Wednesday, January 30, 2008

Group Polarization and Credulous Bayesians

Ed Glaeser is a very smart man. I can think of a lot of groups that this paper (PDF) describes quite well, there is a deep ring of truth to it all:


"Unlike perfect Bayesians,Credulous Bayesians treat offered opinions as unbiased and independent and fail to adjust for the information sources and incentives of the opinions that they hear. There are fourproblems here. First, Credulous Bayesians will not adequately correct for the common sources of their neighbors’ opinions, even though common sources ensure that those opinions add little new information. Second, Credulous Bayesians will not adequately correct for the fact that their correspondents may not be a random sample of the population as a whole, even though a non-random sample may have significant biases. Third, Credulous Bayesians will not adequately correct for any tendency that individuals might have to skew their statements towards an expected social norm, even though peer pressure might be affecting public statements of view. Fourth, Credulous Bayesians will not fully compensate for the incentives that will cause some speakers to mislead, even though some speakers will offer biased statements in order to persuade people to engage in action that promotes the speakers’ interests.


[...]


We then turn to the possibility that an individual’s friends and social networks are not a random sample of the population. A group of people might have skewed views on questions of policy or fact, and group members may not sufficiently adjust for that fact. We formalize this possibility by assuming that noise terms in the sample are correlated, rather than independent, as they could be if the group has been selected on some attribute or taste. Credulous Bayesians underestimate the correlation of the signals and act as if their neighbors are a random sample of the population as a whole. In this case, Credulous Bayesianism again causes more extremism and more error. Here too, larger group sizes (so long as they do not produce representativeness) can make decision-making less accurate. For a wide range of parameter values, more correlation decreases accuracy. This is our first result favoring intellectual diversity."

Glaeser's suggestion is that most people are actually pretty easily influenced by the opinions of those around them. This might sound cynical but if you consider just how long it takes to become genuinely informed about an issue from facts alone we do need to accept that most people will access truth socially. That means that when you socialise with those you agree with it is really easy to become convinced of radical ideas; the extreme ideas you here sound entirely normal because everyone around you holds them.

The lesson is clear: keep plenty of people you disagree with within your social circle (both in real life and on the blogs). It'll keep you sane.

2 comments:

Bretwalda Edwin-Higham said...

This might sound cynical but if you consider just how long it takes to become genuinely informed about an issue from facts alone we do need to accept that most people will access truth socially.

People will not follow research through because they're either lazy or they convince themselves there's no time.

Gracchi said...

This is entirely true Matt- a great reason to keep our blogs in communication!