Reject Your Personality’s Politics

Tyler Cowen points us to a Psychology Today article on how your personality correlates with whether you are conservative or liberal.   When reading below about the correlations, resist the temptation to focus on which side is flattered the most.  Instead, see these as markers of bias.  Unless you have a better than average reason to think your personality type has better insight into what policies are good for society, it is a bias to let your policy beliefs be influenced by your personality.  Instead try to move your beliefs to those you would have had with an average personality.  Those correlations:

Liberals are messier than conservatives, their rooms have more clutter and more color, and they tend to have more travel documents, maps of other countries, and flags from around the world. Conservatives are neater, and their rooms are cleaner, better organized, more brightly lit, and more conventional. Liberals have more books, and their books cover a greater variety of topics.  …

Liberals are more optimistic. Conservatives are more likely to be religious. Liberals are more likely to like classical music and jazz, conservatives, country music. Liberals are more likely to enjoy abstract art. Conservative men are more likely than liberal men to prefer conventional forms of entertainment like TV and talk radio. Liberal men like romantic comedies more than conservative men. Liberal women are more likely than conservative women to enjoy books, poetry, writing in a diary, acting, and playing musical instruments. …

As kids, liberals had developed close relationships with peers and were rated by their teachers as self-reliant, energetic, impulsive, and resilient. People who were conservative at age 23 had been described by their teachers as easily victimized, easily offended, indecisive, fearful, rigid, inhibited, and vulnerable at age 3. …

Conservatives have a greater desire to reach a decision quickly and stick to it, and are higher on conscientiousness, which includes neatness, orderliness, duty, and rule-following. Liberals are higher on openness, which includes intellectual curiosity, excitement-seeking, novelty, creativity for its own sake, and a craving for stimulation like travel, color, art, music, and literature. … conservatives have less tolerance for ambiguity, …

Of course many say the focus of these results is biased:

Yet critics retort that the research draws negative conclusions about conservatives while the researchers themselves are liberal. … a disproportionate amount of the research has focused on figuring out what’s behind conservative behavior. … "There is a bias among social scientists," admits Glaser. "They look for the variables that are unflattering."

GD Star Rating
Tagged as:
Trackback URL:
  • Doug

    This is an interesting blog. A question regarding the advice given with the study results, though: What rubric was used to determine conservatism vs. liberalism? While identification may be fairly constant, in the United States in the 1990s, to be conservative implied a libertarian bent. In the present decade it seems to imply an authoritarian bent. Likewise, a decade ago if you asked me what it meant to be liberal, I would have said that, at its core, liberalism was the belief that the will to act collectively is more persuasive than constitutional or practical concerns. Today, liberalism seems to have replaced statist tendencies with historical and constitutional deference.

    So, I wonder if the study above is measuring personality and experience against political philosophy or, rather, against reaction to current regimes and popular pundits.

  • conchis

    “Unless you have a better than average reason to think your personality type has better insight into what policies are good for society, it is a bias to let your policy beliefs be influenced by your personality.”

    Well, the major strand of argument in Tetlock’s “Expert Political Judgement” is that those with a high need for closure and low tolerance of ambiguity perform worse (in terms of both discrimination and calibration) in predicting the course of future events. That seems like a prima facie reason to let your policy beliefs be influenced by your personality. Admittedly, in Tetlock’s sample there was virtually no correlation between the cognitive style and ideology, but given the preponderance of other evidence, it’s not clear how much one should read into that. It’s also an open question how much this result generalises to spheres beyond that which Tetlock focused on (IR), but the underlying mechanisms seem to be quite general: greater/lesser tendencies to be overconfident, ignore conflicting evidence, update beliefs appropriately etc.

  • michael vassar

    It’s also possible that conservative and liberal politics follow from different values which are correlated to personality. However, it isn’t bias to have preferences, so it is not implied that one should compensate.

    It seems plausible to me that personalities and political preferences do relate to values which directly reference bias however. Potentially, conservatives or (less plausibly to me) liberals could tend to have less desire to be unbiased.

  • Robin Hanson

    Doug, I’d guess they just asked people to label themselves, but you can follow the references to see for yourself.

