What the 2020 census can and can't tell us about race : Code Switch The 2020 census data is finally here! At first glance, it paints a surprising portrait of a changing United States: The number of people who identify as white and no other race is smaller; the share of multiracial people has shot up; and the country's second-largest racial group is... "some other race." But resident census-expert Hansi Lo Wang told us that when you start to unpack the data, you quickly find that those numbers don't tell the whole story.

Painting By Numbers

  • <iframe src="https://www.npr.org/player/embed/1041510105/1047467112" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

GENE DEMBY, HOST:

What's good, y'all? This is Gene hopping in before the show to just let y'all know that we are searching for the next host of the CODE SWITCH podcast, someone to come join us in the other host chair. You know, we're looking for someone thoughtful and funny and collaborative and weird, frankly, who's down to come tell some really fascinating stories about race and identity with us. If you think that might be you or your cousin or somebody you know, get your applications in soon. You can do that by going to npr.org/codeswitch. There's some information there about the posting on our blog. All right. Onto the show.

(SOUNDBITE OF MUSIC)

DEMBY: What's good, y'all? You are listening to CODE SWITCH from NPR. I'm Gene Demby. During the existential sinkhole of 2020 - if we can call it that - when it felt like, you know, the world was coming undone, the federal government was still trying to carry out one of its most logistically complex constitutional duties.

(SOUNDBITE OF ARCHIVED RECORDING)

UNIDENTIFIED ACTOR: (As character) In America, we all count, no matter where we call home, how we worship or who we love. And the 2020 census is how that great promise is kept because...

DEMBY: As you know, every 10 years, the federal government attempts to count every single person living in the United States to determine what the population looks like in terms of income and occupation, age and gender and, extremely relevant for our purposes at CODE SWITCH, race and ethnicity. And that's how we determine all sorts of things like redistricting and federal funding, and some of that initial data is finally out. And that data paints a really fascinating and sometimes confounding picture of how the U.S. population is changing. I mean, you might have seen headlines about how the white population is shrinking and the country is becoming more multiracial. But as Shereen would've said - tear - it's complicated.

And to break down what these numbers do and don't tell us and what these headlines and hot takes are getting wrong, we decided to hit up Hansi Lo Wang. Y'all have heard him on the podcast before. He's a correspondent on NPR's National Desk. He's our resident census expert. Oh, and I should say also, you are a CODE SWITCH O.G. back before we were even CODE SWITCH. But you were here in the very beginning before you decided to own this other beat that's super important. And you've joined us a few times to talk about all the political fighting and maneuvering over the most recent decennial census. There was fighting over which questions should be on it.

At one point last year, which was, you know, during an election cycle and a global pandemic, the Trump administration shut down the census count. Can you just remind us about some of the drama that was surrounding the census count last year in 2020?

HANSI LO WANG, BYLINE: Yeah, I mean, it was a hot mess. The 2020 census was a hot mess, and it was upended by the coronavirus pandemic. It was upended by interference by officials and former President Donald Trump's administration. There was door-knocking that was supposed to start way earlier than it did. That got delayed because of the pandemic. There was counting that was supposed to continue on through to Halloween. That got cut short because of the Trump administration. And all of that really raised the risk of undercounting historically undercounted groups. We're talking about people of color, we're talking about immigrants, we're talking about renters, we're talking about rural residents - all groups that are less likely to fill out a form on their own and really are more likely to be counted when you send a person, a census worker, to go knock on their door. And this was during the Trump administration, the same administration that, before the count officially got underway, tried to get a question about U.S. citizenship on the census forms, a kind of question...

DEMBY: That's right.

WANG: Right - a kind of question that the Census Bureau's researchers have long warned that would discourage households with noncitizens from participating in a count, a constitutionally required count of every person living in the country in order to come up with the numbers that determine how many seats in the House of Representatives, as well as how many votes in Electoral College, each state gets. Plus, this is data that's used to guide an estimated $1.5 trillion dollars a year in federal funding, federal tax dollars, for Medicare, Medicaid - you name a public service, it's distributed in part guided by census data, probably.

DEMBY: In that cauldron of all these things you just outlined for us, the United States government still tried to carry out the census count. The census findings are finally out. You've had a chance to look at them. We've been seeing some really eyebrow-raising conclusions about those findings. In your opinion, what are some of the more, I want to say, questionable takeaways that you've been seeing people arrive at based on the latest census numbers?

