[jhspaybar]

To expand on this a bit, Romney is more likely to take urban republicans in greater numbers over suburban republicans. This is similar to in the general Obama is expected to take urban voters over suburban voters. I suspect this "analysis" could just as easily be done to show Obama "tampered" with votes after the general.

You gave a much better explanation, but in all seriousness folks, flag this story(I already have) and get it off the front page. It's nothing but pseudo statistics from someone who doesn't have a clue hoping to stir controversy over "evil republicans".

[clarkm]

This isn't necessarily true. It's important not to conflate "small precinct" with "suburban/rural area" and "large precinct" with "urban area", since many rural precincts might have more voters than urban ones.

I wish someone would look at the underlying precinct data and redo this analysis while factoring in precinct population density.

[argv_empty]

This isn't necessarily true. It's important not to conflate "small precinct" with "suburban/rural area" and "large precinct" with "urban area", since many rural precincts might have more voters than urban ones.

And in fact, the paper spends quite a bit of time addressing this.