Shifts in demographics of party affiliation
Before the most recent shift in party affiliation household income had been one of the most predictive indicators of party affiliation. Today, frequency of church attendance is a better predictor.

See CNN 2000 exit polls here:
http://www.cnn.com/ELECTION/2000/epolls/US/P000.html

In a recent Oregon poll there was almost no correlation between candidate preference, and income.
See
2004 polling data from Oregon here.
(from http://osrl.uoregon.edu/press/report/2004PresElection.pdf )

It is also interesting to look at party affiliation by education. In the 2000 exit polls it was clear that being at the lower or upper ends of the educational spectrum made one more likely to vote Democrat, and being in the middle of the educational spectrum was more likely to yield a Republican vote. However, when income was controlled for, higher education tended to produce more votes for the Democrat. That is, if you look at all people making between say $50,000 and $60,000 then within that group, the more educated they are the more likely they are to vote Democrat.

The data from the Oregon poll shows a different pattern, however. Oregon does not have a large Black population, (1.6% vs. 12.3% nationwide
http://quickfacts.census.gov/qfd/states/41000.html )
In recent elections the Black vote has been 90% Democrat. The Black vote is also on average less formally educated than the general vote, so it accounts for some of the Democrat’s popularity with lower-education voters nation-wide. 

With the black population almost absent, the Oregon data shows a clear trend of more education leading to greater affiliation with the Democratic candidate.

<HS                    59% Bush , 21% Kerry
HS                      58% Bush , 39% Kerry
Some College       46% Bush , 41% Kerry
College                46% Bush , 47% Kerry
Post-grad             26% Bush , 69% Kerry

This trend intensified in the 2004 election. Bush gained 10% more of the non-high school graduates in 2004 than he had in 2000. He lost those with the most education, post-grads, by 11% both times.

From:
http://www.cnn.com/ELECTION/2004/pages/results/states/US/P/00/epolls.0.html

                            
                      % total of vote       Bush(gain from 2000)            Kerry
No High School    (4%)               49% (+10)                             50%

H.S. Graduate      (22%)              52% (+3)                               47% 

Some College       (32%)              54% (+3)                               46% 

College Graduate  (26%)              52% (+1)                               46% 

Postgrad Study    (16%)              44% (+0)                               55%


Another interesting statistic is the extent to which the ranks of university professors are dominated by the Democratic party. From:
http://www.nypost.com/postopinion/opedcolumnists/35147.htm

One study of 1,000 professors finds that Democrats outnumber Republicans at least seven to one in the humanities and social sciences. That imbalance, more than double what it was three decades ago, is intensifying because younger professors are more uniformly liberal than the older cohort that is retiring.

Another study, of voter-registration records, including those of professors in engineering and the hard sciences, found nine Democrats for every Republican at Berkeley and Stanford. Among younger profs, there were 183 Democrats, six Republicans.

The American Enterprise magazine reported in 2002 of examinations of voting records in various college communities. Some findings about professors registered with the two major parties or with liberal or conservative minor parties:
Cornell: 166 liberals, 6 conservatives.
Stanford: 151 liberals, 17 conservatives.
Colorado: 116 liberals, 5 conservatives.
Back to Vermont page