Bias in Political Conversation

University of Pennsylvania professor Diana Mutz’s book "Hearing the Other Side: Deliberative versus Participatory Democracy looked at survey evidence as to how often people had conversations with others of differing political viewpoints.  In her words:

One logical conjecture would be to expect this form of political behavior to be much like any other.  In other words, it would be disproportionately the province of well-educated, high-income populations.  Indeed, the frequency of general political discussion tracks closely with these characteristics of high socioeconomic status.  But the correlates of cross-cutting conversation are strikingly different.  As shown in Figure 2.3, there are clear patterns of difference with respect to race, income, and education, but they are not in the usual directions.  Nonwhites are significantly more likely to engage in cross-cutting political conversation than whites.  And as income increases, the frequency of disagreeable conversations declines.   Exposure to disagreement is highest among those who have completed less than a high school degree and lowest among those who have attended graduate school.

As sociologist William Weston notes in discussing Mutz’s findings:

I can testify to how easy it is for conversation among academics, the most educated group of people, to turn into a one-position echo chamber. Liberalism is taken to be an IQ test, and the rare conservative is encouraged to be quiet or go elsewhere. For political disagreement I go to the coffee house, which in our town draws a broader range of people than the faculty club contains.

Of course, one explanation would be that what looks like herd behavior and social conformity is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint.  But that’s rather a self-serving explanation, isn’t it?

GD Star Rating
a WordPress rating system
Tagged as:
Trackback URL:
  • http://www.hopeanon.typepad.com Hopefully Anonymous

    I think it’s possible Weston is missing the mark with his commentary (or with his form-fitting this into the crusty liberal/conservative dialectic). I suspect people with less SES status are exposed to in-person disagreement more because they have less control over their environment (who they work with, where they live, where they spend their leisure time). But I think further empirical study will tease these things out one way or the other.

  • http://www.ciphergoth.org/ Paul Crowley

    Of course, one explanation would be that what looks like herd behavior and social conformity is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint. But that’s rather a self-serving explanation, isn’t it?

    It is self-serving, but of course we can’t reject an explanation solely on the grounds of our biases in favour of it, and I’m inclined to think that while no doubt it is not the whole story, it would be strange if this mechanism played no role in explaining what we observe.

  • Ian C.

    “Of course, one explanation would be that what looks like herd behavior and social conformity is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint.”

    In which case these “superior intellects” should be able to provide a proof that liberalism is objectively correct, which they objectively can’t.

  • Xianhang

    To determine which hypothesis, you could look at similar socio-economic groups through time and figure out if the motivations for their agreement.

    Personally, it looks to me like the middle class have a greater opportunity for associative sorting than the lower class. The ability to “customise” your social circle to be congruent to your beliefs is a luxury only a proportion of the population can afford.

  • http://barrkel.blogspot.com/ Barry Kelly

    I personally think it comes down to group identity. The higher up you are on a scale, any scale, and the higher priority you place on that scale’s axis in your life, the more likely you are to vet group membership based on your similarity to people who are also high on the same axis. In other words, people (A) don’t like to spend time with people (B) who disagree with them (A) over things they (A) care about.

    So, if you’re working class, and your family and home area are important to you, or your job is important to you (or church, etc.), the people you interact with will be dominated by people in the same job, same area and same family (etc.) as you. Cross-political interactions should correspond to geographic distribution etc.

    On the other hand, the more educated you are, the more likely you’ll value intellectual stances, and choose to habitually spend more time with people who don’t violate your group’s taboos.

    Objectively correct viewpoints – that’s tragically naive. Fascism / corporatism was born of economic elite settling on what they believed was a correct viewpoint.

  • http://occludedsun.wordpress.com Caledonian

    Objectively correct viewpoints – that’s tragically naive.

    Ironic, considering that you’ve just asserted a view to be objectively correct in the process of decrying such behavior.

    Fascism / corporatism was born of economic elite settling on what they believed was a correct viewpoint.

