First off, as I write this the AP has declared Virginia for Webb—giving control of the Senate to the Democrats.
Okay time for three Senate races…
For a Democratic stronghold I choose our quirky state of Wisconsin. The candidates were Kohl (D) 67%, Lorge (R) 30%, Vogeler (G) 2%, and Glatzel (I) 1%.
What is interesting is that the Republican candidate won absolutely no counties. As a Wisconsinite, I know there is great like for Kohl, but I thought that Waukesha and/or Washington counties surely would have voted Republican since they are conservative counties. Kohl, though scrapped by with 51% and 50%, respectively. As for the Gaps, well, their a bit irrelevant here. In the gender gap, males voted for Kohl 65% and women voted 71%. They have a six point difference, but both overwhelming voted Democrat. As for an ethnic gap, Wisconsin’s minorities don’t add up to any significant percentage of the vote (or maybe in this case the exit poll); however, if minorities were in larger numbers I would bet that they would have voted for Kohl too. The age gap is almost as insignificant as the ethnic gap. There wasn’t a single age group that voted less than 65% for Kohl. The only real difference was between the 30-44 and the 60+ group, but that has only a five point difference (65% and 70%). As for the income gap, their was a slight trend that the more money you made the less you voted for Kohl, but again his numbers were still high among all groups. The biggest difference was between those who made under $15K (84%) and those who made between $100k and $150K (58%). The education gap was insignificant, with every level supporting Kohl with at least 60%. The biggest difference was between Postgraduate education (73%) and College education(60%). In this case, only those with just a college education voted less for Kohl, but still overwhelmingly. Finally, the religion gap. Every religious group voted at least 60% for Kohl. The only significant difference is between those who attend church weekly and everyone else, with 57% to 74% (for the occasional attendees) to 82% (for the those who never attend). The religious gap is understood poorly to make and conclusions though. It is possible that the Republican candidate was more “Christian”, or it is possible that those who attended church weekly or more where told more often to vote (indirectly, hopefully) Republican.
Overall, the gaps had no impact on the election of Kohl. He has developed somewhat a cult of personality that spans all groups. He could run as an independent and still win; however, I think he would not want to face both a Republican and Democratic challenger. On the political spectrum Kohl fits along the center, gathering votes from both sides and consistently gathering votes on both sides.
For the Republican stronghold, I choose the senate race in Texas to examine. The candidates were Hutchison (R) 62%, Radnofsky (D) 36%, and Jameson 2%.
In this race, the Democrat was able to pick up some counties. The gaps are more pronounced in the race. For the ethnic gap, whites voted for Hutchison 68% where African-Americans voted 26% and Latinos 44%. As for the gender gap, it is insignificant with only a two point difference. The age difference has only a seven point difference, those who are 60+ (66%) and those 18-29 (51%). The income gap is big with steady climb between $15K-$30K (44%) and $150K-$200K (71%). The education gap isn’t that significant with those with no college education voting 58% for Hutchison and those with 63%. Finally, I would do the religion gap, but CNN did not post any religious polling data. I can only assume that it didn’t matter what religion you were or if you went to church, you voted for Hutchison. Of course there is the bigger probability/stereotype that those in Texas are more Christian and go to church more. I do find it odd that there isn’t a poll for different sects. For the electorate, only 16% of them where liberals, and the other percentage was divided between moderates (51% voted for) and conservatives (84% voted for). Hutchison probably didn’t have to go as center as Kohl to win his election.
Finally for the switcheroo, Montana! The candidates were: Tester (D) 49%, Burns (R) 48%, Jones (I) 3%.
As expected by a race so close, each had a mixture of counties. The most interesting thing about this race is that just part of the independent bloc could have swung the election either way. One’s vote may not count much, but if you get a group together it matters a lot. For the gender gap, their was only a four point difference between women voting for Tester (52%) and men (48%). The ethnic gap is big (however it is only 8% of the vote), with the category marked “other” (I’m assuming it must be Native Americans since it’s Montana) voting for Tester 61% and whites voting 49%. As far as the age gap, there’s not too much difference. The biggest difference is between 18-29 (56%) and 30-44 (45%); however, it goes up to 53% in the 45-59 age group. The income gap hovered around a ten point difference bouncing between the income levels, of course the lowest income voted the most for Tester. For the education gap, a somewhat proper linear relation forms between the higher the education the more one voted for Tester. The biggest difference being between post-grads (68%) and no high school (40%). As for the religious gap, there were significant differences between those who attended often and those who didn’t (33%-67%, respectively).
For the three races, the gaps didn’t show much. Occasionally a traditional gap did show up, but explaining the gap is the work for psychologists or sociologists. I would like to know why Texas didn’t have any postgraduates in its survey. I find it hard to believe that postgraduates are rarer in Texas than in Montana, especially with one of the world’s few particle accelerators located there in a notable university, Texas A&M. I don’t think Texas is suffering from brain drain, or at least I haven’t heard about it. So then Texan postgraduates: don’t exist, don’t vote, don’t participate in exit polls, or don’t vote in precincts with exit pollsters. This just makes me think that the poll must be flawed in other ways, such as disregarding any other ethnicity other than white in Wisconsin. I know we are not a very diverse state; however, I believe there would be enough for at least a few significant percentage points. I feel that this blog has been a waste of my time, that analyzing media exit polls is worthless. I would rather study (if they took it) the NES results of this election. The only thing I found in analyzing the polls was a common thread of moderates deciding the race. As pundits have been saying, moderates voted this time and choose to reject the Republicans for being too conservative.
Sources:
WI:
http://www.cnn.com/ELECTION/2006/pages/results/states/WI/S/01/index.htmlTX:
http://www.cnn.com/ELECTION/2006/pages/results/states/TX/S/01/index.htmlMT:
http://www.cnn.com/ELECTION/2006/pages/results/states/MT/S/01/index.html