Baked-In Bias: Whites Less Likely To Respond To Survey Emails Sent By Blacks

The explosive subject of race, so prevalent in so many areas of life, has now crept into the parochial confines of the email business. New research shows that White Americans were less likely to respond to an email survey they believed was sent by a Black person than by a White person. 

The research, published last week in the Proceedings of the National Academy of Sciences, reflects a subtle but real form of perhaps unknowing discrimination.  

“A lot of prior studies on racial beliefs have been attitudinal, where researchers asked people about their feelings about minority groups,” states lead author Ray Block, the Brown-McCourtney Career Development Professor at the McCourtney Institute and associate professor of political science and African American studies at Penn State. 

However, “people will often hide or not be truly honest about their beliefs,” in those types of studies, Block adds. “Our measure of discrimination is behavioral. We didn’t care about what people said, we cared about what people did.”



The researchers contacted 250,000 email addresses culled from a nationwide voter registration list and a commercial email list. Included were Asian American/Pacific Islanders, Blacks, Hispanic/Latinos and Whites, reflecting current population breakdowns.  

Recipients were asked to volunteer to take a survey about contemporary political issues — all they had to do was click a link within the email. 

Each person received two emails spaced two weeks apart, one from an apparent Black sender and one from a assumed White person.

How were the purportedly Black and White sender names selected? They were chosen based on perceptions revealed in previous research and in government records. 

The results provided “a snapshot of a subtle form of racial bias that is systemic in the United States,” says an article abstract. “What we term everyday or “paper cut” discrimination is exhibited by all racial/ethnic subgroups — outside of Black people themselves,” the article abstract says.

Specifically, 1.6% of the recipients responded to the presumed White sender, and 1.4% to the Black emailer. The presumed White name was 15.5% more likely to drive a response. 

In the final count, the purported White sender drew 4,007 responses, the Black surveyor drew 3,620. 

Now this may not seem like a prohibitive difference, but magnify it by several hundred in a marketing setting and it could mean the difference between profit and loss.

“Our definition of discrimination had nothing to do with ill intent and everything to do with disproportionate treatment in some kind of way,” Block said. “And we did find that.

He adds the results were the same even when broken down by geographic region. “People might assume discrimination may be worse in certain parts of the country, but we didn’t find that.” 

The study raises many questions, such as: “What if discrimination correlated with partisanship, what if it correlated with opinions about policy?” Block asks. “Future research can explore these and related questions.”

The research, which earned additional support from the National Science Foundation and the Brigham Young University College of Family Home and Social Sciences, was also worked on by Charles Crabtree, Dartmouth College, John B. Holbein, University of Virginia, and J. Quin Monson, Brigham Young University.

Next story loading loading..