When I think about my career, scientific investigation was a large part of it. Therefore, analyzing where researchers go wrong in their analyses has been a significant part of that process. The following “mistake” is almost laughable. No, I take that back. It is hysterically laughable. If you ever had the pleasure of seeing a Gilda Radner skit as Emily Littela, this was a classic, “Oh, that’s very different. Never mind.” (Emily was an elderly woman commentator for a news program. But she was hard of hearing.)
[Source: Epic correction of the decade, by Steven Hayward]
Researchers at Virginia Commonwealth University published a paper in the American Journal of Political Science entitled “Correlation not causation: the relationship between personality traits and political ideologies.” They concluded that:
- conservatives are more likely to exhibit traits associated with psychoticism
- the “authoritarian” personality of conservatives had a genetic basis
They implied that “conservatism” might be a “treatable” mental aberration. So what’s your point, Dr. Filly? My point is that the researchers made a slight error. Get ready to enjoy one of the most epic academic face plants of all time.
Here are some quotes from the manuscript:
Here we test the causal relationship between personality traits and political attitudes using a direction of causation structural model on a genetically informative sample. The results suggest that personality traits do not cause people to develop political attitudes; rather, the correlation between the two is a function of an innate common underlying genetic factor…
In line with our expectations… “Psychoticism” is associated with social conservatism and conservative military attitudes… (Psychoticism’s) link with social conservatism is stronger for females while its link with military attitudes is stronger for males. (While) those higher in Social Desirability are… more likely to express socially liberal attitudes.
OK, say you. So what was the “mistake?” The mistake, answer I, is that the researchers inadvertently transposed the data sets for “liberals” and “conservatives.” Here is a portion of the retraction:
The authors regret that there is an error in the published version of “Correlation not Causation: The Relationship between Personality Traits and Political Ideologies” ... The interpretation of the coding of the political attitudes… in the manuscript was exactly reversed… Thus, where we indicated that higher scores… reflect a more conservative response, they actually reflect a more liberal response. Specifically… we… reported that those higher in psychoticism are more conservative, but they are actually more liberal; and where the original manuscript reports those higher in… social desirability are more liberal, they are, in fact, more conservative.
OMG! Liberals loved to report the “original” findings of this research. I wonder how they will deal with the retraction?