Here at the University of Chicago, we have something called the Chicago Judges Project, by which we tabulate and analyze thousands of votes of judges on federal courts of appeals. One of our key findings thus far is this: In many controversial areas (eg, affirmative action, campaign finance, sex discrimination, disability discrimination, environmental regulation, and more), Republican appointees show especially conservative voting patterns when they’re sitting on 3-judge panels that consist only of Republican appointees. So too for Democratic appointees: They’re far more liberal, in their voting patterns, when sitting with two fellow Democratic appointees, than when sitting on a panel with at least one Republican appointee. In other words, Republican appointees look more conservative when they sit only with fellow Republican appointees, and Democratic appointees look more liberal when they sit only with fellow Democratic appointees.
This is real-world evidence, we think, of group polarization: the process by which like-minded people, engaged in deliberation with one another, typically end up thinking a more extreme version of what they thought before they started to talk. (So, for example, French people who distrust the US distrust the US even more after talking with one another.) Group polarization reflects a form of information aggregation, or at least opinion aggregation, that sometimes leads in unfortunate directions. It’s a big contrast to the price system, Wikipedia, and open source software.
Here’s a related phenomenon: hidden profiles. In a deliberating group, shared information (information held by many or most) usually has a much bigger effect than unshared information (information held by a few or just one). The result is that groups have “hidden profiles,” in the form of information that doesn’t get out, or that has less impact than it deserves. Some big mistakes by private and public organizations (including faculties!) are a result of hidden profiles.
Here again, the price system, Wikipedia, and open source software do a lot better. (Of course there are important differences among the three, as the terrific comments so far suggest, and on which more soon.) All this raises many puzzles. Here’s one: Are there ways to incorporate what we’ve been learning, from those three, so as to make group deliberation go better?
(originally posted 7/17/05)
“This is real-world evidence, we think, of group polarization: the process by which like-minded people, engaged in deliberation with one another, typically end up thinking a more extreme version of what they thought before they started to talk.”
From the description, it’s not clear that one can rule out a confusion of cause and effect. That is, can one distinguish between the explanation above, and the alternative version that people feel free to express a more partisan version of their thinking when among like-minded people. That is, the political equivalent of being in “mixed-company”, versus “single-sex”.
The difference being in assuming that the “mixed-company” version is the baseline thought, and the “single-sex” version is a change. But perhaps the “single-sex” version is the baseline thought, and the “mixed-company” merely moderation of an expression of it.
We need only look at Senate history to see the difference between e.g. how politicians act with a large party majority, versus when there’s no large party majority. But this doesn’t tell us what they really think.
“Here again, the price system, Wikipedia, and open source software do a lot better.”
Umm, “better” as judged how? What was the other part of the comparison?
“Are there ways to incorporate what we’ve been learning, from those three, so as to make group deliberation go better?”
Actually, I think the lessons too many people have been learning from those systems make group deliberations go WORSE!
But the idea with judges, I think, is that there is at least a theory that something “objective” is going on – the understanding of the law. You raise a good point, but if you believe (as I’d like to) that judges aren’t merely “discussing opinons” but are rather “seeking truth” as closely as is possible, then the fact that a different point of view will change the results of their seeking – to something less inclined with how they would tend to react – means that they, in some sense, seem to have more information.
But I do agree, Seth, that the waters are muddled a bit – you do, for instance, have to believe things like the idea that persuasive argument is a useful tool in the discovery of truth.
Seth, the other part of the comparison was apparently systems with more limited knowledge sharing among participants. One common impression of how the Wikipedia encyclopedia works is that it’s a very open process, with lots of shared information and competition to produce the best possible end result (the ideal open source model). That’s true to a large degree in many content areas. It’s far less true in policy matters, where politics as usual and the methods used by political groups everywhere play a substantial and growing role.
Group polarisation seems to be something the blogosphere is particularly good at encouraging. The neutral point of view principle for Wikipedia encyclopedia articles tends to make that somewhat more difficult, though interest groups in specific areas can make it difficult to achieve neutrality or include all relevant material, simply because they will tend to persevere more than typical participants. It’s one area where explicit pro and con adjunts intended to reduce the whitewashing effect might be of use, in addition to the core article.
The University of Chicago study reveals fascinating empirical correlations. It also invokes questions. For example, is there a method as to how federal judges are assigned to cases? Are there any three judge committees with judges from the same ideological background that don’t fall under the information aggregate phenomenon? If so, what is the percentage? Do three judges sharing the same ideological background create higher odds of having a judiciary opinion appealed because of the obvious ideological slant? And so on!
From a jurisprudential point of view, it seems that a perfectly just society or one striving to approximate that end would implement a corrective strategy where the assignment of judges would yield a fair and balanced aggregate of information bearing upon a particular case at hand rather than having one where judges may be swayed by ideological solidarity and showmanship.
the findings presented – exemplified on judges – look to me quite credible, both if the group drives or unblocks the individuals. I dont see though, how the price system should be free of hidden profiles. its simply an illusion that markets are more transparent than anything.
The result is that groups have “hidden profiles,” in the form of information that doesn’t get out, or that has less impact than it deserves. Some big mistakes by private and public organizations (including faculties!) are a result of hidden profiles.
and not so hidden profiles.
Here again, the price system, Wikipedia, and open source software do a lot better.
au contraire, profesor sundstein. on-line communities and efforts are far more the product of groupthink and do a much poorer job at reflecting diversity of opinion than do prices.
for example, in the open market, the overwhelming majority of PCs uses MSFT’s costly windows operating system. (we state this as a fact, not to defend MSFT or invite a flame war.) this MIGHT be interpreted as the general public having a positive view of MSFTs products, but at the very least MSFT market dominance reflects the choices made by a large number of diverse individuals.
log onto to slashdot and say anything positive MSFT and you will be quickly modded down to “flamebait” and subjected to the most vile ad hominen attacks imaginable. the thing is that the slashdot community is itself broad and diverse and very well-informed, but due to its open source roots, the groupthink imposed by the community is incredibly rigid.
in the real world, there are many things other than opinion that bind human beings together. price is a common denominator among many people with disparate views and opinions.
in on-line communities, opinion is often the only thing that binds groups of individuals together. the profiles are not so hidden and the influence is palpable. an analysis of individual wikipedia entries might very well reflect the same bias of opinion found in many on-line communities.
Professor Sunstein,
I know you are visiting Harvard Law next year where I will be a 2L. Is this data from the Chicago Judges Project available to others? If so, where? If not, where might I be able to find out more about the project?
Thanks,
Noah Popp
Professor Sunstein.
Thanks so much for pointing me in the right direction. And of course for commenting on our blog!
Noah