A featured presenter at an Oklahoma race-relations program says Christianity is linked to racism.

Participants were also told survey data showed “stark, indisputable differences” between the views of black citizens and all other races in Oklahoma—even though those sweeping proclamations may have been based on a survey subsample of few as 44 black Oklahomans, a sample size far smaller than what experts say is necessary to produce reliable information.

“There’s a very high correlation between the most racist attitudes in America and white evangelical Christianity, which of course is very strong in the state of Oklahoma,” said Gary Peluso-Verdend, executive director of the Center for Religion in Public Life at Phillips Theological Seminary in Tulsa.

“Black Oklahoma and white Oklahoma look very, very different,” said Kevin Jessop, CEO of Evolve Research.

Peluso-Verdend and Jessop were both featured in the July session of Advancing Oklahoma, an online program described as “a lengthy conversation about race and race relations in Oklahoma.” The program is offered to the members of Leadership Oklahoma, The Oklahoma Academy, Oklahoma Center for Nonprofits, Oklahoma Center for Community and Justice, and the Oklahoma Hall of Fame. Paycom is the presenting sponsor for the event while 12 other entities are listed as lower-tier sponsors.

During the July session, Jessop presented the findings of an online, mobile-focused survey of 547 Oklahomans who were asked several race-related questions by Evolve Research.

The respondents were 57 percent female and 41 percent male, and the racial breakdown was 8 percent black, 79 percent white, 6 percent Hispanic, 14 percent Native American and 4 percent other.

Majorities of all racial groups said they are happy in their workplace, although whites were the group most likely to report happiness. The share of white and black Oklahomans who said they feel like they “are taken seriously at work” was almost identical—78 percent of whites and 74 percent of black respondents.

But black respondents were also less likely to say they felt empowered at work, had a voice, had their opinion heard, or felt supported—despite most saying they are taken seriously.

Black Oklahomans surveyed by Evolve Research were also the most likely to say they “behave differently when I am around people of a different race or ethnicity” (38 percent of black citizens compared to 13 percent of whites, 15 percent of Hispanics, 9 percent of Native Americans, and 27 percent of “other”).

When asked if they had “experienced racism aimed directly at myself,” black respondents were the most likely to answer in the affirmative (81 percent), but a substantial number of all other groups reported having the same experience—including more than one-in-four white respondents.

A majority of respondents in all racial subgroups also said they have “witnessed racism aimed at a friend, family member, or colleague.”

Asked if black people “have to act like white people to succeed in my workplace,” 37 percent of black Oklahomans agreed compared to 41 percent who disagreed. Among white Oklahomans, 56 percent disagreed.

The poll presentation did not specify if poll respondents were provided a definition of the phrase “act like white people.”

While 69 percent of black respondents said race/racism was one of the biggest issues in Oklahoma, just 21 percent of whites and 25 percent of all Oklahomans responded the same.

“You can clearly see the lines where they’re drawn between, for example, black or African American populations and white populations,” Jessop said. “I mean, there are some stark, indisputable differences here.”

However, based on the demographic information provided for the survey, it appears that as few as 44 black citizens—8 percent of the 547 respondents—may have accounted for all respondents representing that demographic in the survey. The raw number of black respondents who agreed that they had to “act like white people” to succeed at work may have been as few as 16.

In 2011, Matthew Knee, then a Ph.D. candidate at Yale University specializing in campaigns and public opinion, warned that the small sample size for many subgroups in polls can result in a margin of error so large for that subgroup that “there is typically inadequate data about many or most subgroups to say very much about what they think.”

“The views of most minority groups such as African-Americans, Hispanics, Jews, and Asians cannot be measured accurately by a standard national poll, although there do exist very large polls or polls specifically designed to properly sample these groups,” Knee wrote.

In a “brief guide to polling” posted on the site of Ballot Box Scotland, the organization warns that findings associated with “subsamples” in many polls “are not accurate as standalone findings.”

The guide notes a poll of 1,000 people can provide a margin of error within about 3 percent but adds that findings tied to a subsample group of 84 people in that same poll may have a margin of error of 10.7 percent.

“That’s why subsamples should almost always be avoided, and you should definitely ignore anyone making a big deal out of them,” the guide stated.

The very small size of racial subsamples in the Evolve Research survey and associated large margin of error did not prevent presenters from making broad statements about life in Oklahoma.

“We have a big difference in how important race is, and I think racism is—and white privilege and the like—is connected to that,” Peluso-Verdend said. “So I think that makes for a really tough starting point.”

Asked if “Individuals are not treated equally in Oklahoma due to their race or ethnicity,” 51 percent of whites agreed, 83 percent of black individuals, 71 percent of Hispanics, and 65 percent of Native Americans.

Those who believe Oklahomans are not treated equally tended to be younger (76 percent of those ages 18-34 versus 38 percent of those 55 and older) and urban (62 percent among those in urban settings versus 48 percent in rural areas). Seventy-six percent of those ages 18-34 also claimed they had witnessed racism aimed at a friend, family member, or colleague, versus just 39 percent of those age 55 and up.

“Age seems to influence exposure to race and racism in this state,” Jessop said.

When asked if the poll gave any definition for racism, Jessop said the firm “was pretty open with our guidance because we didn’t want to steer the conversation.”

He also said online and social-media exposure was linked to increased perceptions of racism in Oklahoma.

“People who see that people are not treated equally, they’re more likely to get their news, both national and local, from digital sources,” Jessop said. “And that speaks to the algorithms that serve you your news. If you were reading issues on race and racism, then the algorithm is more likely to serve you stuff on race or racism.”

Jessop expressed bewilderment that some Oklahomans who said that race is not an issue in Oklahoma took that stance while acknowledging they have seen instances of racism at some point in time.

However, because a majority of those who say race isn’t a problem also say they are comfortable discussing race and are open to conversation, Jessop said officials “have an opportunity, especially among guys who … see everyone as equal in Oklahoma, to educate them that that’s not the case, and that is clearly not what this statistically sound data is telling us and is showing us.”

The next session of Advancing Oklahoma, scheduled for August, will focus on gaining “empathy and understanding from the Trail of Tears, Oklahoma Land Run, and Tulsa Race Massacre” and understanding “how historical trauma impacts communities of color.”