FWIW I didn't downmod you. Both high and low levels of education correlate with voting Democrat. And higher income generally correlates with voting Republican.

FWIW I didn't downmod you. Both high and low levels of education correlate with voting Democrat. And higher income generally correlates with voting Republican. The tricky part is that higher education also correlates with higher income. I believe I saw the data disaggregated in a post on Gelman's blog. Will post in an edit if I can find it. EDIT: www.historycentral.com/elections/12008/exit/Income.html [sound] http://mapscroll.blogspot.com/2009/03/income-and-2008-us-election.html http://www.fivethirtyeight.com/2009/03/rich-and-poor-voters-in-2008.html The second link is the same as the third link from Gelman's writeup (syndicated by Nate Silver), except the image links in the second link aren't broken. The data shown indicates that Republicans tend to be on average wealthier than Democrats, until you get to the ultra high income levels. If you think about it, this does start to suggest that there are aggregation issues going on; in general the rich pay more in taxes than the poor, and the middle class and rich tend to vote Republican, so something counterintuitive is going on if "red states" are net tax recipients. I can't seem to find the original post that did the analysis, but hopefully it seems more plausible now; if you really want to explore the issue you can get into the PUMS sample: http://www.census.gov/acs/www/data_documentation/data_main http://www.census.gov/acs/www/data_documentation/data_main