    No, Fascism was born of political activists resurrecting old concepts about the identity and power of the State.

  • http://www.scottaaronson.com Scott Aaronson

    As a test case, I expect that disagreement about (e.g.) evolution versus creationism, or the reality of the moon landings and the 9/11 attacks, is also smaller among the better-educated. Is that also evidence of social conformity and herd mentality?

    I suspect a key point is that educated people do argue constantly about politics, but their spectrum of opinions is very different than the general population’s. And therefore, if you try to measure disagreement among the educated using the general population’s opinions as a baseline, then it’s going to look, misleadingly, like the educated people all agree with each other. While in reality, you’ve got fierce debates between the socialists and the communists, the Democrats and the Naderites, the libertarians and everyone else… :-)

    Can anyone familiar with Mutz’s book comment on how or whether she deals with this?

  • Sociology Graduate Student

    Very good point Scott. This needs to be broken down by specific opinion.

    I do think that being correct will explain a small share of the finding on some issues.

    Hopefully Anon is also right that more educated people are more able and probably more motivated to select into like minded groups. Perhaps it is partially due to the fact that more educated people are more aware of and/or concerned about ideological consistency. Therefore, when they disagree with someone they will be more likely to (rationally) infer that the other person disagrees with them about many/most things.

  • http://profile.typekey.com/jhertzli/ Joseph Hertzlinger

    As a test case, I expect that disagreement about (e.g.) evolution versus creationism, or the reality of the moon landings and the 9/11 attacks, is also smaller among the better-educated. Is that also evidence of social conformity and herd mentality?

    According to a recent post at Crooked Timber, educated global warming “deniers” tend to more extreme.

  • James Andrix

    is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint. But that’s rather a self-serving explanation, isn’t it?

    I don’t think it fits the fact either. They don’t all agree, they are just close enough that they don’t want stigmatize each other for it. There are probably dozens of differences in theory and policy that two ‘agreeing’ liberals can hold. (or conservatives, I don’t think it matters.) in a group, only one or zero of them could have found the ‘correct conclusions’. The rest are stumbling.

  • Ned

    “Of course, one explanation would be that what looks like herd behavior and social conformity is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint.”

    No, we can reject the explanation that liberal (or conservative, for that matter) opinions are generally correct. Here’s why (unfortunately, this was not my original idea): opinions on different subjects are correlated (for example, you can pretty much predict someone’s opinion on gun control if you know his/her stance on abortion, although the two questions are not related). Now, if one worldview (for example, liberal) is based on more accurate judgment, you would not observe this pattern (grouping of opinions across various issues into two clusters – liberal and conservative). Instead, you would see one cluster (the correct opinion holders), and the rest of data points would be scattered, as individuals make random errors in judgment on various topics.

  • http://metaandmeta.typepad.com/ Q the Enchanter

    Well, it’s only as self-serving as the declaration that the liberal consensus among academics “looks like herd behavior.” (Indeed, can anyone think of any explanation for the phenomenon that wouldn’t be self-serving to someone?)

  • William Newman

    I agree with Scott Aaronsen that groupthink, partisan wishful thinking, or outright dishonesty are not the only explanations for conformity, and that an important alternative explanation is that many opinions are eliminated by requiring technical rigor. But I have seen some patterns that seem to resist that alternative explanation.

    If academe was merely selecting for technical rigor, wouldn’t we expect embarrassingly-dumb-in-hindsight mistakes to have no particular left/right bias? Are there goofs by academic rightwingers comparable to Bellesiles’ _Arming America_ award or the _Social Text_ hoax?

    And did all those communists end up in sociology as a side effect of technically rigorous selection for objective excellence?

    Also, anecdotally from Cornell ca. 1989, both my two memorable examples of silly grad student doublethink enabled by groupthink were distinctly left-wing.

