks people this question: “What is this person’s ancestry or ethnic origin? (For example: Italian, Jamaican, African Am., Cambodian, Cape Verdean, Norwegian, Dominican, French Canadian, Haitian, Korean, Lebanese, Polish, Nigerian, Mexican,
Taiwanese, Ukrainian, and so on.)” This map shows the density of people who answer simply “American” to that question:

These are obviously not Native Americans. They’re white Americans who decline to indicate any European ancestral country of origin. It seems the highest concentration of these “unhyphenated Americans” is right here in Kentucky.

Now consider the following map of results from the 2008 presidential election to the 2004 presidential election (data from The New York Times).

The blue counties are those where Obama did better in 2008 than Kerry did in 2004 (just about everywhere). The red counties are those where Obama did worse.

The next two maps show the 2008 Democratic primary results: Obama vs. Clinton. In the first, Obama won the green counties and Clinton won the blue states.

In this one, Obama won the green states and Clinton won the red states (data from Dave Leip’s site):

These show that Obama performed dismally in states like Arkansas, Tennessee, Kentucky, and West Virginia, even among Democratic primary voters.

I noticed this general pattern shortly after the 2008 presidential election, but I was at a loss to explain why areas where people identify as “unhyphenated Americans” would seem to have very low opinions of President Obama. It can’t just be a partisan thing, because Obama did better than Kerry in some very, very conservative Republican states in the Rocky Mountain West.

This week a paper was published in Social Science Quarterly entitled “Barack Obama’s ‘American’ Problem: Unhyphenated Americans in the 2008 Elections“. The authors (Brian Arbour and Jeremy Teigen) are interested in this same question: what’s the connection between unhyphenated Americans and the lack of support for President Obama? They show statistical evidence that counties with higher levels of unhyphenated Americans were indeed less likely to vote for President Obama, even controlling for partisanship, religious context, racial context, and other sociodemographic factors.

They explain: “[These are] Americans who have abandoned or lost consciousness of their European heritage” (2). Why is this the case? The authors offer a few explanations: 1) perhaps its for patriotic reasons, 2) perhaps lower educational levels mean that people are simply unaware of their ancestral countries of origin, and 3) perhaps it’s because sufficient generations have passed that these Americans are simply forgetting their ancestral European heritage or no longer care about it.

Why are they concentrated in the particular geographical belt between Texas and Pennsylvania? They share some geographic history:

This region of Obama’s weakness follows the line of pre-Civil-War western migration of highland southerners, the farmers of the hardscrabble southern hills who never had enough money to buy land in the fertile Deep South and who moved from the Appalachian highlands across the upperSouth and into the near Southwest (Key, 1949). Historians and pundits have long associated this region and pattern of migration with the Scots-Irish (Fischer, 1989; Webb, 2004; Joseph, 2009). These are regions of the highland South where rocky and hilly terrain made plantation farming difficult and, as a result, African-American populations have always been smaller than in the Deep South. This region also corresponds to the areas where unhyphenated Americans are most highly concentrated.

As far as the link between “unhyphenated American” identification and support for President Obama, they write:

Unhyphenated Americans are concentrated in the South, the one major region of the country where Obama has never lived. Obama appears worldly, sophisticated, and urbane, whereas the aggregate data about unhyphenated Americans suggest that these adjectives would not apply to [most of them]. In addition to these mutable characteristics, two immutable characteristics of Barack Obama may lead unhyphenated Americans to see themselves as distant from him. One is his foreign roots, as epitomized by his name. Obama’s father was Kenyan, and Obama’s immediate connection to immigrant roots and a foreign country contrasts strongly with those who believe their ancestors come from America. Second is Obama’s race. As noted, unhyphenated Americans come from regions and demographic groups who are not well known for their progressive views on civil rights or their acceptance of integration and diversity.

To me, this sheds considerable light on the “puzzle” described at the outset and I commend the authors for their interesting and insightful analysis.

Nativism and opposition to health care reform

In 2009, former President Jimmy Carter came under a lot of criticism for claiming that at least some of the opposition to President Obama and his policies (specifically health care reform) was racially motivated (see story here). Of course, most people who oppose Obama are not racists, nor were most of the people who did not support his health care reform plan. Research has shown, however, that at least some of the opposition to the president and his policy proposals is linked to racial prejudice (see here, here, here, and here, e.g.).

