Why That “Follow-Back” on Social Media Is Not Guaranteed

These kinds of “follow-backs tend to happen quickly on social media, usually within minutes or hours of following,” says Maryam Kouchaki, a professor of management and organizations at Kellogg.

Though people might follow someone back relatively quickly and instinctively, there is a wide range of factors at play in the decision-making process, from personal preferences and business reasons to unconscious bias.

In a new study, Kouchaki collaborated with Krishnan Nair of the University of Illinois Urbana-Champaign and Mohsen Mosleh of the University of Oxford to explore how these factors can influence the way social-media connections are formed.

They studied the behavior of thousands of people on the social-media platform X, focusing specifically on interactions between people of different races (Black or white) and with different political beliefs (conservative or liberal).

Overall, the researchers find that people on X are significantly less likely to follow back people who are Black than those who are white.

What’s more, it turns out that this trend prevails across ideological lines. Both liberals and conservatives are less likely to follow back people who are Black than white.

“There was no difference between liberals and conservatives on this,” Kouchaki says. “It shows that liberals are also susceptible to this intuitive bias that is based on stereotypes. Racial minorities are at a disadvantage in these networks because of this bias.”

Testing follow-backs

The research team set out to test whether people would be less or more willing to follow back the social-media accounts of Black versus white people on X. And with partisan animosity on the rise, they also decided to test how people’s political leanings might affect their decision.

“We wanted to know if conservatives and liberals would respond differently to minority accounts,” Kouchaki says. “Would they follow them back? Would they assume Black profiles were liberal? How would they react if the Black profile was conservative?”

The team created 18 profiles on X, or research accounts, for their study. Half were made to look like they were owned by Black people with a politically liberal, conservative, or neutral stance. The other half were made to look like they were owned by white people, divided the same way. All of the X profiles were of men, and each account started with 250 politically neutral followers. (The researchers pre-tested the X accounts with hundreds of people to confirm that they appeared authentic.)

Each account had a pinned post that indicated the person’s ideological leaning. And every other day, the research team refreshed each account by sharing a new post from a mainstream outlet that aligned with the person’s politics.

Over the course of 14 days, they used their 18 accounts to follow 6,000 real accounts on X. The real accounts consisted of a politically balanced sample of people who had shared posts from politically left-leaning or right-leaning media outlets and were not aware of the study.

Then the researchers kept track of which real accounts followed their 18 accounts back.

“The important thing is that we examined this behavior in a setting where participants were unaware that they were being observed,” Kouchaki says. “That’s how you find out how people act when they think no one is looking.”

Where liberals and conservatives align

Across all X accounts, people were 24 percent less likely to follow back accounts owned by Black people than white people. And the pattern held true whether the people making the decision were conservative or liberal.

“You could have expected that conservatives were going to be more discriminating in following back,” Kouchaki says. “But we did not find that; there was no difference among political leaning when it came to following back Black accounts.”

When the research team examined the most extreme right-leaning and left-leaning accounts, the results still held. “Even with extremely liberal users, you still find the racial bias,” Kouchaki says.

A reason for reflection

If liberals are more likely to express concern about racial discrimination than are conservatives, as some research
suggests, then why did liberals and conservatives have such a similar response to following back Black people on social media?

The researchers point out that there are two different ways people process thoughts: deliberative and automative. In situations where there is a motive to appear unprejudiced, people may be more likely to take a slower, more deliberate approach to their actions, such as when deciding what type of content to post publicly on social media. But when the motive to appear unprejudiced is low—such as the mostly private decision of deciding whether to follow back another person—automatic processes play a larger role. And people, regardless of their political leaning, may be subject to more bias during these automative processes.

“There is a common assumption that liberals behave differently toward minorities than conservatives do,” Kouchaki says. “While prior research shows that liberals are more likely to self-report being less racist and care about these issues, this study shows that in making a quick, intuitive decision where they don’t know that they are being watched, that is not true.”

In a society that is often divided on the topic of racial discrimination, it can be important to take a step back and reflect on the impact that automatic, instinctive decisions can have on others.

“I hope this encourages people to reflect on whether they are falling into these automatic biases and to begin the process of self-improvement to better align with their values,” Kouchaki says.