Attitude extremity is measured by a questionnaire on three dimensions: economic, social, and environmental.
I notice that of the 8 economic questions, three are essentially the same question rephrased (labeled "eqwlth", "eqincome" and "goveqinc") and two additional questions are very closely related ("helpsick" and "helpoor"). It's not clear how these 8 questions do a very good job of measuring attitude extremity along the economic axis.
Questions along the other axis also seem too nuanced for a simple 1-5 answer (agree-strongly to disagree-strongly). For example are nuclear power stations "extremely dangerous" to the environment. Well, in my mind, of course they pose some danger, but less so than the coal plants they often replace. I'd have to disagree strongly that they are "extremely" dangerous (i.e. be classified as having an extreme attitude). I'm not trying to start an argument over the pros and cons of nuclear power here, but just wanted to point out that the questions selected by the researchers introduce their own bias where shallower answers seem more centrist.
Perhaps I'm too sensitive about these kinds of classifications. I consider myself a libertarian and usually feel poorly categorized by surveys.
Yeah, really. I have extremely left-wing opinions, to the point that most Americans would indeed consider me a left-wing extremist. However, these questions seem to me to fail to capture the interesting degrees of variation. For example, there's no option given for the right-wing position that poor people are not merely poor due to their own faults, but that in fact government policy should be used to appropriate what little they have and make them even poorer, or that foreigners should not merely be left outside the United States via restrictive immigration, but bombed into oblivion. Neither is there an option for saying that we should not merely use taxation to redistribute wealth to the poor, but actually seize the means of production and abolish wage-labor. The statement about Irish, Italian, and Jewish Americans fails to capture what really happened in the post-WW2 decades to integrate those minorities. The questions on science versus religion contain no options for fundamentalist religion or transhumanism (despite the fact that Evangelical Christians make up 20% of the total population!). The question about marriage doesn't ask whether husbands should be able to beat or rape disobedient wives. I can't even answer the question about driving a car, because I don't.
The interesting thing about studies of extremism lies in finding out exactly which views are considered the normative mainstream, and which views are considered "out-of-sight".
In US terms I'm a liberal, and I agree there's a lot to question in this study.
I'd certainly want to see a replication with more thorough methods and more informed question design before drawing any conclusions.
I'm not even sure it makes sense politically. In this model, who are the swing voters? If older people have entrenched views and younger people don't vote, how does this information correlate to election-winning voter swings in the gap between the two?
The measurement of the target variable, here attitude extremity, seems a little shaky to me.
First of all, using min and max values is not very smart because of outliers and complete disregard of the "coherence" of the opinions, i.e., if standard deviation is small or big (comparatively).
Secondly, the formula is not symmetric. Maybe that's ok because the distributions aren't but using the maximum in the denominator of the attitude extremity formula just increases the effect my first point.
I think a definition in terms of percentiles would have been probably much better. This would implicitly define attribute extremity in terms of non-conformity but that's IMO exactly what it is, it would just sound less dangerous and more 'revolutionarish'.
This topic roughly reminds me of Jeff Schmidt's book "Disciplined Minds"[1], where he argues that professionals are likely to be more conservative and trusting of the government that other people, in part due to the process of professional education. Here is an excerpt from the first chapter:
> Evidence that professionals are not independent thinkers has been around for a long time but has generally been ignored, in part because people don't know how to make sense of it. The Vietnam War produced some revealing examples, which are worth looking back at.
> On 12 January 1971, the federal government indicted Philip Berrigan and other East Coast antiwar activists on felony charges of plotting to impede the Vietnam War through violent action. The activists' agenda supposedly included blowing up underground heating pipes in Washington to shut down government buildings, kidnapping presidential advisor Henry Kissinger to ransom him for concessions in the war and raiding drafting boards to destroy records and slow down the draft.
> The Justice Department prosecutors chose to hold the conspiracy trial in Harrisburg, Pennsylvania, a conservative area where a randomly chosen jury would be heavily against the defendants. However, before the jury was selected at what came to be known as the Harrisburg-7 trial, a group of left-leaning social scientists supporting the defendants interviewed a large number of registered voters in the area to try to figure out how to get a sympathetic jury there. They discovered, amongst other things, that college-educated people were more likely than others to be conservative and trust the government. Thus, in court, during the three weeks that it took to examine 465 potential jurors and pick a panel of 12, lawyers for the defense quietly favoured skilled blue-collar workers and white-collar workers without a lot of formal education -- nonprofessionals, although the sociologists and lawyers never used that term.
> The lawyers were uneasy about doing this, however, because it went against their intuition. The notion of closed-minded hard hats and open-minded intellectuals is widespread and is reinforced by mass-media characters like dock-loading worker Archie Bunker and his college-student son-in-law, "pinko" Mike. In fact, All in the Family made its television debut the very day of the Harrisburg indictments, 12 January 1971; by the time the trial and jury selection started, it had been on air for over a year.
> Ignoring these false stereotypes paid off. The government put on a month long, $2 million extravaganza featuring 64 witnesses, including 21 FBI agents and 9 police officers. The defense called no one to the witness stand. After seven days of deliberation, the jury was not able to reach a unanimous decision, and the judge declared a mistrial; but with 10 of the 12 carefully selected jurors arguing for a not-guilty verdict, the government dropped the case.
I am not sure. That said, it does seem that people from the US place an unusually large degree of trust in business, and less trust in government [1]. I have read that there has been a great deal of money spent in the US on pro-business propaganda over the last century [2], so perhaps results like these are not so surprising!
The field of economics really has done itself a disservice in its relevance as a science by becoming so intertwined with politics.
Wait- what? The authors' credentials on the abstract page are different from their credentials in the pdf (which the site won't let me re-open). Hmmm.
Also, early on the authors fail to include an apostrophe that is needed to make the word 'individuals' a possessive plural. A detector is starting to go off in my mind.
I suspect that the PDF states the authors' positions when the paper was written and the SSRN page states their current positions.
According to http://www.michaelmakowsky.com/about/ Michael D. Makowsky is currently a professor at Clemson University (as listed on the SSRN page) and was previously a professor at the Center for Advanced Modeling at Johns Hopkins University (as in the PDF).
I notice that of the 8 economic questions, three are essentially the same question rephrased (labeled "eqwlth", "eqincome" and "goveqinc") and two additional questions are very closely related ("helpsick" and "helpoor"). It's not clear how these 8 questions do a very good job of measuring attitude extremity along the economic axis.
Questions along the other axis also seem too nuanced for a simple 1-5 answer (agree-strongly to disagree-strongly). For example are nuclear power stations "extremely dangerous" to the environment. Well, in my mind, of course they pose some danger, but less so than the coal plants they often replace. I'd have to disagree strongly that they are "extremely" dangerous (i.e. be classified as having an extreme attitude). I'm not trying to start an argument over the pros and cons of nuclear power here, but just wanted to point out that the questions selected by the researchers introduce their own bias where shallower answers seem more centrist.
Perhaps I'm too sensitive about these kinds of classifications. I consider myself a libertarian and usually feel poorly categorized by surveys.