How Race Mattered on Nov. 4, and Why the National Exit Poll Is Wrong

SAVE AS FAVORITE

VIEW FAVORITES

By John Wilson (about the author) Page 1 of 2 page(s)

opednews.com Permalink

For OpEdNews: John Wilson - Writer

Most analysts have downplayed the role of race in the Nov. 4 election, largely based on reports about the national exit poll. However, I've discovered that there's a serious statistical error in the national exit poll. It cannot be correct, if you look at the separate state exit polls.

The common view about race in this election was given by ABC News:

Race was perhaps a surprisingly minor factor. Nineteen percent of all voters called it at least somewhat of a factor in their vote; 80 percent said it was not a factor at all. But Obama's margin was essentially the same among those who called race a factor – 53-45 percent – and those who said it was not, 51-46 percent. There were, however, differences by race. Seventeen percent of whites called race a factor, and favored McCain by 61-37 percent. Whites for whom race was not a factor voted for McCain by a narrower 53-44 percent. Meanwhile one in three blacks called race a factor in their vote, and, like all blacks, favored Obama almost unanimously.

I've put together a spreadsheet listing the exit poll data on racial views for each state. It is statistically impossible for the 9% and 19% figures to be correct. In fact, out of 36 states with exit poll data on the race-based voting question, only California (9% significant, 16% a factor) fell below the reported national figure of 9% significant and 19% a factor, and it only did so by a bare margin.

The states without exit poll data on the influence of race are Alabama, Connecticut, Idaho, Kansas, Mississippi, Nebraska, North Dakota, Oklahoma, Rhode Island, South Dakota, Texas, Utah, Vermont, Wyoming, and District of Columbia. (If anyone can find this data, please let me know; I've checked MSNBC, CNN, and Fox News with the same lack of results.)
Considering that the remaining states include high-population states with high levels of racism such as Alabama, Mississippi, and Texas, it's inconceivably that these missing states could lower these numbers. In fact, the correct national numbers on race would almost certainly be much higher than what I report below.

To get an accurate analysis of race, I calculated the influence of race, and then adjusted it for the population of each state in the 36 states where we have results. Assuming that I didn't screw up, I found that nationwide, 13.24% of voters said race was an important factor (including the small number who said it was the single most important factor), and 21.4% of voters said race was a factor (adding those who called it a minor factor).

Now, it is true, as the reports said, that 53% of these votes where race mattered were for Obama, a contrast with the primaries where race-based voting hurt Obama much more. However, it is mistake to imagine that these votes indicate that race was a positive factor for Obama. The overwhelming number of votes for Obama where race mattered came from African-Americans, who vote for Democrats anyway by a large margin. Obama slightly increased the Democratic percentage of the African-American votes, and slightly increased African-American turnout. But for the most part, the people voting for Obama who said that race mattered were almost certain to vote for Democrats anyway. And it is much easier to admit that race mattered and say that you voted for the black candidate than. It is far less clear that people are willing to admit racism.

I measured the "racist vote" (defined as those who said race was a factor and who voted against the black candidate). I should note that this definition represents the bottom-level measure of racism. For example, there are an unknown number of people who voted against Obama because of race and refused to admit it to the pollsters. There are also probably a small number of racists who considered Obama's race as a negative factor but voted for him nevertheless due to other factors (for example, in nearly all-white West Virginia, the proportion of people who said race mattered but who voted for Obama grew from 9% in the primary to 39% in the general election). I know there are some who would say that voters (of any race) who supported Obama and said race was a factor are also racists. I disagree. That's like saying the NAACP is the same as the KKK because they both are concerned about race. It's not racism to support electing the first African-American president. It is racism to oppose doing so based on race.

The average state was 8.07% racist (as an important factor) and 12.62% racist (as any factor). Because the larger states tended to be less racist, the nationwide average of these 36 states was 7.13% racist as an important factor and 11.5% racist as any factor. (By contrast, according to the inaccurate numbers reported by the national exit poll, only 4.23% of voters were racists who considered race an important factor, and 8.93% voters were racists who said race was a factor at any level. So the levels of racism on Nov. 4 were 69% and 29% higher than what's currently being reported.)

The most racist state in these polls was Louisiana, where 14.84% of voters against Obama felt race was an important factor, while 23.78% of anti-Obama votes said race was a factor in their vote. Closely following Louisiana were Alaska (14.0% and 21.08%) and Kentucky (14.3% and 19.8%).

The least racist state in these polls was California, where only 3.78% of voters against Obama said race was an important factor, and only 5.92% overall said race was a factor. Close behind California were New York (3.41% and 7.44%), Washington (3.51% and 6.84%), and Illinois (3.96% and 7.68%). The biggest surprise might be Georgia, where the numbers were only 4.81% and 8.4%. Maybe it's notable that the Obama campaign fought hard for Georgia, while the most racist states had essentially no Obama commercials or campaign workers.

However, we shouldn't discount the power of racism just because Obama won. In presidential elections, 11.5% of the electorate isn't a small factor; it's a massive group of voters (more than 14 million people).

How big of an influence was racism? The following states would have turned to Obama if the McCain voters who said race was an important factor had switched sides: Arizona (10), Missouri (11), Montana (3), South Carolina (8), West Virginia (5). That's a total of 37 electoral votes. And these states would have turned to Obama if the McCain voters who said race was a factor of any kind had supported Obama: Georgia (15), Kentucky (8), Louisiana (9), Tennessee (11). That's 43 more electoral votes.

Speculating among the states without exit polls on race, it seems likely that Texas (34), Mississippi (6), South Dakota (3), and North Dakota (3) would have gone for Obama without racism, since they were fairly close, based on the levels of racism in similar states. That's 46 more electoral votes.

Altogether, without the influence of racism, by my calculation, the only states McCain would have won would be Alaska (3), Idaho (3), Wyoming (3), Utah (5), Alabama (9), Arkansas (6), Oklahoma (7), Nebraska (4), Kansas (6): a total of 46 electoral votes, compared to Obama's 492. And Obama's popular vote margin would have easily been in double digits.

1 | 2

www.obamapolitics.com

John K. Wilson is the author of five books, including "Barack Obama: This Improbable Quest" (Paradigm Publishers, 2008), www.obamapolitics.com, and "Patriotic Correctness: Academic Freedom and Its Enemies" (Paradigm Publishers, 2008). He is the (more...)

The views expressed in this article are the sole responsibility of the author
and do not necessarily reflect those of this website or its editors.

Contact Author

Contact Editor

View Authors' Articles

Book Recommendations for "2008 Election Presidential Barack"

Say It Like Obama: The Power of Speaking with Purpose and Vision
by Shel Leanne

$21.95
Lowest New Price $12.67
Number of pages: 224
Publisher: McGraw-Hill

The Obama Nation: Leftist Politics and the Cult of Personality
by Jerome R. Corsi

$28.00
Lowest New Price $0.85
Number of pages: 384
Publisher: Threshold Editions