How Much Does Race Still Matter?

Earlier this month, I wrote about research by social scientists at Brown and the University of Michigan who reported that despite the fact that President Obama won a higher percentage of the white vote than any Democratic presidential nominee since 1976, racial resentment had increased during Obama’s first term.

Over the past three weeks, a number of experts in race relations have brought contrary findings to my attention.

Seth K. Goldman and Diana Mutz of the University of Pennsylvania find that the Obama 2008 campaign, in and of itself, had a strong, positive impact on racial attitudes. The two have co-written a book, “The Obama Effect: How the 2008 Campaign Changed White Racial Attitudes,” which will be released later this year by the Russell Sage Foundation.

In October 2012, Goldman published a closely related paper in Public Opinion Quarterly arguing “that the Obama campaign produced a significant and substantive decline in white racial prejudice.” Goldman compared the relatively sharp decline in prejudice during the last six months of the 2008 Obama campaign with the much slower reduction in prejudice over the previous 20 years, as measured by public opinion data from the American National Election Studies, the General Social Survey and the National Annenberg Election Study.

Goldman points out that “the ‘Obama Effect’ was dramatic, reducing racial prejudice by a rate between 5 and 14 times faster than the secular [long-term] trend of decline in prejudice over the previous two decades.” He provided Fig.1:

Standardized Change in White Racial Prejudice per 6-Month Period
Historically (1990–2008) and during the 2008 Campaign (July 2008–January 2009)

To measure levels of prejudice, Goldman used questions from the three surveys, asking whites to rate

whites and blacks on three scales, ranging from hardworking to lazy, intelligent to unintelligent, and trustworthy to untrustworthy.

The higher the score, the higher the level of prejudice.

The enormous television exposure of the public to Obama, both on news shows and in campaign commercials, was crucial to the decline in racial prejudice during the 2008 campaign, according to Goldman:

Throughout the campaign, innumerable images of Obama and his family contradicted negative racial stereotypes and changed the balance of black exemplars in mass media in a positive direction, thus causing reductions in prejudice among political television viewers.

It did not matter, Goldman found, whether stories about the Obama campaign appeared on conservative or liberal television shows; both Republican and Democratic commercials featuring Obama had a prejudice-diminishing effect:

even exposure to conservative programs that criticized Obama’s politics reduced prejudice because these programs nonetheless portrayed him as countering negative racial stereotypes.

Fig. 2 shows a sharp decline in prejudice among whites living in the 25 most contested states where the airwaves were saturated with political ads, compared with the much more modest changes in states that were not contested and thus not heavily advertised in:

Change in White Racial Prejudice in the Top and Bottom 25 States in Television Advertising Spending by the Obama Campaign.

Goldman reports that

racial prejudice declined to a greater extent among McCain supporters, Republicans, and conservatives” than among Obama supporters, Democrats, and liberals “because exposure to a counter-stereotypical black exemplar should be most informative and surprising for those with more negative preexisting images of blacks. On the other hand, counter-stereotypical black exemplars should not provide much new information for those with already low levels of prejudice.

A separate study by the political scientists Susan Welch, of Penn State, and the late Lee Sigelman, of George Washington University, “The ‘Obama Effect’ and White Racial Attitudes,” used data from the A.N.E.S. survey to measure racial prejudice and found a steady decline in prejudice beginning in 1992, when poll questions on the topic were first asked:

Obama’s emergence as an important figure in American politics, his candidacy, and his election did seem to accelerate that trend (as well as reflect it). Thus, Obama’s pursuit of and election to the nation’s highest office can be seen as one of those historical events — such as World War II, the civil rights movement, and the passage of the Civil Rights Act of 1964 — that move mass public opinion in a more egalitarian direction.

Welch and Sigelman write that

on one hand, the Obama campaign enhanced the salience of race in general. On the other, his own characteristics in particular — his obvious intelligence, discipline, and interpersonal appeal — were made salient topics for media and public discussion.

They also note that

Obama is the type of black political leader who has been historically most popular among whites — one who was not part of the civil rights movement, who accommodates rather than confronts, and who maintains close personal and political ties to whites.

Welch and Sigelman find the strongest shifts in a positive direction to be among the youngest and oldest voters. The youngest voters, they argue, have fewer “deeply ingrained negative racial stereotypes” than older voters and their “views of African Americans in general were open to being shaped by the first ‘new’ national political leader that they encountered.”

Welch and Sigelman suggest two possible factors to explain the diminution of bias among the oldest and historically most prejudiced voters. First,

because Obama so clearly defied the racial stereotypes that had prevailed when older whites were growing up, he directly challenged their prejudices and caused many of them to rethink their image of blacks in general.

Second,

Mortality is selective on education and cognitive functioning as well as on health. Thus, it may be selective on intolerance, directly or indirectly, because tolerance is related to education and cognitive functioning. If that were so, the older generation would become more tolerant as the number of its members shrinks.

Fig. 3 gives five age cohorts of whites and their increasingly positive views of black Americans from 1992 to 2008 — the youngest, those who came of voting age during the presidencies of either Bill Clinton or George W. Bush, and the oldest, who first voted before the civil rights movement gained momentum and who had attained the age of at least 73 by 2008. Before the Obama campaign, older voters revealed more racial bias than other age groups. Those differences disappeared altogether by 2008 in terms of evaluating the work ethic of blacks and whites:

The findings of Goldman and Mutz, taken together with those of Welch and Sigelman, stand in contrast to the two studies I reported on Feb. 6. Michael Tesler of Brown and David Sears of U.C.L.A. (whose work I discussed) found that “racial resentment” became a stronger predictor of voters’ negative assessments of the Democratic Party and of their likelihood of switching parties after Obama became the nominee than had been true in previous years. “Indeed,” Tesler and Sears wrote,

the effect of these attitudes on both general election vote preference in 2008 and presidential approval in April 2009 were considerably greater than they had been at any other time in the preceding decades.

