Analyzing the 2008 Presidential Election in the South

Pearl K. Ford, University of Arkansas

Angie Maxwell, University of Arkansas

Todd Shields, University of Arkansas

FAYETTEVILLE, Ark. – When University of Arkansas political scientists analyzed surveys conducted shortly before the 2008 election in two representative Southern states, they found that voting behavior was significantly influenced by “a deep, subtle and modern symbolic racism.”

In a paper published in the June 2010 issue of Presidential Studies Quarterly, Pearl K. Ford, Angie Maxwell and Todd Shields examined the effect of the Obama candidacy on candidate support in Arkansas and Georgia. Their research contributes to the scholarly understanding of symbolic racism and for the first time examines its impact on a national election. Their research results spanned political parties and the patterns held true for both Republicans and Democrats.

“To understand the New South,” Shields said, “we’re going to have to understand the new politics of race.”

Ford points to Obama’s election as “a significant achievement.” Yet, she said, “There is continual evidence of racism in America and the American South that still needs to be addressed. The symbolic racism we’re measuring has significant impact on policy and political choices.”

Despite claims by some scholars that race is no longer a primary factor in southern politics, the researchers wrote, “Our analysis of the way in which symbolic racism was activated by the first African American presidential candidate indicates that in the 2008 election symbolic racism was a strong and significant predictor of candidate support in both a peripheral and deep southern state.”

Much of the research on symbolic racism began in the post-Civil Rights era. When people were surveyed about their attitudes toward civil rights, Ford said, researchers at that time noticed “a disconnect between what people said and the policies they supported. For example, people would say they believed that all children had the right to attend integrated and quality public schools, but they were against policies such as busing or the redrawing of school district lines to achieve equity in education.”

The blatant, public racism of the Jim Crow era had declined; but, the Arkansas researchers wrote, a system of beliefs remains that “denies the ongoing struggle for equality experienced by African Americans.” Previous studies have shown that symbolic racism “is closely related to white opposition to various public policies that are indirectly linked to race, such as housing, busing and crime.”

Symbolic racism is often linked with conservative values. Maxwell traced this connection back to the 1964 presidential campaign of Barry Goldwater, who ran on a platform of small government and won four Deep South states, a break-through for Republicans. Subsequently, she explained, Richard Nixon’s campaign “figured out that what appealed to Goldwater voters was the part about race.” Small government philosophy “symbolically came to mean to a lot of the South that the federal government should not intervene in civil rights issues.”

“Not all conservative values are based in racial conflict. What we need to understand is where do the non-racial attitudes end and where do the racially based attitudes begin,” Ford said.

“It’s important to me that we understand the white South and why it votes the way it does without being dismissive or demonizing,” Maxwell said. “If we identify symbolic racism, we can start to get an accurate picture of how the races really feel about each other, and that’s the first step to progress.

“We need to understand real conservative values that have nothing to do with race, such as concern with the tax burden or sending money to China,” Maxwell said. At the same time, she noted, “You have to be conscious that race lies below some conservative values so that you can reject the symbolic racism part of those values and pass along the positive values.”

Past research on symbolic racism in elections had analyzed local races. The 2008 election presented the Arkansas researchers with an opportunity to examine whether symbolic racism would explain election outcomes in the first presidential race with an African American candidate from a major party.

The two states they chose offered different situations to test. In Arkansas, there was not a large number of black voters, and Obama never visited the state to campaign. In Georgia with its significantly large African American population, the Obama campaign targeted areas where it was possible to register significant numbers of black voters. The unique campaign environments of the two states allowed the researchers to test whether “the mere presence of an African American presidential candidate at the top of the Democratic ticket” was sufficient to activate racially conservative attitudes in Arkansas or whether Georgia’s intense presidential campaign was necessary “to bring symbolic racism to the front of voters’ decision-making processes.”

In the month before the 2008 election, the researchers asked participants in the Arkansas Poll and the Georgia State Poll a set of established questions for measuring racial attitudes. For example, they asked for agreement or disagreement with statements such as “It’s really a matter of some people not trying hard enough; if blacks would only try harder they could be just as well off as whites,” or “Generations of slavery and discrimination have created conditions that make it difficult for blacks to work their way out of the lower class.”

“A substantial scholarly literature in political science and psychology continues to demonstrate that these attitudes are separate from other conservative attitudes, such as attitudes toward the size of the government or national debt,” Ford said.

Both states went for John McCain in the election. Obama had campaigned heavily in Georgia and brought additional voters to the polls, increasing the Democratic total vote by 34 percent over the previous election. There was little presence by the Obama campaign in Arkansas, and at the presidential level, Democrats lost 11.3 percent of support from 2004.

In the past, scholars had questioned the usefulness of the symbolic racism scale. They had suggested that voters who were more educated and more politically sophisticated would score lower on measures of symbolic racism, not because they were less racially conservative, but because they better understood complex political issues. The less politically sophisticated might score lower, it was suggested, because they were more likely to “embrace individualistic explanations for complex political problems.”

Because the Arkansas researchers included a measure of political sophistication in their models, they were able to test for the differences in the effects of symbolic racism across levels of political sophistication. Their research showed that symbolic racism played a strong role in decision-making across the board.

“The big ‘ah-ha’ moment for me was when we saw people who were self-identified as liberal and Democrat and were unemployed, yet they weren’t voting for Obama. That’s the kind of incongruence you don’t usually see,” Shields said.

In the end, the researchers found that “despite alternative explanations ranging from ‘Hillary backlash’ to political sophistication to economic evaluations, our models consistently indicated the importance of symbolic racism in evaluations of the two main presidential candidates.”

The researchers concluded that the future of southern politics “may depend on the extent to which new voters continue to participate and the extent to which symbolic racism continues to play a role in the choices made by white voters.”

Ford and Maxwell are assistant professors of political science in the J. William Fulbright College of Arts and Sciences. Shields is a professor of political science and the director of the Diane D. Blair Center of Southern Politics and Society.