Race, class, and Americans’ perspectives of achievement gaps

Perhaps even more than usual, this Martin Luther King Jr. Day calls for reflection on race relations and racial equity in America. We have just emerged from a year characterized by police brutality toward people of color, resentment and unrest rooted in that brutality, and the most divisive, bruising presidential election in recent memory.

Education provides a window into the state of racial inequality in the United States and, potentially, the nation’s perceptions of it. Earlier in 2016, we released a study of Americans’ perspectives on race- and wealth-based test score differences. Research on “achievement gaps” has shown large, persistent test score differences between white, black, and Hispanic students, as well as between students from wealthy and poor families. With little previous research on how Americans perceive and explain these gaps, we surveyed the U.S. public about test score gaps between students of different races and classes. We found much greater concern about wealth-based gaps than race-based gaps. At the time, we did not have the foresight to envision the racially and socioeconomically divisive election that would ensue, nor to consider the implications of our findings for a presidential election. Now, however, we see reason to reexamine our data through a new lens and to explore the views of low-income, white Americans, whose perspectives on race and class have attracted considerable attention since November.

First, we step back to review the study’s basic approach and findings. We surveyed a nationally representative sample of 1,000 adults, asking questions about their degree of concern about test score gaps, their support for specific gap-closing policies, and their explanations for why these test score gaps exist in the first place. Importantly, since we hoped to assess whether Americans feel differently about race and wealth gaps, we embedded a randomized experiment. We randomly assigned about one-third of the 1,000 respondents to answer questions about gaps between white and black students, one-third to answer questions about gaps between white and Hispanic students, and one-third to answer questions about gaps between wealthy and poor students. The questions that we asked were otherwise identical. By comparing the answers of these randomized groups, we could test whether Americans feel differently about race-based and wealth-based inequalities.

Our data indicate that they do. We observed much greater concern about, and willingness to address, test score gaps between wealthy and poor students than test score gaps between white and black or Hispanic students. In particular, 64 percent of respondents said it is “essential” or a “high priority” to close the wealthy-poor gap, while only 36 percent said that about the white-black gap and 31 percent about the white-Hispanic gap. Consistent with this, respondents were more likely to support a proposal aimed at closing achievement gaps—like offering bonus money to attract experienced teachers to certain schools—if it was intended to benefit students from poor families than if it was intended to benefit black or Hispanic students. Perhaps most unnerving from the perspective of racial attitudes, we found that almost half of respondents assigned to either the white-black or white-Hispanic gap group (44 percent of each) said that “none” of that test score gap could be explained by discrimination or injustice in society.

In the originally published study, we disaggregated our findings by respondent race and income level, finding that Americans tend to be particularly concerned about gaps when they belong to the lower-scoring racial or socioeconomic group. For example, low-income respondents (those from families earning less than $30,000 per year) were especially concerned about test score gaps between wealthy and poor students relative to their level of concern about gaps between white, black, and Hispanic students. However, we did not disaggregate our findings by race and income level simultaneously, which would have enabled us to see the perspectives of low-income, white respondents in particular.

Looking again at our data, we see intriguing—and perhaps in some ways surprising—patterns in the responses of low-income, white Americans. Consistent with claims that this group, politically, tends to be more concerned about class than race, we found a large difference between low-income, white respondents’ concern about wealthy-poor test score gaps and their concern about racial test score gaps. For example, on a scale from 1 to 5, with 5 being the top priority, these respondents rated the importance of reducing wealth-based test score gaps as a 3.84, on average, and of reducing race-based test score gaps as a 2.99.

Notably, however, what accounts for this difference is not a relative indifference about race-based inequality but rather an acute concern about wealth-based inequality. In fact, the mean level of concern about race-based test score gaps appeared larger among low-income, white respondents than others in our sample, although these differences were not statistically significant.  From our data, it seems that low-income, white Americans are no less concerned than the broader population about the challenges facing black and Hispanic students; they are simply more concerned about the challenges facing the poor.

In the context of our broader findings, we remain sobered by the views expressed in this survey. A large portion of the American public remains generally unconcerned about test score gaps between white and minority children, and many Americans attribute the gaps that exist exclusively to minority parents and children rather than to broader social or historical causes.  However, our findings indicate that, in this context, it would be wrong to ascribe these views just to low-income, white Americans. In reality, we may need much broader changes in public attitudes toward educational inequities before we should expect policymakers to feel much pressure from the public to close today’s test score gaps.