Monthly Archives: July 2010

More on Kentucky’s “red state” status

It may come as a shock to non-Kentuckians that registered Democrats actually out-number Republicans in the state, and by a fairly sizable margin. A full 57% of voters in Kentucky are registered as Democrats while only a mere 37% identify as Republicans.

Yet Kentucky is also a reliably “red state”, voting solidly Republican in the last three presidential elections and in most congressional elections.

In previous posts it’s been shown that Kentucky Republicans are very conservative while Kentucky Democrats are liberal on economic matters but rather conservative on social issues. We can observe further evidence of this by taking a look at voting patterns within partisan groups in Kentucky. This data is from the 2004 Voter News Service exit poll, so it’s a little dated, but still useful.

2004 Presidential election:

  • KY Democrats: 71% Kerry, 28% Bush
  • KY Republicans: 7% Kerry, 92% Bush

2004 Senatorial election:

  • KY Democrats: 81% Mongiardo, 18% Bunning
  • KY Republicans: 16% Mongiardo, 84% Bunning

There was also a constitutional amendment on the ballot to prohibit homosexual marriage, a good proxy for social ideology:

  • KY Democrats: 63% yes, 35% no
  • KY Republicans: 88% yes, 11% no

It’s interesting to note that a solid majority of Democrats in Kentucky (63%) voted in favor of traditional marriage in 2004.

The general trend that can be observed is that more Kentucky Democrats vote Republican than Republicans vote for Democrats. The overwhelming vote in favor of traditional marriage by Kentucky Democrats suggests that their social conservatism leads many of them to cross party lines when it comes to national elections, which might help explain why a state full of Democrats produces a solid red state in recent elections.

Government subsidies to fuel industries

A recent study by the Environmental Law Institute has found that “the federal government provided substantially larger subsidies to fossil fuels than to renewables.” This is presented graphically here:

As interesting as this finding is, I wonder what this graphic would look like if the numbers were displayed as a proportion of the total amount of money spent by the respective industries on each type of fuel. In other words, I wonder if government subsidies represent a proportionally larger share of the renewal fossil fuel industry than the fossil fuel industry.

Social capital and the economic downturn

A recent editorial by Doyle McManus of the Los Angeles Times discusses the “psychological fallout” of the current economic situation. He makes the following interesting observation:

But here’s something more surprising: As the recession deepens, participation in civic activities — community organizations, volunteer groups, even church attendance and social clubs — is likely to drop. Sociologists once assumed that during hard times people would naturally band together, if only to protest their plight or to give each other solace. It turns out that the opposite is true: Economic distress causes people to withdraw.

This concept of “participation in civic activities” is also known as “social capital” and for various reasons has numerous detractors among political scientists. Nonetheless, McManus’s article makes for an interesting read:,0,5206014,print.column

Partisan judicial elections

Bluegrass Politics reported on July 13th:

A federal appeals court on Tuesday cleared the way for Kentucky judicial candidates to raise money and run with political party affiliations, but ordered a lower court to consider whether the candidates can offer specific positions on issues. Judge Jeffrey Sutton of the Cincinnati-based U.S. 6th Circuit Court of Appeals said banning campaign fundraising and party affiliations violated the First Amendment’s guarantee of free speech. “Elections are elections, and the same First Amendment applies to all of them,” Sutton wrote for the three-judge panel.(

There are good arguments for non-partisan judicial elections. With no party affiliation, the election should theoretically focus on the characteristics of the candidates and judicial philosophy rather than how they would rule on particular issues.

In a 2001 article in the American Political Science Review, Professor Melinda Hall addresses these questions by examining state supreme court elections in 38 states that popularly elect their justices. She finds that partisan influences often manage to creep their way into supposedly “non-partisan” judicial elections:

“Nonpartisan judicial elections fail to insulate incumbents from partisan politics or other contextual forces. Both state-level partisan competition and election-specific partisan competition have dramatic effects on the vote received by incumbents. In fact, the simple act of having a partisan challenger reduces the vote share of incumbents by about 22%, even though partisan labels do not appear on the ballot. Clearly, partisan considerations have not been eliminated from these races.”

Given that there is little empirical evidence to support the arguments in favor of non-partisan judicial elections, perhaps the arguments in favor of partisan elections are worth a second look.

Food for thought


More information available here:

Predicting the midterm elections…

Ezra Klein has a very insightful article out in Newsweek in which he discusses what we should (and should not) focus on as we try to predict what’s going to happen in the midterm elections this November.

For decades now, political scientists have been building election models that attempt to predict who will win in November without making any reference to candidates or campaigns. They can get within 2 points of the final vote, and they don’t need to know anything about the ads and the gaffes and the ground games. All they really need to know about is the economy.

He goes on to discuss research which shows that the best predictor is the level of real disposable income in the public: the more money voters have in their pockets on election day, the better the incumbent party will do in the aggregate.

This also implies that a host of other “traditional” factors (BP spill, Kagan confirmation hearings, health care votes, deficit spending, etc.) will likely not make much of a difference to the overall results of the midterm elections.