WANG: Well, one of the first headlines I saw when the - what's known as the redistricting data, the second major set of census results where you get the information about race and ethnicity - when that came out, one of the headlines I saw was, quote, "America's White Population Shrank." And I get it. We're in the news business. You know, there's a limited character count with headlines but...

DEMBY: (Laughter).

WANG: ...That's a misleading headline. That's a headline that really just doesn't capture everything that's going on. This is complicated stuff. And when you - when you're talking about the, quote-unquote, "white" population, what these results show, it really depends on how we're defining white. If we're talking about people who checked off only the white box on the census form to answer the race question - and the size of this group and its share of the total U.S. population have dropped over the past decade, down almost 9%. But if you expand the white population to also include people who checked off white as well as one or more of the other racial categories, then the white population has actually grown since 2010, up almost 2%.

DEMBY: That is really fascinating. What is it about the finding of white population decline, like, what is it that we're missing in that story?

WANG: To just say this - the white population has declined oversimplifies all the dynamics here. There's a lot of things happening. I think, you know, just one point here, 1960 - it wasn't until 1960 until every household could self-report their racial identity, which means that for the majority of this country's history, this race data was really determined by government workers looking at someone and judging, oh, I guess you look white. And that's not necessarily how someone would have identified themselves. And so to compare 2020 data with this older data, that's - there's a lot of caveats there we have to keep in mind.

DEMBY: Hansi, I don't know if you remember this. But way back in the days - like, in the early days, of CODE SWITCH - we went to the Census Bureau together. And we sat with a bunch of the quants there. And they were explaining...

WANG: I remember that. I don't know if they call themselves quants. But, sure, yeah.

DEMBY: Oh, yeah. I'm sorry. If that's some sort of slur or some sort of...

WANG: (Laughter).

DEMBY: ...Derogatory term to people who count numbers and do all sorts of complicated, methodological stuff around census data, my bad, y'all. I'm trying. God is still working on me. But we went to the census together, the Census Bureau. And they were asking about a lot of these questions. And they were still, at that time, sort of trying to figure out what they were going to do with big questions like, what do you do with the Middle Eastern/North African population - right? - these people who are - historically, a lot of people have identified as white even though we've seen pushes for those people to sort of be disaggregated from the category that we call white. But also, the thing they underline was that, as you said, every single census survey, every 10 years, has different racial questions on it. Like, they're almost never the same ever. They've never been the same more or less, right? You can sort of look at the census and sort of tell a little bit about what was happening politically around race in the country, if you look at any sort of one, singular snapshot. So why is it so easy for us to - I'm saying us, media folks - to draw bad conclusions based on this data?

WANG: Part of it is because it's just so easy to take these numbers, this data, at face value, and to ignore all that context you laid out. But we have to keep in mind, data, especially census data, is produced. It's not just waiting out there in the wild for the Census Bureau to gather once a decade. It matters how questions are asked. It matters who does the asking. It matters what's happening when questions are asked. We have to remember, there was new write-in areas to allow people to give more detail about their racial and ethnic identities on the 2020 census forms. And the Census Bureau was able to process those responses. So it was writing responses in more detail partly because, for the first time, all households could do the census online. And all of these changes, the Census Bureau says, produced a more thorough portrait. You know, think of it like if the census wears glasses because it's nerdy like me - I wear glasses. And it got its prescription updated. And so now it could see the country's demographics more clearly because of these changes. It's not 20/20 vision, of course, because no census is perfect. But it's clearer. It's better. But that also means we have to be careful when trying to make any historical comparisons with what happened before, because the picture was fuzzier in a lot of ways.

DEMBY: That was the nerdiest possible metaphor, so kudos to you for that.

WANG: (Laughter).

(SOUNDBITE OF MUSIC)

DEMBY: All right, y'all, when we come back, we're going to get into some of the reasons why more people than ever have been checking multiple boxes on the race question. And spoiler alert, it's not because there's been some explosion of interracial coupling. But, you know, we're going to get into all of that after the break. Stay with us.

(SOUNDBITE OF MUSIC)

DEMBY: Gene - just Gene for now. CODE SWITCH.

