Let’s examine Marcus Arvan’s peer-reviewed paper, “Bad News for Conservatives? Moral Judgments and the Dark Triad Personality Traits: A Correlational Study”, published in the 25 July 2011 issue of Neuroethics.
Through the generosity of the “Yale Experiment Month initiative, a program financially sponsored by the American Philosophical Association for the purpose of encouraging experimental research by philosophers”, Avran went on-line and found “302 male, 257 female, 2 transgendered” persons to take part in a survey.
People self-reported, on a 1-7 scale, their fiscal and social conservatism or liberalism. (Arvan self-reported himself as a philosopher.) His subjects self-reported their age, sex, and several other variables that never appeared in Arvan’s analysis (why?). He also asked questions that sought—reminder: to seek does not imply to find—opinion on contentious political topics. He had the idea that conservatives would be more “‘hard-hearted’ or callous'” than liberals. Statistics and small p-values would vindicate him.
How did he come by this hypothesis? “Commonsense,” Arvan tells us, “suggests that there is a relationship between personality traits and moral value judgments.” He noted that “Adolf Hitler”—who he helpfully reminds us was a “mass murderer”—had the “notable personality trait” of “counteractive narcism,” [sic] which is “a type that is stimulated by real or imagined insult or injury” and that the briefly mustachioed dictator entertained the “notable moral judgment” of anti-Semitism. Arvan forgot that Hitler was a non-smoking, vegetarian, teetotaler. Perhaps these (more liberal than conservative) personality traits drove him to the Final Solution?
A typical Arvan question: “Homosexual behavior is: morally wrong, morally bad but not forbidden, morally neutral (neither good nor bad), morally good but not required, morally required.” Now, there are many arguments about the propriety of homosexual behavior, but to suggest it is “morally required” is not common. It is not clear whether this response is understandable to the audience Arvan polled. It is, at the least, ambiguous; at worst, nonsensical. He also had the same question and responses for gay marriage. “Morally wrong” is clear here, as is “morally right”, but “morally required” is not.
The wording on his economics questions is worse. Example:
A government ought to tax its citizens in order to ensure that all citizens enjoy basic life necessities (example: Social Security, which provides old-age/retirement benefits, and temporary assistance for needy families, such as food stamps etc.).
Who but a cold-hearted brute would withhold “basic life necessities” from his neighbor? Arvan’s word choice reveals his obvious bias: it ignores the conservative argument—whether you agree with it or not is irrelevant—that low or no taxes are better guarantors of “basic life necessities” than is government control. A person’s sympathies with his brother is confounded with his judgment of what makes sound tax policy.
And then there is his amateur psychological quiz. Answer (1 through 5 here) high on the question “I am a thrill seeker” and, according to Arvan’s view, you might be a psychopath. Say that you “like to get acquainted with important people” and Arvan says you’re a narcissist. Or agree that “It’s not wise to tell your secrets” and you are Machiavellian. Answer positively on all three and you descend into the “Dark Triad.”
Says Arvan. Not directly, of course, but by calling these questions part of an “instrument” with “face validity” and by quoting others in the “growing body of literature” he rubs a patina of legitimacy on his survey. Finding correlations between arbitrarily labeled questions all the rage and guarantee the road to publication. Over confidence in their results abounds. Why?
It might be true, for example, that all psychopaths would agree that “People who mess with me always regret it” but it is far from obvious, and probably false, that all psychopaths would openly admit it on a public survey. It is also unclear how many non-psychopaths would also agree with this and other similar questions.
Finally comes Arvan’s statistics, which is nothing more than Pearson straight-line correlation between answers on the personality and political and moral questions. Because of the discrete form of the responses, he should not have used this measure: doing so can inflate correlation. Most of his correlations were (in absolute value) in the 0 to 0.4 range, most were near 0. But they all had small p-values, which is unfortunately all you need to publish. Suspiciously, he never reports what percent of respondents were conservative, liberal, or religious, nor does he give any demographic breakdown (besides noting “gender”).
To the results! He found, for example, that psychopaths are more likely to agree that “A government ought to detain suspected terrorists as long as necessary without trial to prevent terrorist attacks”, but only to the tune of a dismal 0.2 correlation. He calls this finding a confirmation of “conservative judgment”, which means he doesn’t follow current politics closely, as calls for eliminating habeas corpus are now more likely to originate from the White House than the editorial page of the Wall Street Journal.
His says his conclusions are “bad news for conservatives.” Time and again Arvan implies that conservatives are Machiavellian, Narcissistic, and psychopathic—“‘morally worrisome’ traits” in his words. Yet he gushes that his conclusions are “provocative” and “significant” and that luckily they raise “many questions worthy of further research.” After putting a good word in for ad hominem arguments, he states he has raised “provocative moral questions worthy of further philosophical research.”
A psychologically more interesting question that they ones “raised” by Avran are how these sorts of papers ever see the light of day. If anybody wants to fund such as study, then I am your man.
Arvan’s work has also been picked up in the press.