After this week’s presidential debate touched on race, I wanted to know something about how the U.S. media covers this urgent topic. To find out, I reached out to Erik Bleich, a political scientist at Middlebury College. His Media Portrayals of Minorities Project publishes an annual report on coverage of racial and ethnic groups in U.S. newspapers and has just released its report on 2019. What follows is a lightly edited transcript of our discussion.

Joshua Tucker: First, tell me a bit about how you do your research.

Erik Bleich: We collect every article from four newspapers with a national circulation — the New York Times, USA Today, the Wall Street Journal and The Washington Post — that mention African Americans, Asian Americans, Native Americans, Latinos, Jews or Muslims. We then look at two essential things: How positive or negative are those sets of articles, and what are the main topics of coverage? This gives us a good sense of what newspapers convey about each of these groups across the year.

JT: What do you mean by “positive or negative?” How do you measure that?

EB: We compare all the words in each article to eight widely used dictionaries of positive and negative words. Our score for each article is benchmarked against the average score of over 40,000 articles from 17 U.S. newspapers from across a 20-year period. So when we say an article is extremely negative, we mean it has a lot more negative words than positive words in it, relative to the average newspaper article.

Stories that are very negative are not always negative about the group or person. For example, there were a lot of very negative Latino articles that covered the El Paso Walmart shooting in August 2019, which drives down the tone of Latino articles for the entire year. Negative articles can at times evoke sympathy, but even in those, they often leave a lasting residue of negative feelings associated with a group or person in the back of readers’ minds.

JT: Given the debate’s explosive discussion of white supremacist groups and everything that has happened this year with the Black Lives Matter movement, what can you tell us about coverage of African Americans in 2019?

EB: African American stories were relatively visible in U.S. newspapers last year and, overall, were relatively close to neutral in tone. Of course, there were a lot of race-related articles about policing. In fact, almost a quarter of all the Black-associated stories mentioned words linked with police or policing. These articles were strongly negative. Some of this coverage included an uptick in stories about New York City’s earlier “stop-and-frisk” policy when former mayor Mike Bloomberg entered the Democratic primary for the presidential nomination. Other coverage of African Americans, however, focused on more positive stories, such as the Democratic presidential bids of senators Kamala Harris and Cory Booker, or museums, opera and art in the African American community.

JT: What is the most striking thing you found for 2019?

EB: Coverage of Muslims stands out as having been far more common and far more negative than that of the other groups that we monitored. There were nearly 9,000 articles that mentioned Muslims in 2019. The next most covered group was African Americans, with closer to 6,000 articles. The tone of Muslim articles was also strongly negative — almost four times as negative as the tone of those touching on Latinos, the group associated with the second-greatest negativity.

JT: What is the focus of the negative articles that mention Muslims? Are these mostly about foreign events and locations?

EB: Yes, that is a lot of what we see. Stories about Syria and ISIS, the Iran nuclear deal, or terrorist attacks in Sri Lanka and New Zealand [where a white supremacist shot and killed 51 Muslims at two mosques] include quite a high level of negative language. But even articles focused on events in the United States are still more than twice as negative as articles about any other group. Around 4 percent of all articles that mention Muslims also mention Rep. Ilhan Omar, for example, and these stories are just as negative as the average Muslim article. Some articles are about her as a target of President Trump’s insults; others are about accusations of anti-Semitism.

JT: What are some other big takeaways from your study?

EB: In addition to analyzing articles that mention a group once or twice, we also zoom in on stories with three or more mentions of a group, and which are more likely to be “about” those groups. In most cases, the tone is relatively similar no matter how many times a group is named. One case is a notable exception. Stories about Jews go from being essentially neutral to clearly negative when Jews are mentioned three or more times. These articles tend to contain more references to topics like anti-Semitism and the Holocaust; a majority of them touch on Israel/Palestine. These topics are more intensely associated with Jews as a group by these newspapers, and they drive greater negativity.

JT: Is coverage of these racial, ethnic, and religious groups mostly bad? Do you find any good news?

EB: No, and yes, to some degree. On the whole, coverage is not markedly negative. For African Americans, Asian Americans and Native Americans, the average article tone is roughly the same as that in the average U.S. newspaper story. Native Americans and Asian Americans are the least-covered groups in 2019, but coverage is very slightly positive in stories that mention them three or more times. While Native American articles included significant coverage of Elizabeth Warren’s DNA controversy, fully 30 percent mentioned “museums.” Among Asian American articles, a large proportion touched on the theme of educational successes.

That said, a closer look might reveal that Native American museum stories tend toward the folkloric, or that articles about Asian Americans reinforce “model minority” stereotypes; I don’t want to imply a unique or simplistic interpretation of these articles. Our overview provides a way to grasp tone and topics, which provides a jumping off point for more detailed analyses.