I called Tesler up to ask him to ask him how his findings jibed with Goldman’s demonstration of declining prejudice. He said that Goldman’s work found that “the effects were very small.” Tesler remains convinced that since Obama emerged as the Democratic nominee in the summer of 2008, “all racial measures are increasing in importance” and that divisions between blacks and whites are increasing. He does not hold back. “Obama,” he told me, “has had a negative effect on racial polarization.”

As I noted on Feb. 6, Josh Pasek of the University of Michigan, Jon A. Krosnick of Stanford, and Trevor Tompson, the director of the Associated Press-National Opinion Research Center, have also produced research finding that from 2008 to 2012 there was a substantial increase in “explicit anti-black attitudes.”

The difference between these conclusions and those of Goldman and Welch may rest upon the differently phrased survey questions on which they base their conclusions.

Goldman tallied survey responses to a battery of six questions from the A.N.E.S., G.S.S., and N.A.E.S. asking whites to rate both whites and blacks “on dimensions ranging from hardworking to lazy, intelligent to unintelligent, and trustworthy to untrustworthy.”

Tesler, as I mentioned earlier, used a scale

constructed from how strongly respondents agreed or disagreed with the following assertions: 1) Irish, Italian, Jewish and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors. 2) Generations of slavery and discrimination have created conditions that make it difficult for blacks to work their way out of the lower class. 3) Over the past few years, blacks have gotten less than they deserve. 4) 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.

The Tesler approach, which is deployed by a number of other scholars, has produced a contentious, and often bitter, dispute between warring camps within academia.

The leading critics of the Tesler approach are Edward Carmines, a political scientist at the University of Indiana, and Paul Sniderman, a highly respected political scientist at Stanford. Together with Beth Easter, a doctoral candidate at Indiana University, they contend in a 2011 article, “On the Meaning, Measurement, and Implications of Racial Resentment” that “rather than being a measure of racism, racial resentment measures primarily racial policy attitudes.” They make the case that Tesler is not measuring racism, but is instead essentially “asking respondents whether blacks need, require, or are entitled to help and assistance.”

If that is the case, Carmines and his colleagues argue, the questions used by Tesler are not just tapping into attitudes toward African-Americans, but are eliciting race-blind conservative support for self-reliance over dependence on government.

Obama’s 2008 and 2012 victories lend weight to the optimistic assessments of Goldman and his collaborators. Tesler and Pasek to the contrary, there appears to be a far greater consensus, not only among political scientists but among strategists, candidates and the electorate at large, that race prejudice is in decline.

Welch and Sigelman write:

Americans’ racial attitudes have been driven by both demographic changes and specific events and trends. In most cases, these events and trends have moved opinions rather gradually in a more positive direction, but in some arenas, the shift has been minimal or even retrograde.

They go on to add that

the gradual growth of the black middle class, the increase in the number of elected black officials at all levels of government, the growing presence of African Americans in prominent positions in business and the arts, and the rise in the political and economic prominence of members of other ethnic groups along with women of various races and ethnicities all have presumably propelled and reinforced the growth of more favorable attitudes toward African Americans.

Are the Tesler and Goldman findings irreconcilable? Both are based on changing responses to different sets of poll questions. Race prejudice can and does decline, as documented by Goldman and Welch, while the salience of racial issues can increase, as shown by Tesler and Sears. Obama’s candidacy inevitably brought race to the forefront, and the intense media focus on Obama and his family as “positive exemplars” forced many Americans to re-evaluate negative stereotypes.

The gains in 2008 were not irreversible, however. Goldman’s more recent research shows that with the cessation of Obama’s prominence in campaign advertisements, the end of the presidential election news cycle, and with the lowered profile of Michelle Obama and their two daughters, by the 2010 midterms, prejudice had shifted back to pre-2008 levels.

As readers will recall, Republicans not only wrested back control of the House in 2010, but elected majorities in 54 state legislative chambers, the highest number since 1952, and won key gubernatorial contests in pivotal states like Wisconsin, Michigan, Pennsylvania, Ohio and Florida.

While prejudice continues its decline at varying rates of speed, the more pressing question becomes: what will happen as Americans of all races adjust to the fact that Hispanics have displaced African-Americans as the nation’s dominant minority, and now outnumber blacks 50.5 million to 37.7 million?

The United States faces the ascendance of racial and ethnic minorities when the political response to rising levels of debt is forcing adoption of austerity policies at every level of government. One dominant issue in the coming years will be the competition between urban blacks, Hispanics and whites over increasingly scarce government resources. These conflicts will concern not only attitudes but fundamentals: who gets medical care and who does not; which children go to quality schools and which do not; who has food security and who does not.

Domestic racial and ethnic conflict, in turn, could well be subordinated to transnational turmoil as America’s pre-eminence declines and as global competition for markets, for food, for energy, for high-tech professionals, for arable land and for sanctuaries from climate change intensify.

How salient is race? I would say the ability of the United States to go from legal segregation a half century ago to the election of a black president suggests there is enormous elasticity in the American political system, and that the country has the capacity to deal with what it now faces, both inside and outside its borders.