When NPR announced the hiring of Gary Knell as CEO in October, Joel Dreyfuss of The Root, an African-American oriented website, published an open letter challenging Knell and NPR to work harder to diversify its staff and programming. Dreyfuss said he wanted NPR to be "a reflection of the America I live in," which, of course, is part of public radio's mission.
Dreyfuss's demand has been a running one for decades among some racial and ethnic minority advocates and listeners. Discrimination lawsuits, messy firings and various diversity attempts helped fuel the attention. So, over recent months, I have been building my own notes in an attempt to measure just how good a job NPR is doing now. What I find so far is that, racially and ethnically, NPR is not doing badly, and is getting better.
To see if Latino, black and Asian listeners find programming that appeals to them, I broke down NPR audience figures by higher education and income. I discovered that within these categories, the levels of representation of the minority groups and whites are not far apart. Minority staffing in the newsroom and on air, meanwhile, continues to improve. NPR does significantly better than the industry averages in radio, television and newspapers. But then, we expect NPR to do better.
Knell has publicly committed himself to further improve diversity, not just of race and ethnicity, but also of age, geography and what he calls "thought." He is aided by the vice president for diversity, Keith Woods, whose role has taken on greater internal importance in the wake of the fiasco over the firing of Juan Williams.
Additionally, Ellen McDonnell, the executive director who oversees the news shows, is in the midst of creating and implementing a strategy to push the shows, as she puts it, "to bring in more diverse voices to reflect the rich diversity of this country." Her effort follows on similar earlier projects, but brings a methodological strategy that she is developing with Columbia University's schools of business and journalism.
With national elections approaching in November, the political coverage in particular is too Washington-focused and filled with white voices, she said. She doesn't want to eliminate those voices, she said, but explained: "That lens is too narrow."
In my own study, I don't actually quantify a breakdown of the news coverage by race, and to my knowledge no one else has either. I tried, but found that it is impossible to classify stories as black, Latino or Asian. Most stories cross over and are of interest to many groups—even stories that might focus primarily on one racial group. There were so many caveats that any numbers I tried to come up with were useless.
But I do have a separate caveat. To look at race and ethnicity does not mean that I believe NPR should write any goals into stone. Race and ethnicity still matter in America, but less as time goes by. I used to teach immigration policy at Harvard, and that background tells me that the United States is the single most successful example in world history of a multi-racial and multi-ethnic society. Sociologists and market researchers today have identified what some call "a new mainstream" in which the educated and the young identify with each other more than with their ethnic and racial roots, though the roots don't disappear.
Indeed, I am also trying to follow how NPR is doing in covering the white working and middle class. My sense is that the national news media, of which NPR is part, has done a poor job in recent decades of covering this segment of Americans. The recent coverage of the Tea Party, which to some degree represents this group, has been mostly political, not social in character. I am a critic of some of Charles Murray's earlier books, but feel that he is on to something in his latest, Coming Apart. His study uncovers a huge cultural chasm that has opened between poor and working class whites, on the one hand, and upper income and highly educated whites on the other. This latter group includes most national journalists.
Looking at NPR, the overwhelming majority of its radio audience is in fact white -roughly 87 percent, according to research pulled together for me by Lori Kaplan of NPR's Audience, Insight and Research Department. This is substantially higher than the 77 percent of adult Americans (18 and older) who are white. Asian-Americans make up nearly 4 percent of the audience, but roughly 3 percent of the adult population. African-Americans and Latinos, however, are under-represented among NPR's listeners. Blacks make up nearly 12 percent of the adult population but just a little more than 5 percent of NPR listeners. For Hispanics, the numbers are 14 percent versus 6 percent.
Fine. But irrelevant.
Using the total adult population is the wrong baseline. NPR appeals overwhelmingly to college-educated Americans. As you can see in the accompanying chart, more than two-thirds of NPR listeners have a college degree, compared to less than one-third of all Americans. Because college graduates earn more than people with just a high school degree or less, the NPR audience trends to higher-than-average income, too. Forty-five percent of listeners have a household income of more than $100,000 a year. A quarter of all American households earn that much.
The sophistication of its audience—combined with its sheer size of more than 25 million people listening to an NPR program each week—reflects the trust in NPR news, as well as the appeal of its storytelling. Whether these listeners have a college degree or not, they are smart, informed, interested in the world around them and they tend to be opinion leaders. It is an audience that has few or no other alternatives for independent and fair news on radio. It is also the audience that NPR wisely goes after.
So, it seems to me that the important question in looking at minority listenership is what percent of the audience is NPR attracting among college graduates within each racial and ethnic category. I went back to Lori Kaplan. The gaps now narrow significantly.
This chart shows adult listeners with a college degree in the four principle racial and ethnic categories. The first bar records what percent within each group listens to NPR at least once a week, regardless of income. The second bar measures listenership in households that earn more than $100,000 a year.
What you see is that the groups aren't so far apart in how they index. Among all income levels, more than 11 percent of whites with a college degree listen to NPR. This compares to 9 percent of Asians with a college degree, nearly 7 percent of Hispanics and 6 percent of blacks.
The gap is even narrower among those with household incomes of more than $100,000. Asians in this category actually out-index whites. Nine percent of these wealthier college educated Asians listen to NPR, compared to less than 8 percent of whites. The percentage among Latinos is close behind at slightly more than 6 percent, followed by African-Americans at 5 percent.
In an ideal world, the same percentage of all groups would listen to NPR. This would imply that the programming and news coverage speaks to them equally. Many factors might explain the current differences, among them marketing. But the size of the gaps suggest that while NPR should work to close them, the differences are not major.
A qualifier: Some audience members list belonging to more than one race or ethnic group. The number of people who do this is small and doesn't substantially change the above results.
