I did not completely anticipate the presidential election of 2008. My guy did win, but the margin of victory was relatively slender nationally, at 54-46%. I thought the contrast in candidate and ticket quality was much greater than that, and I've surmised that Mr. Obama's race may have been an unmentionable factor a "Bradley Effect" that in this case did not turn the election tide.
Racial animus is tough to track by direct means, however, since it has fallen out of fashion, at least publicly. I've suggested to colleagues my theory that 'cognitive dissonance' aroused by the Civil Rights Act of 1964 may have begotten a more accepting populace. Some agree, but others suggest that racial animus continues to simmer just beneath the civilized surface. Few race bigots are sufficiently proud of that heritage to reveal their biases to researchers. So, if it endures, it has been hard to measure.
Enter Seth Davidowitz, an inquisitive, tech-savvy Ph.D. candidate who has published a recent paper in which he claims to have indirectly identified a "significant and robust" racial animus impact on 2008 results, using Google search volume as an indicator (Is there anything Google can't do?). He believes the impact may have cost Mr. Obama three-to-five percentage points in the popular vote, thus narrowing a healthy 58ish-42% margin to half that difference.
In a nutshell, Mr. Davidowitz tracked the Google search volume for the Famous and Fundamental Racial Epithet during the period following the 2004 all-Caucasian election in some two-hundred media markets, comprising 99% of the electorate. He also tried to wean-out use of The Term in other, more neutral contexts, such as rap lyrics searches or among African-Americans. He observes that in those markets where the noted search frequency was high, Mr. Obama did markedly worse than his predecessor Mr. Kerry's totals, as compared to those areas where the search frequency was much lower. For the record, he found that most of West Virginia, upstate New York, rural Illinois and Ohio and parts of the Mississippi delta country ranked highest in effect, whereas largely Hispanic Laredo, TX, Utah and our fair region were among the lowest.
He further dissects his 3-5% drop into components parts, as follows: Democrats who stayed home, non-voters who turned-out for McCain/Palin, and Dems who crossed the voting aisle. He finds that most of the impact was probably about equally divided between the latter two groups.
But what about any countervailing pro-Black vote? Davidowitz found that few white voters fit that description, and that black-voter effects were attenuated by their relatively small numbers and the historical tendency of blacks to vote Democratic, anyway.
The data also revealed a surprisingly high incidence of use of The Term and its plural in Google searches -- roughly as often as words like Hispanic, charity or migraine. The term African American was searched only about three times more often. Davidowitz also notes that, when The Term was used, before and after searches contained a similar racially derogatory tone.
In summary, Davidowitz is satisfied that anonymity whether in the voting booth or while Googling allows folks to express prejudices with which they wouldn't want to associate their names (regular readers of these Town Square Forums may recognize that phenomenon). He also believes that those prejudices are real, enduring and significant in the Big Picture.
So, if true, what do these data portend for 2012? It's always difficult to gauge an incumbent's likely totals when s/he's the only near-certain candidate having an actual alternative choice sharpens the focus considerably. It's also true that this year's crop of Republicans is a remarkably Right-leaning and "imperfect" bunch. Their similarly conservative, primary-voting base may not allow them to move toward The Center in the general campaign, and, in any event, they are unlikely to draw the kind of popular fascination that Obama's uniqueness and personal magnetism spiced into 2008.
At the same time, the Lefties are bitterly disappointed by Mr. Obama's failure to deliver on their dreams of a transformative, post-partisan Progressive, so they may come out only grudgingly, or stay home. His move to the middle has already occurred, but his outcomes have been less than anticipated, especially in the domestic arena. It's not at all clear yet whether next year's margin will be within that 3-5% head start that his opponent, likely a white male, will enjoy.
It seems likely that race still matters in America whether it matters enough to sway the next presidential outcome remains to be determined.