(SOUNDBITE OF MUSIC)

DEMBY: We've been talking to NPR's Hansi Lo Wang. He's the correspondent who's trying to make sense of the census. And Hansi was saying that all those headlines about how the U.S.'s white population was declining, yeah, that's not really what's happening. The actual story of people identifying as white is harder to sum up that neatly. So I asked if there were any other census numbers that have been reported on that he had been side-eyeing. And he told me, one thing he's noticed is that the category of people who check off more than one race is often sidelined or ignored.

WANG: And so this is another pet peeve of mine as a census reporter because...

DEMBY: OK.

WANG: ...This 2020 census is telling us, according to this data, the country is becoming more multiracial. And, you know, we also have to keep in mind, being able to choose more than one racial category in the census is a relatively new thing for the U.S. Census. I talked to Jenifer Bratter, sociologist at Rice University, who studies multiracial identity. And she told me we should keep in mind that the U.S. has been multiracial from the very beginning.

JENIFER BRATTER: I think there's something in our American racial consciousness that needs to cast it as new. But it's absolutely true that we have historical records, demographic data and just general sense of history that tells us that mixed-race people have always been a part of the American landscape, just as a desire to understand the nuances of this group as the exception has also always been a part. But it's very, very true that mixed-race communities are a persistent feature of the American racial landscape.

WANG: And we have to remember, being able to check off more than one box for the race question, that only started with the 2000 census. So in that context, sometimes I wonder, how much is the U.S. really becoming more multiracial, versus how much is the United States government allowing people more of an opportunity to say that they're multiracial in the census? And, you know, some people may be hearing, oh, you know, the U.S. is becoming more multiracial. This must be because there are more babies being born to parents who identify with racial groups that are different from one another. But Jenifer Bratter told me that it's - you know, that's likely part of it. But it doesn't explain why the multiracial population almost quadrupled over the past decade.

BRATTER: It seems, with this type of increase, that it's more than likely a shift in how people are identifying, as opposed to an introduction of more people who identify as multiple - with multiple racial categories. We know that racial identification can shift and often does shift. And for the multiple-race population, the shifting is even more likely.

WANG: You know, one of the changes the Census Bureau made to the census forms last year was - I mentioned earlier - these new write-in areas. They were specifically under the Black and white check boxes for people with non-Hispanic origin. So people had even more space to give more detail about their identities. And there were prompts under white for German, Irish, English, Italian, Lebanese, Egyptian. Those are examples of what you could write in. And again - important to keep in mind - the federal government's standards consider people with roots in the Middle East or North Africa officially categorized to be white under the official guidelines. And under the Black category, there were prompts for African American, Jamaican, Haitian, Nigerian, Ethiopian, Somali. And the bureau also changed how it sorted through those responses. And it actually was able to count more characters that people wrote in. So a lot of that detail now is represented in this data. And it likely contributed to more people being counted as multiracial.

(SOUNDBITE OF MUSIC)

WANG: And, you know, there's another factor I've been tracking that I know CODE SWITCH has also done reporting on.

DEMBY: Yep. I think I know where you're going.

WANG: Cheaper DNA tests.

DEMBY: So cheaper DNA tests allow people to have a more sort of granular idea of, like, which parts of the world they have ancestry in, right? But of course, like, what you're saying here is, like - all right, so under the category white, you have categories like Lebanese and Egyptian of, like - as examples on the U.S. Census. But as anyone who's been in the United States over the last 20 years and seen the way, like, Arab Americans have been treated and racialized, I imagine a lot of people don't consider themselves white at all, and don't feel like they're having very white experiences at all.

And then you have this DNA test question, which is about this stuff that may not even matter in the way you move through the world. Like, if you find out that you have some Black ancestry but everyone in your family is white, identifies as white, you've never known them, been known to be anything but white - or as often happens, probably even more often, is, like, having indigenous ancestry. This is why race in such bullshit - right? - is that there's this disconnect in so many different directions between the way people, like, live race and the way we capture it. All these things are sort of flying at each other in different directions.

WANG: Yeah. I mean, this is part of the reality in the United States now, that taking DNA ancestry tests have become more common over the past decade. And, you know, regardless of the caveats, some people, it matters a lot to them. And it is helping to shape how some people are reporting their racial identities on the census. I talked to two people who took DNA ancestry tests. And they checked off more than one box. I talked to Jonathan Marshall (ph) from Fort Myers, Fla. - grew up in Trinidad, told me he never singularly identified with one racial group. He has grandparents from Barbados, Spain, the U.K. So before he took the test, he already was planning to mark more than one box. But those test results he got changed his understanding of his family tree.