    A physicist argued first that SDI is so technically impractical that we should feel contempt for those few marginal technical people who claimed it might work, and also (lots of time in long walk up hill…) that it was not just a waste of money but very dangerous because the Soviets would rationally launch a preemptive attack. Sophisticated arguments, perhaps, and I was something of an SDI skeptic myself, but still I had to point out the arguments don’t work well together. Surely the Soviets must have a number of technically informed decisionmakers if their missiles and bombs work. I was surprised to find that he was surprised — all the wailing and gnashing of teeth in the physics department, and no one had pointed out that two popular arguments don’t play well together?

    And a philosopher crushed my vulgar objections about some question of interpretation of the US Constitution by pointing out that ultimately documents have no objective meaning. Fine, end of discussion, I will go on vulgarly messing with computer files and information theory and you can think about, um, whatever, and we can play Go. (The Aumannians here have harsh things to say about those of us who agree to disagree, but sometimes what can you do?) But then some weeks later he waxed indignant about the Reagan administration violating the proper interpretation of an arms control treaty. Then having agreed to disagree or not, I had to point out that the two positions are hardly compatible. “Documents have no objective meaning” sounds impressive and is hard to refute. “Documents inconvenient to the left have no objective meaning” sounds less impressive. And again, this point was clearly a novelty to him. I took that as evidence that in the sophisticated philosophical circles where he had learned the “no objective meaning” principle, the unspoken rule about when to apply the principle must be universally respected.

  • josh

    Maybe you don’t get to the top by picking fights.

    I wonder what happens if you account for the internet.

  • http://occludedsun.wordpress.com Caledonian

    Is that also evidence of social conformity and herd mentality?

    If there are powerful, logical arguments and extensive, overwhelming evidence in favor of a position, we need look no further to explain why most people hold a position. If there are equally compelling points against a position, that seriously narrows down the reasons why a social group might adopt it regardless.

    Politics is rarely clear-cut, and people’s arguments for their beliefs virtually never are. There are different goal sets, different methods, and different beliefs about methods. Add rhetoric and diplomacy into the mix, which involve people not speaking their beliefs but saying what they believe will bring about the desired result, and you get modern political discourse.

    And a philosopher crushed my vulgar objections about some question of interpretation of the US Constitution by pointing out that ultimately documents have no objective meaning.

    And the conventions that link symbols with numbers are ultimately arbitrary. That doesn’t mean that 2+2=5.

    To paraphrase George Orwell, that person must have been a philosopher to make such an argument: no ordinary person would be such a fool.

  • http://hanson.gmu.edu Robin Hanson

    Stuart didn’t notice that I’d already posted this result last November, and I wanted to see if anyone would notice.

  • ad

    Caledonian, if memory serves, Orwell remarked in the same essay that a man who wanted to predict the course of the Second World War would have been better served by talking to factory workers than the intelligencia (based on his own experience).

    More intelligent people might be more likely to come to the correct conclusion than the less intelligent – but only if they are genuinely trying to find the correct conclusion. If they are simply trying to keep in with their friends, they merely have the ability to string plausible sounding arguments together to support greater idiocies.

    Of course, one explanation would be that what looks like herd behavior and social conformity is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint.

    It occurs to me that successful businessmen are likely to be smarter than the average person, and they are more likely to be conservatives. They also have much more experience of actually running organisations than most academics. So perhaps it is they who have settled onto the objectively correct viewpoint?

  • Leonid

    “I can testify to how easy it is for conversation among academics, the most educated group of people, to turn into a one-position echo chamber.”

    This is not necessary true of all countries.

    In the US the major dividing issues in politics are religion (which usually predetermines the attitude on abortion, gay marriage, school prayer etc) and taxes.

    Scientists very rarely follow traditional religions and, being state employees, they are more favorable to increased government spending.

    In some other countries the major dividing issues run across the scientific community, resulting in much more diverse spectrum of opinion.

    Another factor which reinforces the conformity in the US academy is a very high degree of separation between academics and the rest of the country.

  • Robert

    “Of course, one explanation would be that what looks like herd behavior and social conformity is really just what happens when a bunch of superior intellects independently settle on the objectively correct viewpoint.”

