January 20, 2005

Revisiting Election 2004 Exit Polls...Again

Marc Comtois
O.K., I promise not to write about this again for a while (as I've already done here, here, here, and here). At the risk of beating a horse that is, if not dead, then is at least trotting to the glue factory, I think the following should give conservatives here in RI some "Hope." Professor Andrew Busch has written that, contrary to popular belief, the country is not getting more polarized and that
. . . a closer examination of the voting data shows decreased, not increased, polarization. If 2004 had been a really polarizing election, one would expect that Bush's vote percentages would go up in the red states compared with 2000, but that they would go down in the blue states. But this is not what happened. A comparison of the Bush vote in 2000 with his vote in 2004 shows that in the 29 red states, he gained an average of 3.3 percentage points. In the 19 blue states, he gained an average of 3.0 percentage points. (In the three switchers, he gained an average of 1.7%.)

Bush gained big in reliably liberal bastions like Hawaii (+8 percentage points), Rhode Island (+7), Connecticut and New Jersey (+6), New York (+5), and Massachusetts (+4). Altogether, he improved his vote proportion in 48 states—of which only 5 improved by less than 1%. His vote share dipped in only two states, one very blue (Vermont, where he fell from 40.7 to 38.9%) and one very red (South Dakota, from 60.3 to 59.9%). An examination of voters by type of community shows that Bush's biggest gain by far was among big-city dwellers (+13 percentage points), while his suburban and rural support remained stable.
Busch elaborates further that the President significant gains in many traditional liberal/Democrat demographic categories, too. As to his point that the President made big gains in cities, and inasmuch as Rhode Island is often viewed as a political "city-state", I wonder what the voting numbers were in Providence? I suspect that there is still some ground to be made there. (via Powerline)