Implicit Attitudes and the AZ Immigration Law

The study of “implicit biases” is becoming more and more common in research seeking to understand the nature of attitudes toward minority groups and governmental policies that affect them. My own dissertation research employed a study of “implicit nativist biases” – subconscious preferences for one version of American culture over another.

The American Psychological Association recently published a press release containing an interview with Dr. John Dovidio at Yale University. He explains this about implicit attitudes:

Implicit biases are beliefs (stereotypes) and feelings (prejudice) that are activated without intent, control, and often conscious awareness. These are habits of mind that develop through cultural as well as personal associations. Whereas most people no longer consciously endorse stereotypes and prejudice, the majority of people still harbor implicit biases.

Further, he argues that the presence of implicit racial attitudes toward Hispanic immigrants among most Americans will make it very difficult, if not impossible, for Arizona police officers to avoid racial profiling when carrying out the recently-passed Arizona immigration law:

APA. Will this new law lead to racial profiling within the state?

Dovidio. Stereotyping, prejudice, and biases in how people perceive and react to members of other groups typically occur automatically and with limited conscious control. These automatic processes are even more influential when people feel threatened or are under time pressure – common experiences for police officers – and thus will lead to systematic and racially/ethnically biased profiling.

APA. The Arizona governor has said that police in the state will be trained to properly apply the law. Will police officers in the state be able to overcome their implicit or unconscious bias?

Dovidio. Training of the type that is being proposed cannot consistently mitigate the effects of these implicit, and often unconscious, biases. Training may make people more aware of the potential for biased implementation of the law and help them understand better what they should be doing, but research has shown that training by itself cannot eliminate the systematic forces of implicit bias that operate unintentionally, often without awareness and the ability to control it. Training should help limit blatant abuses, but implicit biases will still play an important role in how the new police powers actually play out on the street.

The entire interview is available here:

More about implicit attitudes, as well as online implicit attitude tests that you can take, are available here:

Explaining Support for the AZ Immigration Law

Given that I just spent two years writing a dissertation on the determinants of immigration policy attitudes in the United States, I thought I might say a few words about the immigration law that was recently passed in Arizona. Essentially, this law “orders immigrants to carry their alien registration documents at all times and requires police to question people if there’s reason to suspect they’re in the United States illegally” (

Why might a person support or not support this law? Those who support the law say that it’s a simple law-and-order thing and it’s nothing personal against Latinos and not about preserving a “traditional” American culture from the “Spanish invasion.” Those who do not support the bill often think that its supporters are racists or “nativists.” So which one is a more accurate reflection of the motivations behind this law?

Here are the results from a logistic regression estimation of the determinants of support for a proposal to allow local police to check the immigration status of immigrants in Illinois, taken from a public opinion survey sample in 2006. In other words, here are a bunch of potential explanations for why a person might support a similar law (racism, political ideology, education, religiosity, etc.) and the computer sorts out which ones are strongly associated with support for the law:

Logistic regression Number of obs = 100
Wald chi2(11) = 17.45
Prob > chi2 = 0.0952
Log pseudolikelihood = -41.652413 Pseudo R2 = 0.2993

DVpolicech~k Coef. Std. Err. z P>z [95% Conf. Interval]

female -.407708 .6753963 -0.60 0.546 -1.73146 .9160445
age .0447008 .0286731 1.56 0.119 -.0114973 .100899
educ .3117166 .3079368 1.01 0.311 -.2918283 .9152616
income -.4021055 .2275677 -1.77 0.077 -.8481299 .043919
econnat 1.069313 .4764342 2.24 0.025 .1355191 2.003107
attend .0622249 .2149337 0.29 0.772 -.3590374 .4834872

Cforeignborn -22.67927 9.100616 -2.49 0.013 -40.51615 -4.842389
PrejHisp3 -.0183704 .2050105 -0.09 0.929 -.4201836 .3834428
ideo -1.002764 .567494 -1.77 0.077 -2.115031 .1095041
party5 .1159735 .2344021 0.49 0.621 -.3434462 .5753932
DVnativism1 .7167384 .3591134 2.00 0.046 .012889 1.420588
_cons -1.908899 2.643325 -0.72 0.470 -7.089721 3.271924

This analysis indicates that there are a few different things that explain why people support such a policy. A person’s gender, age, education, religiosity, political partisanship, and levels of anti-Hispanic racism do not matter. But a person’s perceptions of the economy, political ideology, feelings about American culture, and number of immigrants live around you DO matter.

All other things being equal, those who live in areas with lots of immigrants do NOT support this policy, those who think the economy is doing poorly, are political conservatives, and are more nativist (i.e. believe that “our American way of life needs to be protected against foreign influence”) DO support it. And living around other immigrants matters MOST in explaining support, with your economic perception and ideology mattering second most, and your thoughts about American culture coming in last.

Bottom line? Those who support the bill are not racists, despite claims to the contrary. They’re worried about law-and-order and a detrimental effect of immigrants on the economy. And some of them think that immigrants might “threaten” a particular American way of life. So cries of racism by those who oppose the law are unfounded. But it’s not purely a “law-and-order” thing either, as there are also economic and cultural threat considerations that play a role as well.