Politics, Personality and People With Red Faces
Conservatives might want to hold off on gloating about a recent social science screw-up, as rich as it may look. A team of researchers are being accused of letting their biases show by connecting conservative views with something called “psychoticism,” then doing a 180 and admitting in a correction that it was liberals who deserved the unflattering-sounding label.
For anyone who enjoys a spectacle of shame aimed at social scientists, liberals, and especially liberal social scientists, the story is a little too good to be true. For one thing, psychoticism is not as terrible as it sounds.
It turns out that psychologists use a perverse system of nomenclature. Psychosis is a symptom characterized by loss of contact with reality, but psychoticism is a whole other thing -- a personality trait that’s been more strongly connected to creativity than to mental illness. Maybe conservatives will want it back, given all the emphasis these days on “innovation.”
The term is confusing, admits University of Pennsylvania psychologist Robert DeRubeis. All labels for personality traits, including psychoticism, were invented to describe normal human variation. Most people are unfamiliar with the term psychoticism because it’s not an important part of contemporary mainstream psychology. As personality traits go, it’s not part of what are known as the big five: extroversion, conscientiousness, openness, neuroticism and agreeableness.
What psychoticism means depends on whom you ask. DeRubeis sent a test with an answer key, which revealed that you could get a high psychoticism score by saying you don’t care much about manners, morals, marriage, family, obeying laws, being in debt or what other people think of you. Your psychoticism score goes up if you say you have enemies or that you’ve experimented with illegal drugs.
The 2012 paper that started the whole flap associated the trait with “authoritarianism,” though the questions would suggest the opposite -- that perhaps on the low end of the scale would be uptight, pearl-clutching church ladies and on the high end, Silvio Berlusconi. Most reasonable people would be in the middle. “You wouldn’t think that having a modicum of psychoticism would be a good thing,” DeRubeis said, but as with all personality traits, balance can be an advantage.
It still might be interesting to see if this trait has anything to do with political leanings. Penn State University political scientist Peter Hatemi, who was one of the authors of the corrected paper, says that’s what he intended to find out. His conclusion was that psychoticism didn’t push people in either direction along the liberal-conservative spectrum.
He admits to making a goof. Data gathered by other researchers showed a correlation between psychoticism and liberalism. Hatemi said he and his colleagues showed that this correlation was spurious, though they wrongly said the spurious correlation was between psychoticism and conservativism. He said he and his colleagues added a correction last summer, but it doesn’t change his conclusion that this trait doesn’t make people more conservative or more liberal.
Last week, however, the website Retraction Watch seems to have gotten Hatemi’s claim backwards with a post headlined, "Conservative political beliefs not linked to psychotic trait, as study claimed," thus implying the study claimed a link exists when it claimed the opposite.
But something has to cause people to be liberal or conservative. Are people born tending toward liberalism or conservatism or do political leanings emerge from contact with other people? If you switched Bernie Sanders and Ted Cruz at birth, would they be running on the opposite platforms? If Hillary Clinton were brought up shooting moose in Alaska, would she have run as a Republican? (In actual fact she was brought up in Illinois, her father was conservative and she was an active young Republican.)
University of Nebraska political scientist John Hibbing put the nature-nurture influence on politics to the test by comparing sets of fraternal and identical twins. If a trait is controlled partly by genes, it will be shared more often by the identical twins, who share all their genes, than the fraternal twins, who only share about half. Both types of twins would presumably share similar environments.
Comparing twins this way has been used to show that there’s a genetic component to autism, religiosity, sexual orientation and a host of other traits. Hibbing found that political leanings, too, were shared more often by the identical twins. That research is controversial even if it’s not surprising that political views are shaped by a complex mix of internal and external factors.
Hatemi said a number of studies have investigated a possible connection between politics and the kinds of personality traits that psychologists like to measure with their questionnaires. His conclusion is that none of them influence people’s politics in predictable way. Others disagree, he said, but there’s no reason his claim should offend people on either side of the political spectrum.
The story that’s generating much chatter about the research error is built mostly on assumptions, including the notion that the researchers must be typical liberal academics. Hatemi said one of his collaborators is an Anglican priest and another is a supporter of the conservative party in Australia. He’s served in the military, he said, and while his views are mixed, “all of my adult life I have been connected to national defense and support of our service men and women in some form or another.” It’s probably not what anyone would have assumed.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Faye Flam at firstname.lastname@example.org
To contact the editor responsible for this story:
Jonathan Landman at email@example.com