But implicit racism -- prejudice unacknowledged in public and, in many cases, hidden from conscious awareness -- is a factor in opposition to Obama's health policies. That's the conclusion of a provocative new paper that's one of two research reports on prejudice and the president just published in the Journal of Experimental Social Psychology.
Participants in a yearlong study who scored low on implicit prejudice found the proposed health care plan equally appealing whether it was attributed to Bill Clinton or President Obama. However, those who scored high in implicit prejudice supported the plan when it was linked to Clinton, but opposed it when linked to Obama.
A team of scholars led by psychologist Eric Knowles of the University of California, Irvine recruited 285 Americans (236 white, 43 Asians, six Latinos) from a Stanford University database. (Blacks were deliberately excluded, since the issue at hand was how non-blacks feel about blacks, and whether those feelings affect voting decisions and political opinions.) In late October 2008, the participants took tests designed to measure both explicit and implicit racial prejudice.
To gauge levels of implicit racism, they performed the Go/No-go Association Task, a variation on the Implicit Association Test. Knowles describes it this way:
"The Go/No-go Association Task has individuals categorize words -- specifically, stereotypically African-American names and words that carry pleasant or unpleasant meaning -- into the categories ‘black' and ‘bad' or ‘black' and ‘good.' Some participants have great difficulty categorizing black names and pleasant words at the same time, while finding it easy to simultaneously categorize black names and unpleasant words. These participants are deemed to mentally associate the black category and bad things."
In a second assessment the following week, participants were asked about their attitudes toward then-candidate Barack Obama. Three weeks later, they were asked who they voted for in the presidential election. Finally, in early October 2009, 230 of the original participants were asked about the current health care debate.
Half of them were asked about their support or opposition to the Democratic health care proposals, and asked to rate six potential concerns about the policy. The results: "Subjects who showed no bias against blacks (in the original test) were about evenly split on Obama's health care plan, with 48 percent opposed to the plan and 52 percent supporting it," Knowles reports. "However, subjects with an anti-black bias were opposed to the Obama plan by almost two-to-one, with 66 percent opposed and 34 percent supporting it."
The remaining 130 participants "were randomly assigned to read a description of health care reform framed either as being President Obama's plan or Bill Clinton's 1993 plan. The description was identical across conditions and described elements common to both plans. After reading the description, participants rated their attitude toward the plan."
The results were quite striking.
"When the health care reform plan was framed as former President Clinton's idea, a majority of both high- and low-prejudice subjects (65 percent and 66 percent) supported the plan," Knowles said. "However, when the plan was framed as Obama's idea, support among biased subjects fell to 41 percent, while support among low-bias subjects remaining essentially unchanged (70 percent)."
Knowles acknowledges this is not definitive proof of racism. "We do think that some of our subjects like Obama less than Clinton as an individual," he said. "However, because these participants happen to be higher in prejudice, we think it's a race-based dislike. High-prejudice participants (in our study) were prone to believe all sort of putatively non-racial negative things about Obama -- that he's a socialist, a radical, unpatriotic, etc. That these attributions correlate with prejudice indicate to us that this is a racial thing."
The second study confines its analysis to the 2008 election, but comes to the same conclusion. A research team led by psychologist B. Keith Payne of the University of North Carolina, Chapel Hill surveyed 1,056 Americans before and after the election. Their levels of implicit and explicit racism were rated and compared to their voting behavior.
In this study, implicit prejudice was measured using the Affect Misattribution Procedure, in which photos of either black or white faces were flashed onto a computer screen, followed by a Chinese ideograph. "Respondents were instructed to judge whether each ideograph was pleasant or unpleasant while avoiding influence from the photos," the researchers write. "Unintentional influence of the primes on judgments can be used to measure attitudes toward the white and black faces."
"We found that implicit and explicit prejudice predicted voting behavior in subtly different ways," the researchers report. Not surprisingly, those higher in explicit prejudice were less likely to vote for Obama and more likely to vote for Republican candidate John McCain.
But those with higher levels of implicit prejudice were less likely to vote for Obama, but they were also less likely to vote for McCain. "Instead, they were more likely to either abstain, or to vote for a third-party candidate," the scholars report.
Knowles and Payne both concede that some scholars question the validity of these implicit racism tests. But both strongly defend their findings, with Knowles noting that levels of implicit prejudice successfully predicted opposition to Obama's health care plan "after a lag of almost a year."
"Because most people wish to appear fair-minded -- both to others and themselves - they also embrace more principled, ‘color-blind' rationales for their race-based views," Knowles and his colleagues write. They conclude that "while our findings do not corroborate the view that opposition to the president is motivated primarily by racial prejudice, they clearly rebut those who argue that opposition to Obama and his policies have nothing to do with race."
Payne and his colleagues agree, arguing that their findings suggest "implicitly measured prejudice is indeed associated with unambiguous and meaningful discriminatory behaviors." But they add that their study also suggests the impact of explicit racism on voting behavior may be underestimated.
"Our findings suggest that Mr. Obama was not elected because of an absence of prejudice," they conclude, "but despite its continuing presence."
