Unlike the case for neighborhood integration, where regional differences have essentially disappeared, inter-racial marriage attitudes continue to show strong differences between southerners and non-southerners.  In 2004, 38 percent of white southerners opposed or strongly opposed inter-racial marriage while 26 percent of white non-southerners said the same thing.  Thus, although there are declines in opposition in both regions, the regional difference persists.  Similarly, those with less education are consistently more opposed to inter-racial marriage, but even among the least educated, opposition has declined.  In 2004, 43 percent of those without a high school diploma opposed or strongly opposed unions between blacks and whites, compared to 1 in 4 among those with at least some college.  Across all levels of education, opposition from 1990 to 2004 has been nearly cut in half.  Finally, from 1990 to 2004, there is a clear effect of cohort on attitudes toward inter-racial marriage, with younger cohorts reporting less opposition than older cohorts.  By 2004, 54 percent of those in the pre-Civil Rights cohort were opposed or strongly opposed to inter-racial marriage (compared to 81 percent in 1990), while just 16 percent of those in the youngest cohort felt the same way in 2004 (compared to 56 percent in 1990).   
 

Finally, one additional social distance question now has sufficient trend data to warrant reporting.  Whites were simply asked how close they feel to whites and then how close they feel to blacks.  We calculate a difference score to identify the degree to which whites feel closer to other whites than they feel to blacks.  Using this abstract question about closeness—related to, but not completely the same as, willingness to share social space—we see that there has been little change over time:  whites continue to feel closer to whites than they do to blacks, with just about one-half in 2006 reporting a difference in their ratings of whites and blacks. 
 
 

Miscellaneous Questions 
 

Table:  3.6 
 

Related to the question of closeness, there is a longer time series in which whites are asked to evaluate their feelings toward whites and blacks on what is called a “feeling thermometer”—where 0 is “cold” and 100 is “warm.”  This question, too, is quite remarkable in its lack of change:  whites, on average, report consistently cooler feelings toward blacks than toward whites.  Though the means have fluctuated over time, in 2004 whites’ feelings were about 5 degrees cooler toward African Americans than whites.   
 

Similarly, there has been the continuation of slight declines in the percentage of whites agreeing that “blacks shouldn’t push themselves where they’re not wanted”—in 2002, 38 percent of whites agreed or strongly agree with this sentiment.  This reflects a considerable drop from the mid-1960s, when the vast majority of whites agreed with this sentiment (78 percent); and is caused by a slow but steady decline ever since that high point.  
 
 

General Summary of White Racial Attitudes 
 

Over the years since the 1997 edition of Racial Attitudes in America was published, the survey record on trends in racial attitudes shows improvement, stagnation, or declines, depending on the dimension of racial attitudes on which one focuses.  The principle-implementation gap largely persists, though two of the implementation questions that continue to be included on surveys show an interesting disengagement with the issue—that is, increasing percentages of respondents opt not to answer the question, instead saying they have “no interest” in the issue.  On questions of government expenditures and preferential treatment, whites are stagnant:  there is little change in levels of support, and in general there is rather lukewarm support, if not outright opposition, to the kinds of policies and programs presented by these survey questions.   
 

Questions of social distance and stereotyping show perhaps the clearest signs of improvement:  fewer and fewer white Americans readily endorse statements that blacks are less intelligent and hardworking than whites; and fewer verbally object to increasing levels of inter-racial mixing in neighborhoods and in marriage partners.  These trends must be interpreted with caution, for they may reflect at least to some extent changes in social norms about what kinds of answers ought to be reported on surveys rather than changes in actual levels of stereotyping and in openness to living with and marrying African Americans.  Studies that use increasingly sophisticated measurement approaches that can disentangle the possible explanations for the pattern of changes are certainly warranted.  This might include the use of experiments within surveys, the further development of measures of unconscious stereotyping, as well as in-depth qualitative studies of racial attitudes.  One might interpret this pattern of changes to reflect quite strong changes in racial norms (Schuman et al. 1997) that now apply to questions of this type.  That is, it may no longer be acceptable to admit to these kinds of attitudes in a semi-public setting like a survey interview.  This in itself reflects a change in racial attitudes in this country even if it does not reflect changes in the hearts and minds of Americans.  What the implications of this change are, however, is subject to interpretation. 
 

Finally, there are a set of questions that have become less racially liberal over time.  Specifically, questions that ask whether African Americans continue to experience racial discrimination and if the consequences of past discrimination and slavery continue to shape the experiences of today’s African Americans.  Essentially, the trend is for fewer whites to acknowledge that African Americans are adversely affected by past and persistent discrimination.  This trend is of consequence, since support for policies targeted toward helping African Americans are closely related to whether or not an individual believes that African Americans continue to face these kinds of barriers.