    Michael, I was clear to talk about what is good for society, not just you.

    Conchis, yes, some personality features may plausibly correlate with accuracy.

  • michael vassar

    Arguably different people value different things for society, not just for themselves.
    Utilitarianism is in many respects not well defined, and at any rate there is no strong prior for giving it precedence over other forms of consequentialism.

  • Robin Hanson

    Michael, can you honestly say “if I had a different personality, then what is good for society would be different”? If not, my point stands.

  • Mike Bendzela

    I disagree with the whole clubby, dichotomizing tone of this.

    What’s more interesting is not how our personalities (which have infinite permutations) influence our “beliefs,” but how quick we are to polarize, associating ourselves with one “side,” disassociating ourselves from the “other side.”

    “(1) When individuals having no established relationships are brought together to interact in group activities with common goals, they produce a group structure with hierarchical statuses and roles within it.

    (2) If two in-groups thus formed are brought into functional relationship under conditions of competition and group frustration, attitudes and appropriate hostile actions in relation to the out-group and its members will arise and will be standardized and shared in varying degrees by group members.”

  • Robin Hanson

    Mike I had in mind inferring your personality’s political position on a continuum of political positions, not a dichotomy.

  • Curt Adams

    The accusation of bias seems biased to me. Apart from the childhood personality reports, I don’t see anything pejorative about conservatives. The only perjorative comment is that liberals are called messy. I’ve heard the early personality reports before, and they align with what John Dean wrote about authoritarion personalities as adults in “Conservatives without Conscience”, where he referred to a substantial scientific literature. If anything, the authors of this study made an effort to avoid negative comments by referring to childhood personality traits rather than adult ones.

  • TGGP

    If, as Bryan Caplan suggests, the political beliefs of the average person are highly biased, is it such a good thing to move toward the average? Furthermore, is there any reason to believe that a person who is already biased will have an accurate perception of what beliefs they “would have had with an average personality”? To me it seems the corrective nature of such advice would be about the same as telling people they should shift their personalities toward the average.

  • Robin Hanson

    TGGP, obviously if you know of other biases then you should correct for those as well, and if you can’t tell what your politics would be with the average personality, then my advice to move there isn’t much use to you.

  • michael vassar

    In extreme cases I can definitely say “if I had a different personality then what is good for society would be different”. For instance, if I had the personality of a member of most non-human species, then what is good for society would be very different.”
    The Brave New World explored these issues fairly well. If one has the personality of almost any participant in that society, that society is FAR better than ours, as those of our values that don’t exist in that society also don’t exist for you. For us, the “Brave New World” society fails to embody many traits which are necessary for a good society.
    Value is inherently subjective, and like all subjective things, that in no way makes it not “real”.

  • Groovy

    I bet that liberal wrote that. 😉

  • Robert Scarth

    Michael Vassar said “In extreme cases I can definitely say “if I had a different personality then what is good for society would be different”.”

    I don’t think you need to be extreme at all. Consider free trade. If trade is free then the nation is in aggregate better off – this is an uncontestable fact. However if I believed that social stability was more valueable than material wealth, and that it was not captured by a market process then I could consistently accept that while free trade might make us materially better off we would be better to restrict trade and so keep society stable. Another person might not value social stability so highly and believe that trade should be free. I think this is quite a common and regular example.

    This example comes about because, as you point out, value is subjective.

  • Robin Hanson

    Robert, social stability might be better for society than material wealth, and whether you thought so might depend on your personality, but whether stability was more good for society would not depend on your personality.

  • Eliezer Yudkowsky

    I don’t consider myself a conservative or a liberal, but, speaking strictly as a rationalist, I find this study’s results *highly* suspicious. When a group of steadfastly Republican researchers conduct a similar study and find the same results, I’ll start paying attention.

  • Eliezer Yudkowsky

    Robin, if you think there’s a continuum between Blue and Green, that’s buying into the “dichotomy” nearly as much as if you thought there were only two discrete points.

  • Robin Hanson

    Eliezer, surely there is a continuum between blue and green, which also extends past them both. There are of course many other dimensions one could identify and focus on.

