Erika Hall is an assistant professor in the Organization & Management department at Goizueta Business School at Emory University in Atlanta.

Despite the fact that the word “Negro” has been widely considered offensive for decades, the term remained in our country’s official vocabulary for a surprisingly long time. Only last year did the Census stop listing it as an option, and only earlier this month did the Army officially end the practice that allowed service personnel to be addressed as “Negro.”

But what remains – “African-American” and “Black” – should still be up for a bit of investigation.

As a proud woman of color, I am comfortable with both terms. However, I’ve noticed that socially awkward cocktail party conversations become even more awkward when a white attendee begins to refer to people of my racial background — and nervously grapples with whether to call them black or African-American.

After frequently witnessing this unease, I became interested in investigating whether the term black had a negative connotation to white Americans. So, along with colleagues Katherine Phillips and Sarah Townsend, I conducted a series of studies to determine whether white Americans perceived African Americans more favorably than blacks.

In one study, we randomly assigned white participants to associate words with either blacks or African-Americans. Specifically, they selected 10 terms out of a list of 75 (e.g. aggressive, ambitious) that they felt best described each group. The participants that evaluated blacks chose significantly more negative words than those who evaluated African-Americans. Notably, whites did not associate more negative words with “Whites” than with “Caucasians.”

Then, we conducted another experiment, using a fictional crime report to see whether a criminal suspect would be perceived more negatively when he was identified as black or African-American. Whites showed more negative emotions toward a suspect when they read that a black male suspect “was found running east on Lake Street” than when the same suspect was described as African-American. Participants showed their emotional response by indicating how “warm” or “cold” they felt toward the suspect on an interactive thermometer.

These results are not confined to our lab – they live and breathe in the pages of American newspapers. Using analytic word count software, we evaluated hundreds of crime reports in major newspapers from 2000 to 2012. We created a custom dictionary to measure the presence of the terms African-American and Black, as well of the presence of over 300 negative emotion words (e.g. violent, hatred, and enemy). The use of the term Black was positively associated with a negative emotional tone in an article. We found no such impact for the use of the term African-American.

Naturally, we were interested in nailing down the “Why?” question. Perhaps, each term evoked different individuals. For example, if White Americans were told that an African-American man was at the door, would they expect a refined gentleman who looked like former U.S. Secretary of State Colin Powell? If they were told that a black man was at the door, would they expect a more thuggish man who looked like a character from the hit crime series, the Wire? We wondered whether whites perceived blacks as lower socioeconomic status than African-Americans, and we speculated that whites’ feelings toward blacks (vs. African-Americans) could be explained by this factor.

In our final study, white participants viewed an employment application where a job candidate, who was otherwise identically presented, was identified as either African-American or black. Participants were asked to estimate the annual salary of the person, guess whether they were in a managerial role at work, rank their estimated educational level (from no high school degree to a professional degree), and estimate whether they came from a high or low socioeconomic status.

By significant margins, white participants believed that the black applicant was lower status, with less education and less annual income than the African-American applicant. Moreover, only 38 percent of participants who evaluated the black applicant believed he could be in a managerial position, compared to 70 percent of participants who evaluated the African-American applicant.

Although these racial labels cannot account for all of the intense racial discrimination in this country's history, it does makes us ponder: How many of our youth would have been more rightfully vindicated in the justice system if they were first identified as an "African-American" rather than "Black" suspect?


  1. Associated Press (2014, November 6). Army drops use of term Negro in document. The Washington Post.
  2. Yen, H. (2013, February 5). US stopping use of term ‘Negro’ for Census Surveys. Associated Press.
  3. Figure 1 is based on data that is included in our paper on the consequences of identifying racial group members using African-American vs. Black: Hall, E.V., Phillips, K.W., & Townsend, S.M. (2015). A rose by any other name? The consequences of subtyping “African-Americans” from “Blacks.” Journal of Experimental Social Psychology, 56, 183-190.