    Even if liberalism cannot be proven objectively correct, smarter and better educated folks would be more likely to come to similar, objectively correct, conclusions about things that ultimately affect worldview. For instance, this demographic might come to the objectively correct opinion about religion and religious documents, such as the bible, which would allow them to look at issues like abortion, environmentalism, gay rights, etc. from a different angle. If rationalism forces people to eliminate certain types of arguments, it makes sense that smarter and better educated folks might tend to come to the same conclusions on policy issues.

  • http://www.scottaaronson.com Scott Aaronson

    No, we can reject the explanation that liberal (or conservative, for that matter) opinions are generally correct. Here’s why (unfortunately, this was not my original idea): opinions on different subjects are correlated (for example, you can pretty much predict someone’s opinion on gun control if you know his/her stance on abortion, although the two questions are not related). Now, if one worldview (for example, liberal) is based on more accurate judgment, you would not observe this pattern (grouping of opinions across various issues into two clusters – liberal and conservative). Instead, you would see one cluster (the correct opinion holders), and the rest of data points would be scattered, as individuals make random errors in judgment on various topics.

    Interesting argument, but I don’t think the conclusion follows. It might be, for example, that those with right-wing positions on both abortion and gun control define themselves largely by their hatred of “intellectual elitists,” and therefore adopt opposite opinions from academics just out of spite. Or it might be that they have one large wrong belief (possibly religious in nature, possibly unknown) from which all the other smaller wrong beliefs flow as consequences. I’m not actually advocating either theory, just pointing out that there are all sorts of reasons why seemingly-unrelated variables could be correlated.

    Let me even make a further claim: if I didn’t know what the liberal vs. conservative positions were on (say) gun control, I’m pretty sure I’d be able to predict them, given my knowledge of the liberal vs. conservative positions on lots of “unrelated” issues like abortion and gay marriage. And this is so, even though I might not be able to articulate a convincing general rule that let me make such a prediction. But this predictability, if true, is manifestly incompatible with “liberalism” and “conservatism” being two arbitrary tribal allegiances whose positions on various issues amount to independent coin tosses.

  • Constant

    if I didn’t know what the liberal vs. conservative positions were on (say) gun control, I’m pretty sure I’d be able to predict them, given my knowledge of the liberal vs. conservative positions on lots of “unrelated” issues like abortion and gay marriage.

    It’s pretty hard to be sure you’re not suffering from hindsight bias, especially if, as you say, you wouldn’t be able to articulate a reason for your view.

  • ad

    I can testify to how easy it is for conversation among academics, the most educated group of people, to turn into a one-position echo chamber.

    I think that is mostly worrying for reasons that have nothing to do with politics. If academics are unwilling to criticise each others ideas, whether for reasons of academic politics or any other reason, what use are they?

  • Caledonian

    If rationalism forces people to eliminate certain types of arguments, it makes sense that smarter and better educated folks might tend to come to the same conclusions on policy issues.

    No, because “smarter and better-educated” != “rational”.

    I don’t think most people appreciate that what is generally meant by intelligence has nothing to do with the quality of decision-making. Intelligent people create more elaborate and intricate justifications for their beliefs. It takes distinctly different qualities to sort sense from nonsense.

  • Leonid

    Scott: “if I didn’t know what the liberal vs. conservative positions were on (say) gun control, I’m pretty sure I’d be able to predict them, given my knowledge of the liberal vs. conservative positions on lots of “unrelated” issues like abortion and gay marriage.”

    I doubt that. Let’s take the issue of gay rights for instance. Based on the present situation, one may take the support of the left and the opposition of the right for granted. Yet things were quite different in the past. In early 1930s, homosexuals held many prominent positions in NSDAP and, if Roehm had more luck, homosexuality might have been legalized. On the other hand, many ideologues of the left (including Karl Marx, Erich Fromm and Jean-Paul Sartre) considered homosexuality to be a bourgeois vice, sometimes even directly linking it to fascism. Later many left-wing movements found it politically convenient to reverse their attitudes (ironically when some communist parties in the Western Europe turned to advocating gay rights their financial sponsor in the East was still sending homosexuals to the Siberian prison camps).