  • pdf23ds

    Robin, I think Michael’s point is that different people with different values might have different opinions on which societal *outcomes* would be good or bad, not different opinions on which policies would lead to which outcomes. Given a single, detailed, view of a hypothetical society, one person would say “this society is bad” and another “this society is good”. When one asks “If my personality were different, would what is good for society be different?”, it assumes that “what is good for society” is some objective thing. But it’s really not. It’s a subjective thing.

    An example: If your current society contained two mutually antagonistic groups, with extreme animosity between them, members of each group might think that a better society would be one where the other group had been killed.

  • Robin Hanson

    pdf, yes of course people can have different preferences over the sort of society they would prefer. And if they admit they just have different preferences, there is nothing to disagree about. But usually when people speak about what is “good for society,” they talk as if what they intend to refer to is a more common concept, about which they disagree.

  • pdf23ds

    Robin, I think that even when people of different ideologies *think* they’re talking about a common concept of good society, they usually have different ones in mind. So they end up having low-level discussions (about policies or politicians, for instance) that are distorted by high-level differences in values driven by their personality.

    And to some degree, this is complicated by the fact that people *aren’t* very pragmatic about many areas of policy. They desire certain policies in themselves, and not just for the effects. Abortion is an obvious example, and police power and sentencing rules a less obvious one. So, two people might agree on what would make a good society on the *most* abstract levels, (say, that it would provide everyone opportunities to fulfill their own happiness,) but be unwilling to change their positions on some lower-level pragmatic issue even if it were shown that the lower-level issue did not lead to the higher-level results.

    So I think the distinction you’re trying to make isn’t really that clear.

  • Robin Hanson

    pdf, my advice to move to the political opinions you would have if you had an average personality applies only to those who think their political opinions are about a common concept. If you think it is about your preferences, then my advice doesn’t apply to you.

  • Joseph Hertzlinger

    I’m dubious about studies (e.g., the “whiny kid” study) based on a sample size of 95 in an atypical community. The conclusion might be plausible (speaking as a reactionary crackpot, I was tempted to change my blog’s name to “The Whiny Ex-Kid”) but the study isn’t.

  • ChrisA

    There is more than a grain of truth in the study, I mean if you went to a poetry reading, would you not expect to find more liberals there than at the country music festival? (Now I am a libertarian and wouldn’t be caught dead at either…) It reminds me of that old saw about the conservatives being repulsive but right and the left being romantic but wrong.

    In terms of the task of trying not to let your personality influence your policy beliefs, I will take a pragmatic approach because, in almost all cases, policy beliefs can be treated as entertainment as the involvement of most people in changing policy is limited to voting, which has minimal impact. Only if you are the position of being able to change policy do you need to worry about this bias. Perhaps politicians should use the wisdom of crowds to make policy, taking the median position from opinion surveys, unless you have clear information that the general public are not party to.

    Another bias that comes to mind that you have to deal with as a policy maker is the (commonly suggested) one of becoming more conservative as you get older – I assume you would compensate in the same way.

  • Robin Hanson

    Chris, the changing of opinion with age might be explained by increasing info with age, so it is not clear whether to reject that change. If opinions aged due to info, then you would want to change your opinion to the opinion you expect to have when you are old.

  • pdf23ds

    That reminds me of a bias pointed out by Paul Graham. Parents seem to be more conservative about their children than they were with their own lives as a child. This seems to be because, as a child, you get some solid benefits by climbing that big tree, but as a parent, you get none of those benefits and have to deal with all of the problems it causes (like broken limbs).

    Something similar might be going on with old age conservatism.

  • TGGP

    pdf, regarding abortion and policy vs. effect, check out this from Walter Block. If technology reached the level imagined there, I think a lot of people on opposite sides would come together.

  • billswift

    All of you arguing about beliefs vs actual “good” society really need to read Thomas Sowell’s A Conflict of Visions, and his Vision of the Anointed.

  • Indy

    Quite a simplistic view of personality. Let me guess, you’re a liberal. I used to be liberal, but now I’m conservative. So I must be schizo! I just got fed up with the moving moral target of the progressives. Next thing you know they’ll be advocating sex with kids. Oh yes, I read lots of books, am messy, love to act and paint, and believe in God.