JONATHAN MARSHALL: My mom's mom, she's of Asian descent, which previously I thought was Japanese but turns out - after doing the ancestry test, turns out she's Chinese. So all of that is just - I'm just a product of all of that mixing.

WANG: And Jonathan said that without those DNA test results, he probably would have checked off another Asian group. And I also talked to Jill Marie Maldonado (ph) of Manchester, Conn.

JILL MARIE MALDONADO: Especially as a Puerto Rican, we don't really have a lot of information about who we are and our ancestry and all of that. It's sort of kind of given that you're Puerto Rican, and that you have African ancestry, Spanish ancestry and Indigenous ancestry, but it's sort of, like, that myth that they gave us. And as I was learning about things - like, for example, there was an Italian immigrant that came to the island. There was British immigrants. There were immigrants from a lot of different, like, groups and things like that. I wanted to know where my breakdown was, just to get to know a little bit more of that, the people that kind of survive all the odds to get me here, you know what I mean?

DEMBY: Yeah, we've talked a lot about Puerto Ricans and the mythology around Puerto Rican racial identity in Puerto Rico and how complicated that is and how that sort of thing that being codified by governments in Puerto Rico and that's not - but what's actually happening again on the ground with people's identities is much more textured and nuanced and, as Shereen might say, complicated.

WANG: It is complicated, and it got so complicated in Jill Marie Maldonado's home that there was actually a family fight.

DEMBY: Ooh, drama.

WANG: When the census came around, Jill Marie brought up these DNA test results with her mother, Maribel Rodriguez (ph), who's also taken a DNA ancestry test.

MARIBEL RODRIGUEZ: She was saying that we shouldn't be saying that we were white because we were mixed.

MALDONADO: I refuse to just check the white box.

RODRIGUEZ: So that was kind of, like, well, I always marked the white, and she was like, well, we're not.

MALDONADO: 'Cause it would deny those people that came here to the island and survived all the odds against them for me to be here.

RODRIGUEZ: In Puerto Rico, when I was growing up, we were all white.

MALDONADO: She was like, I'm not Black. And I'm like, you have African ancestry in you. I've seen your DNA breakout.

RODRIGUEZ: She can be very convincing.

DEMBY: But also, I mean, that's - again, not to, like, hammer home the, like, sort of messiness, the real life lived experience of it, it's like, if Blackness is, like, a political category - right? It's not just a racial category - if all these things are political categories, and she has been white her whole life, and she has not lived as a Black person, does some Black ancestry, as determined by an ancestry test, like, make her Black? I don't know. I don't know the answer to that question, honestly.

WANG: It's complicated, to quote Shereen.

DEMBY: Yes, absolutely.

WANG: And when you're looking at the census form, it's asking you to self-report. The way that Jill Marie Maldonado reported the race for herself and her mother was American Indian, Black and white. And so they were among the nearly 34 million people counted in last year's census as identifying with more than one race. But the questions you've brought up really brings home what I've been talking to with a lot of sociologists who focus on the effect of these genetic ancestry tests. And many of them are worried about how this can affect census data, data that is used to enforce civil rights laws, measure racial disparities. You know, many people who buy these DNA ancestry tests for themselves identify as white but not Latino, and they often want to claim a certain ancestry, often Native American ancestry. And this is a finding from a sociologist, Wendy Roth at the University of Pennsylvania.

WENDY ROTH: What these companies are telling you is that you have a likelihood of having a certain amount of your genetic markers associated with certain populations more often than others. So there's a lot of caveats there that are not always clear to test-takers. I generally suggest to people that the information you get from taking a genetic ancestry test is one piece of information that you can put together with everything you know from your family, from your historical research, from your genealogical information. But in and of itself, it doesn't tell you your race or ethnicity.

DEMBY: Right, because race is not scientific. There's no such thing as race that happens in the, like, natural world, right? We're making these categories up all the time. And so telling you that you are, I don't know, 75% sub-Saharan African - it might tell you that you're Black, right? It probably tells you that you're Black but doesn't necessarily tell you that you're Black. Like, that's not how any of this stuff works. That's not how any of this works, Hansi.

WANG: Cat's out of the bag, Gene.

