Exceedingly popular these days in the social “sciences” are studies purporting to show that conservatives are a deranged bunch, full of outré psychological traits that make them vulnerable to authoritarian come-ons. Many have been the social-psychologists and behavioral and political scientists who have claimed to find that conservatism is, if not exactly a form of insanity, then at least an unfortunate mental impairment.
The real impairment, however, appears to be in the statistical skills of some of these “scientists,” who persist in publishing howlers. And the most recent embarrassment for the field is a doozy.
A team of academics published in the American Journal of Political Science in 2012 their findings that conservatives tended to score high on the psychoticism scale. Now, “psychoticism” doesn’t mean one is psychotic, but it isn’t exactly a term of endearment. According to the textbook The Psychology of Personality, psychoticism “is associated with the tendency to be impulsive, cold, not empathic, unconcerned about the rights and welfare of others, and antisocial.” The very definition of a conservative!
At least that’s what Brad Verhulst and Lindon Eaves of Virginia Commonwealth University and Peter K. Hatemi of Pennsylvania State University must have thought. The good professors took a big barrel-full of preexisting study data and compared where thousands of subjects fell on the psychoticism scale (their “P-scores”) to where they fell on the political spectrum. Their analysis of the data found “that those higher . . . in psychoticism are more conservative.”
This research not only earned plaudits, it was widely quoted in other social-science journals. The only problem was that the researchers got tripped up in the “coding” of the subjects’ responses (an error that was caught by psychologists Steven G. Ludeke and Stig Hebbelstrup Rye Rasmussen). The skeptics’ exposé led the authors of the original paper to publish a correction that began: “The authors regret that there is an error in the published version of ‘Correlation not Causation: The Relationship between Personality Traits and Political Ideologies.’ ”
Oh, and what an error it was! You see, the authors admitted getting the “coding of the political attitude items . . . exactly reversed.” In other words, “where we indicated that higher scores in Table I reflect a more conservative response, they actually reflect a more liberal response.” Instead of being associated with conservatism, high P-scores “are actually more liberal.”
Why did it take so long for such an extraordinary screw-up to be discovered? Why is it that hundreds of other social scientists cited the original research without questioning it?
Could it be that the original results fit so neatly with the prejudices of the academic left that they couldn’t imagine they were wrong? One thing is for sure: Had Verhulst, Eaves, and Hatemi found, from the get-go, that liberalism was associated with undesirable traits, their research wouldn’t just have been challenged, it would have been put through the wringer. You call that methodology? Let’s see your data! Let’s recrunch those numbers!
Instead, the left-wing monoculture that is academic social science finds it all too easy to believe the things they want to believe, even when they’ve got it exactly backwards.