This then brings us to staffing. Diversity in staffing is important for two reasons. One, it assures that there are voices in the newsroom raising the concerns and points of view of different segments of our society. Two, it assures that all listeners hear voices with which they can identify on air, making NPR their radio. Your radio.
Woods said it best in a speech at the University of Colorado Boulder this past week:
Let me speak first in terms of ideals. Journalism is the only profession mentioned in the U.S. Constitution. That's how attached it is historically to democracy. Freedom of the press carries with it an incumbent responsibility that tends to get lost in the noise that sometimes envelops conversations about the First Amendment...
My profession does not always act responsibly. But the ideal that underlies our actions, honorable or otherwise, is simple, evergreen, and at its core, democratic. That ideal is expressed in the three ethical pillars upon which responsible journalism stands:
Tell the truth as fully as possible;
Remain independent of undue influence;
Minimize harm to vulnerable people;
So if you ask, "Why does it matter that journalism 'sounds like America,' " I would say this: Because that is how ethical journalism sounds. Tell the truth as fully as possible means you have to seek out that truth wherever it will be found.
Blacks, Latinos, Asian Americans, Native Americans and Hawaiian Islanders, or "people of color," make up 23 percent of NPR's newsroom, meaning reporters, editors, producers and managers, according to NPR's Human resources Department. This compares to just 7 percent for radio in general, according to a survey by the Radio Television Digital News Association and Hofstra University. The television news industry average approaches that of NPR at a rounded off 20 percent. Daily newspapers fall far behind at 13 percent, according to a survey by the American Society of News Editors.
Seen another way, NPR falls short. People of color make up roughly 28 percent of the adult population, compared to their 23 percent representation in the newsroom. But once again, the more important measure is by staffers with college degrees, as almost all journalist and management positions require one.
People of color in 2010 made up roughly 20 percent of Americans with college degrees, according to Kaplan. She pulls her numbers from the demographic source widely used by news media organizations, the GfK MRI Doublebase, which in turn reflects U.S. Census numbers. This 20 percent overall compares to the 23 percent of NPR reporters, editors, producers and managers who are people of color.
NPR does not quantify the education levels of its staff, but almost all—if not all—of these journalists and managers have college degrees. We thus can feel confident that we are pretty much—and perhaps totally—comparing apples to apples. The newsroom numbers I use, moreover, do not include support staff that does administrative work. In other words, we are just looking at the people who bring you the news, and therefore control it.
Seen this way, NPR shines in its minority hiring.
Let's look more closely at each individual racial and ethnic minority group. Seven percent of U.S. college graduates are African American. Blacks make up 12 percent of the newsroom—much more than their 7 percent weight among college graduates. Hispanics, however, are slightly under-represented. They make up six percent of the Americans with college degrees but five percent of the newsroom. Asians do exactly the reverse. They are five percent of Americans with degrees and six percent of the newsroom. The Native American sample size is too small to draw many conclusions. There is one person among NPR's journalists and managers who said he or she was Native American, according to Human Resources. This is 0.2 percent of the newsroom, compared to the 0.6 percent of college graduates who are Native American.
Given the large black population in Washington, D.C., where NPR is headquartered, it may make sense that they are the largest minority group in the newsroom. I haven't looked into this. My anecdotal sense, however, is that black males—as opposed to black women—are under-represented, but that is a separate issue.
The bottom line here is that in terms of the nation's largest racial and ethnic minority groups—blacks, Latinos and Asians—NPR staffing may have arrived. Why do I hedge by saying "may"? You need college graduates to produce the level of NPR's journalism. Many staffers actually have graduate degrees.
But to "sound like America," does NPR need a staff that more closely mirrors the total demographic weight of each ethnic and minority group?
I don't think so, but I recognize the legitimacy of the tension among the three goals of reflecting the American population, reflecting NPR's target audience and reflecting journalism's talent pool. There is no one good answer. But I think it is safe to say that in resolving this tension, NPR is not doing badly, and possibly very well. By my way of thinking, it is truly doing well.
The following chart with information prepared for me by the Human Resources Department further breaks down NPR's newsroom by racial and ethnic categories, and by managers, journalists and support staff.
What follows are two charts that give more detail on NPR staffing, according to NPR's Human Resources Department. The first chart separates journalists, senior editors or managers, and support staff. The second chart includes all of NPR's staff. As you can see, people of color are distributed throughout the hierarchy and are strongly represented among senior management.
I am a news ombudsman, not a corporate one, and so I consider NPR's non-news staffing and corporate management to be outside my purview. But as all the numbers were given to me by Human Resources, I pass them on to you. In general, the numbers update NPR data released in 2009 for all departments and represent a small but respectable improvement in two years. This is especially so in upper management, production and editorial positions.
I can also say with certainty that NPR has a Latino ombudsman, its first.
In the coming weeks, I will be doing a quantitative regional bias study and a qualitative look at who gets interviewed on air. I welcome suggestions on other ways to analyze diversity.
2010 and 2011 Doublebase GfK MRI: MRI data generally has a 5% margin of error. The sample size is 52,000 persons across the US and the data source is used extensively across US Media organizations.
From MRI on how respondents select their race:
It is up to the respondent to decide. We show the respondent a card and ask: please tell me the number next to the race or races you consider yourself to be. There are 5 options: White, Black/African American, American Indian or Alaska Native, Asian or Other.
Learn more about the 2010 Doublebase GfK MRI online.
(ACT 1 based on Arbitron Nationwide, Persons 12+, Fall 2011, Monday-Sunday Midnight-Midnight)
The Arbitron numbers are based on two different methodologies — a 7 day paper diary which relies on recall and a passive meter carrying device. The total surveyed between the two Arbitron methodologies, is approximately 350,000 persons across the U.S.
Lori Grisham contributed to this report.