|Obama's|| Share ||of Vote|| % of Dem. Voters |
|State||White||Black||Hisp.|| ||Black %||Hisp. %|
The "Simple Average" is the mean not weighted by size of state.
A reader points out that Obama's California exit poll numbers are inflated:
Unless I made a mistake with my calculator, these exit polling numbers for Obama aren't correct, i.e. they don't match his actual vote count, being a few points too high.Perhaps the other states are too?
Normally, they'd be "adjusted" to tie-out with the final results (that included a huge number of mailed ballots), but I'd half suspect that the media was reluctant to lower the initial (inflated) non-black numbers which they'd shown.
Since his Hispanic and Asian numbers are already pretty low, I'll bet his white numbers are the inflated ones. They should probably be about 7 points lower.
Also, keep in mind that exit polls' figures for a demographic group's share of the vote are, by necessity, largely rigged. The polling company guesses ahead of time that Hispanics will cast X % of the vote, so they decide to hire Y Spanish-speaking pollsters and conduct interviews in Z Hispanic-dominant precincts.
For example, I doubt that Hispanic voters outnumber black voters 29% to 6% among Democrats in California. Perhaps Hispanics were less likely to use absentee ballots, which have become hugely popular, and more likely to show up at the polls and thus be exit poll-interviewed?