It recently occurred to me that opposition to health care reform might also be linked to nativism, at least for some people. Why might this be the case? Nativism is the attitude that a distinctly American way of life needs to be protected against foreign influence. Might some people have perceived health care reform to be “foreign” or “un-American”?

A 2010 NPR story describes a gathering of Tea Party activists in southern Nevada. Referring to the new health care reform law, one activist stated: “I think it’s totally un-American, really.” Headline speaker Sarah Palin said: “Something’s not quite right when Fidel Castro comes out and says he likes Obamacare, but we don’t like Obamacare?” Another activist in attendance said: “If I was socialist I’d probably be for him [Obama], but I’m not a socialist.”

This is, of course, anecdotal evidence, but it suggests that some may have perceived health care reform to be distinctly un-American and associated with non-American cultures or systems such as Cuban communism or European social democracy.

A 2010 Hawkeye Poll, designed and fielded in part by yours truly, contained survey questions both on nativism and support for the health care law. Among Americans who opposed the health care reform law, 78% said that they believed that “our American way of life needs to be protected against foreign influence.” In other words, 78% of those opposed to health care also expressed nativist sentiments.

This can be further examined with a multivariate regression analysis. This isolates the effect of one variable (nativism) on another variable (support for the health care law), while controlling for the effect of all other variables in the analysis (such as demographics, religiosity, racial prejudice, and political ideology and partisanship). This analysis reveals that higher levels of nativism are associated with greater opposition to health care reform, even controlling for a host of other factors (for statistics geeks: p=0.006).

Not surprisingly, partisanship is the strongest factor in predicting opposition to health care reform (strong Republicans are 63% more likely to oppose the law than strong Democrats). Ideology comes next (strong conservatives are 51% more likely to oppose than strong liberal). Nativism is the third strongest predictor of opposition to the health care law, with strong nativists 34% more likely to oppose the law than non-nativists. In other words, even Democrats and liberals who are strong nativists are more likely to oppose the health care reform law. Anti-black racial prejudice also makes a difference, with those who are higher on racism 26% more likely to oppose the health care law. Apparently, nativism has a stronger effect on support for the health care law than racial prejudice.

The most interesting finding is that none of the other demographic variables made a difference once these other factors were controlled for. In this sample, at least, opposition to health care can be explained by a combination of partisanship, ideology, nativism, and racism.

The third and final piece of evidence is the results of a nativism “list experiment” that was also included in the 2010 Hawkeye Poll. A list experiment is useful because some people might be shy or embarrassed to admit that they think that “our American way of life needs to be protected against foreign influence.” So this experiment is able to measure “hidden” agreement with a particular statement or attitude. It does this by splitting up the survey respondents into two groups. Each group gets this question: “I’m going to read a list of things statements about things that some people think. Please tell me how many of the following items you agree with. Not which ones, just how many.” The first group gets four statements that are completely unrelated to the attitude you want to measure (in this case, nativism). The second group gets the exact same list and adds a fifth statement about protecting our American way of life against foreign influence. Theoretically, the difference in the average number of answers given by the respondents between the two groups is the amount of the population who has that attitude. In this survey experiment, “hidden” nativism is referred to as “latent nativism” (as opposed to “express” nativism).

The results of the survey experiment show that 62% of those who opposed health care reform were also latent nativists (p<0.00001 = very strong relationship). On the other hand, only 9% of those who supported health care reform were latent nativists, and this 9% difference was not “statistically significant”, meaning the results could have just as likely occurred at random.

Collectively, all this provides strong evidence that nativism accounts for at least some of the opposition to health care reform. Again, this is not to say that all people who oppose health care reform are nativists. However, this does suggest that it would be inaccurate to say that opposition to health care reform is purely a philosophical position based on the size and scope of government.