    Naturally it may be theoretically possible to predict such reversals, but this seems to be not a trivial task.

  • http://entitledtoanopinion.wordpress.com/ TGGP

    Jeffey Friedman summarizes Philip Converse on such issues here. Bryan Caplan gives a preview of that issue of Critical Review here.

  • Romeo Stevens

    liberalism doesn’t recognize property rights. without recognized property rights there will be constant conflict over whom should get what.
    liberalism supports enforced socialized welfare. enforced socialized welfare creates a disincentive to be competitive.
    liberalism supports collectivism. ‘majority makes right’ is just a different type of ‘might makes right’.
    People see the U.S. constitution as a rock solid foundation for society. It is not, even as originally envisioned, the constitution is a precarious balance between the rights of the individual and the collectivist safety of the state. Since then we have had 232 years of supreme courts reinterpreting what the constitution means, almost always in favor of more governmental power.

    The founding fathers would have been horrified by the intrusiveness of taxes, the existence of a permanent army, our needless involvement as ‘world police’, the passing of responsibility for minting our money from congress to the semi-private Fed, the blatant market manipulation resulting from the unification of the financial institutions (and government complicity in such), our massive contributions to corrupt NGO’s such as the world bank, the gutting of state sovereignty, the legal bribery of politicians (the loopholes could be easily closed) etc.

    we’re in a sorry state and I don’t see things getting better except in places like singapore, hong kong, dubai, and other capitalist city-states. they’re profitable enough to waste billions on ridiculous architectural challenges, have high standard of living and low crime rates. they must be doing something right.

  • http://occludedsun.wordpress.com Caledonian

    Naturally it may be theoretically possible to predict such reversals, but this seems to be not a trivial task.

    The fact is made more complex by the reality that people often find it convenient to conceal their intentions and motivations, and so describe their positions using names and terms that are grossly inaccurate.

    Conservatism, as a political movement, has little to do with conservatism as a concept. A similar point holds for liberalism – what is formally called ‘classical liberalism’ is incompatible with Liberalism as a political identity today, and is generally considered a form of conservatism.

  • http://www.hopeanon.typepad.com Hopefully Anonymous

    With expressions of belief like Romeo Stevens, I see the fun in Cowen’s suggestion that people assign probabilities (of truth/accuracy) to their beliefs.

  • http://www.hopeanon.typepad.com Hopefully Anonymous

    “I don’t think most people appreciate that what is generally meant by intelligence has nothing to do with the quality of decision-making. Intelligent people create more elaborate and intricate justifications for their beliefs. It takes distinctly different qualities to sort sense from nonsense.”

    Caledonian, I disagree. I think intelligent people (of the only type of intelligence that matters IMO, the ability to maximize persistence odds) are those that sense from nonsense. This is different than (1) creating elaborate and intricate external propaganda to get others to act in the intelligent person’s interest, or (2) creating and internalizing personal belief ideas and narratives that on their surface are irrational, but optimize management of system 1 “subconscious” intelligence to meet the rationally optimized goals devised by the system 2 intelligence.

  • Robert

    Caledonian, Even if “smarter and better-educated” != “rational”, I think that smarter and better-educated should tend towards rationalism. I think this because the educational ideal (at least in the math,science,humanities university environment) is to cultivate rational thinking, and presumably, more intelligent people can better acquire and use the techniques education develops.

    You are right that intelligent folks can create intricate justifications, but it isn’t clear to me that this ability means they cannot distinguish rationalizations from rationalism.

