Someone left a lengthy response to yesterday's post on literacy. My reply got too long to wedge into the comment box, so I decided to make it an independent post. You might want to read yesterday's comment for context first.
The overwhelming majority of the country was purple, not red or blue, in the 2004 election. Which is to be expected with only a 3% popular vote margin nationwide.
The literacy study didn't break things down to the level of individual publications, so there's no way to tell (from this data) whether people are reading The Nation or The National Review. Presumably a bit of both. Access to the Internet in particular means access to a vast array of opposing views, which can only help informed debate.
Even a bad New York Times article contains more facts than a good Fox News (or CNN, or CBS, or pick your favorite TV news) story. That's simply the nature of the medium. You can fit far more information into 3000 words than you can into three minutes. (Try it. Read the front page of any newspaper aloud for three minutes and see how far you get.) As for bias, well, the most recent major embarrassment for the Times involved Judith Miller's blatantly pro-administration reporting.
The statistics correlating social ills with voting are actually quite interesting. For example, Massachusetts, a blue state if there ever was one, has lower teen pregnancy, divorce, and crime rates than ultra-red Texas. That probably has more to do with other social variables than with politics, but provides at least one counter example to the claim that pro-Bush areas have fewer social ills. (Yes, some of the statistics quoted are a little old. Anyone who can find links to newer data is welcome to post them.)