Authoritarianism and the debt negotiations

A few weeks ago I wrote a post about the role of “authoritarianism” in the 2012 GOP nomination, arguing that the higher level of authoritarianism among GOP primary voters will likely make it more difficult for Jon Huntsman to win the nomination (see post here). I also explained that “authoritarianism” is not being a Nazi or a fascist, but rather it is a personality characteristic that leads people to have a higher need for order, seek “clarity in the face of confusion” and eschew nuance and ambiguity. One of its key characteristics is that it leads people to see the world in terms of “black and white” instead of “shades of gray.” I also explained that even though all Republicans are not authoritarians, there is strong statistical evidence that there is a higher average level of authoritarian thinking among Republicans than Democrats in the general U.S. public.

What does this have to do with the current debt ceiling/deficit negotiations? I would think that people who see the world in terms of “black and white” would be less likely to support political compromises because they involve nuance and value trade-offs.

In a recent column, David Brooks (himself a moderate conservative) expressed dismay with the current Republican leadership for their unwillingness to compromise on the debt issue. He writes:

They [many in the Republican leadership] do not see politics as the art of the possible. They do not believe in seizing opportunities to make steady, messy progress toward conservative goals. They believe that politics is a cataclysmic struggle. They believe that if they can remain pure in their faith then someday their party will win a total and permanent victory over its foes.

Compare this to the following chart by Kevin Drum at the MotherJones website:

To me, all this descriptive evidence makes sense through the lens of political psychology. In the contemporary political world, Republicans are more authoritarian, and thus less prone to seek compromise. We should not be surprised when see this actually play out in the real world.

Quotes on public opinion

As a quick follow-up to my last post, here are some quotes about public opinion that I’ve listed at the top of my syllabus for this fall’s “Public Opinion and Voting Behavior” course:

“There are certain times when public opinion is the worst of all opinions.”  – Sabastien-Roch Nicolas de Chamfort

“Public opinion: an attempt to organize the ignorance of the community, and to elevate it to the dignity of physical force.” – Oscar Wilde

“Too often we … enjoy the comfort of opinion without the discomfort of thought.” – John F. Kennedy

“Public opinion in this country is everything.” – Abraham Lincoln

Public opinion is on Democrats’ side in debt negotiations

This is worth a read for anyone interested in the ongoing debt negotiations:

Nate Silver provides persuasive evidence that the Republicans, by refusing assent to any deal that includes any form of tax increase (or expiration of existing tax cuts), are putting themselves far out of step with public opinion – even among Republicans. He discusses the results of a public opinion survey which shows that the average Republican in America would prefer a deal that’s about 1 part tax increases and 3 parts spending cuts, while the average Democrat wants it about half and half.

I think that his evidence is persuasive and so I accept this conclusion. I have a few random responses:

  • No big surprise. Research has consistently shown that members of congress are ideologically more extreme than their partisan counterparts in the general public.
  • I don’t think this information will suddenly make the Republican leaders re-think their negotiating position. Even though public opinion might be closer to the Democrats, the debt negotiations are a very “wonkish” issue that most average voters don’t know all that much about and most will likely not base their vote on this issue next November. (Unless, of course, the U.S. defaults on the debt and we’re plunged back into a mega-recession. Then people will start to care, but because of its effect on the economy on not because of the details of a policy negotiation.)
  • As a general rule, I personally don’t believe it is always right for elected officials to base their decisions solely on public opinion polls. (That’s why we have a representative democracy instead of a direct democracy.) So I don’t necessarily fault Republican leaders for not conforming to popular opinion on this particular issue. (I fault them for other reasons, but not for failing to conform to popular opinion.)

So voting is stressful… but not for me!

This research article suggests that voting is a physiologically and emotionally stressful activity for most people to engage in. The researchers found that people going to vote tend to have higher levels of cortisol, or a “stress hormone” present in their systems than people who aren’t in a voting situation.

This suggests that voting is perceived to be a “burden” for most people. Personally, I think this is unfortunate. Democratic governments require only the most limited input from their citizens: casting a vote every couple of years to choose elected officials do to their decision-making job for them. In my view, that’s a pretty good bargain for the privilege of not being ruled by an authoritarian dictator.

I wonder what the researchers would find if they repeated the sample procedure on political geeks like myself who get their jollies out of making voting decisions and filling out a ballot in the voting booth? Increased endorphin levels, perhaps?