Polarization

[madspark]

Polarization is bad, m'kay?  Unless, of course, you _want_ a split...

http://www.orgnet.com/leftright.html

http://www.orgnet.com/divided.html

Fascinating!  And not terribly surprising...

Comments

[oliana0]

I dunno. I think that using Amazon's people who purchased this also purchased that to guage an understanding of who's reading what may be a bit skewed.

For example, just because Amazon says that people who bought A also bought B doesn't mean they didn't also buy X, Y, and Z, it could just be less frequent.

Also, do we know that Amanzon doesn't also skew the selection by the type of book? Is Amazon more likely to include a book about subject A on another book about subject A's page, or a book about subject B, even if they are bought by similarly sized groups of people?

[madspark]

It's not definitive by any means, but the data is tough to come by so we get to peer at what is available.

Also, Amazon itself will tend to focus communities "If you bought this, you will like that", creating additional overlap.

Of course there are a zillion external factors that affect what is read and where it is bought, but in a large sample (and Amazon is nothing if not large) these also smooth out.

Also, the very definite wasp-waist is interesting, and not a given at all; also, this narrow separation was there in both the 2003 and 2008 samples.

If nothing else, it points to an area of valuable study.

And, also, when was the last time any of us read any Anne Coulter? Or something else from across the philosophical aisle?

I have no visceral reason to doubt these results.

[oliana0]

Anne Coulter burns my eyes. I probably wouldn't mind ready something clever, witty and true from a conservative perspective, but that's not her.

Any suggestions?