  • http://youtube.com/leearnold Lee A. Arnold

    I guessed Mutz’ results long ago! In general there is more open and friendly talk across varying viewpoints among today’s bluecollars than among high-income well-educated. I circulate among both. It should be pointed out that “well-educated” does not mean broadly-educated, in our time. In fact “well-educated” is a bit of a misnomer: it should be “well-specialized” or something. Some of these people you couldn’t trust to take out the trash, because they would put it in the wrong dumpster, or get into an argument with the garbageman. Since Kuhnian paradigms change with time, since for example market economics was presumed to be the default organizing position for society even though none of its theoretical justifications could be calculated further than two-person exchanges, while the presumption of self-interest was a objectionable bias obvious to Adam Smith, Alfred Marshall, and your grandmother and is regularly violated in new experimental studies, it is hard to claim that “superior intellects independently settle on the objectively correct viewpoint.” I realize you were joking. My own friendships extend from extreme left to extreme right; I doubt I could get some of them in the same room together. I think the reason that liberals dominate in universities, (if they really do, I doubt whether it’s true in the physical sciences and engineering,) is because there is a long-term thing lasting decades which could be called “intellectual fashion,” and it gets buttressed by self-reinforcement, hiring practices, and the like. Of course modern conservatives didn’t help their own cause much: they stood up with leaders who were on the wrong side of civil rights, womens’ rights and gay rights, they tend to think that poor people get what they deserve (often relying upon market economics as the justification,) and their recent heroes have been Ronald Reagan and George W. Bush, two uneducated men who increased the debt, shifted the distribution of the tax burden downward, and deregulated health, safety, and environmental issues to no good ends. God knows the modern liberals can be insufferable, but what’s the alternative? A little intellectual humility would first be called for.

  • Shmuel

    “when a bunch of superior intellects independently settle on the objectively correct viewpoint”

    You’ve obviously never read the comments at the Huffington Post or the Daily Kos.

  • http://stuartbuck.blogspot.com Stuart Buck

    Robin did post about this same finding last November, but readers’ long-term memory will be best activated if periodically reminded of the same information. (OK, I forgot.)

    Q says, “it’s only as self-serving as the declaration that the liberal consensus among academics ‘looks like herd behavior.'”

    Maybe I should clarify: as to issues where something like an objective answer exists, the fact that lots of intelligent people believe the same thing isn’t necessarily herd behavior (although it well could be). But many political issues boil down to subjective weightings of incommensurable values, where no objective solution is in the cards (e.g., the right of the fetus to live vs. the right of the woman to control her body, or the intangible value of protecting certain endangered species vs. the value of economic development). When you find a group of people who all take one side of such an issue, I don’t think it’s “self-serving” to suspect herd behavior of some kind.

  • http://occludedsun.wordpress.com Caledonian

    Caledonian, Even if “smarter and better-educated” != “rational”, I think that smarter and better-educated should tend towards rationalism.

    Then I doubt that you appreciate what the function of higher education is. It might be helpful for you to review the origins of the university system.

    I also suspect that you’re using ‘smarter’ in an informal and vernacular sense. The more specific meanings of intelligence do not result in any clear trend towards rationalism.

  • Mark Nelson

    See Inglehart, Welzel in Modernization, Cultural Change, and Democracy. They provide solid data from the World Values Survey documenting how fluctuations in human autonomy are strongly predictive of prevailing human values systems. More autonomy, greater desire for choice, self-expression, individualism. Less autonomy, greater reliance on religion or authoritarian types of government.

  • Roger Sweeny

    Lee A. Arnold,

    I think you are largely right but I would be willing to bet my life that between the beginning and end of the Reagan administration, all 6 of these went up: 1)federal health regulation, 2)non-federal health regulation, 3)federal safety regulation, 4)non-federal safety regulation, 5)federal environmental regulation, 6)non-federal environmental regulation.

    Same for George W. Bush. Twelve deltas, all positive.

    Of course modern conservatives didn’t help their own cause much: they stood up with leaders who were on the wrong side of civil rights, womens’ rights and gay rights…

    It is interesting to compare the attitude of most academics toward conservatives and toward libertarians. Libertarians never stood up with those leaders but they often get lumped in as not being worth taking seriously.

  • T.H.B.

    Uh…now what was the question, again?