DEMBY: (Laughter).

WANG: DNA ancestry tests are out there, and people are making their own interpretations on the census.

DEMBY: Right. And of course, like, all these categories are using Black, Latino, white - like, they all mean certain things in the U.S. context. They don't necessarily mean anything in another context, if you were taking the same DNA test and you, like, lived in a different country with a different racial imaginary, to use a term that we hear a lot from smart race people on the show. Like, while this stuff is also really important to, like, the way the government allocates money and the way we live our lives, it also is complete nonsense (laughter) and doesn't make any sense.

WANG: And there's also a fine line of - you know, the Census Bureau has to follow federal standards that say the best way of collecting race ethnicity data, basically, is through people self-identifying. That leaves a lot of room for everyone to make their own interpretation in a lot of ways, certainly influenced by society, their personal experiences, the political climate, the social climate. But it is self-identification. That's the standard that the federal government - that the census uses.

DEMBY: So what does this box of people in this multiracial category, according to the census, actually look like? Are certain combinations of different racial identities more common than others when we're talking about multiracial people on the census?

WANG: Well, if you look at people who did not identify as Latino, the three largest combinations were American Indian or Alaska Native and white, Black and white and an Asian group and white. And if you look at people who identified as Latino, the three largest combinations were some other race and white, Black and some other race and American Indian or Alaska Native and white. And, you know, it's easy to list off these combinations. But of course, there's only so much that data can tell us. I talked to sociologist at University of Kent Miri Song, who is co-writing a book about multiracial people of white ancestry in the U.S. And she says, we have to keep in mind that there is a lot of nuance that even that level of detailed data doesn't capture.

MIRI SONG: When people tick boxes, you can - OK, you can have a box that says, I am - what? - Black and white and maybe Asian, 'cause you can tick multiply in the U.S. The question is, what does it mean? I mean, I think the really interesting question for me is how five people who tick the same two or three boxes on the census can have very, very different lives. Some of them may have white spouses. Some of them may have, you know, non-white spouses. Some may have co-ethnic spouses with whom they share an ethnic heritage. So, you know, how does the choosing of categories translate into how you live your life? And for that answer, you can only get that through talking to people.

(SOUNDBITE OF MUSIC)

DEMBY: So Hansi, all this stuff is incredibly fascinating and really complicated. We've - obviously, we as people who are in the news have been taking a lot of the wrong conclusions away from these very messy numbers. But what's a number that you've looked into from your reporting on the census that isn't getting the attention you think it should be getting? Like, are there more nuanced or subtle results that are actually for - like, a real big deal that we're just missing?

WANG: I got a number for you. You ready?

DEMBY: OK. Yes. I want to hear it.

WANG: Forty-nine million, nine-hundred-two thousand, five hundred thirty-six.

DEMBY: Forty-nine million, nine-hundred-two thousand...

WANG: Five hundred thirty-six.

DEMBY: ...Five hundred thirty-six.

WANG: Or close to 50 million.

DEMBY: OK.

WANG: And that is the size of the second-largest racial group in the United States after white, according to the 2020 Census.

DEMBY: And which group is that?

WANG: That group is called some other race.

DEMBY: (Laughter) OK. OK.

WANG: Mysterious, right? Some other race. But here's the biggest takeaway.

DEMBY: OK.

WANG: More than 90% of this group, some other race, is Latino.

DEMBY: OK.

WANG: And we have to remember that the federal government considers Hispanic or Latino identity to be an ethnicity and not a race. And so that means that for a lot of people who identify as Latino, when they get to the census race question, it feels just impossible to answer.

I talked to two people who said they really had a hard time answering it - Leilani Garcia Torres (ph) in Brooklyn, N.Y. and Frank Alvarez (ph) from Los Angeles, Calif. And for the 2020 census, both of them identified as Hispanic or Latino.

LEILANI GARCIA TORRES: Both of my parents are from the island of Puerto Rico.

FRANK ALVAREZ: Growing up, you know, we were in a very traditional Guatemalan home.

GARCIA TORRES: We're just historically pretty mixed.

ALVAREZ: I think I just identify with my ethnicity.

GARCIA TORRES: If you look at anyone in my family, you wouldn't really be able to guess a race. We just look vaguely tan, I would say.

WANG: They said they got to the race question, each individually, and they looked at the boxes - American Indian or Alaskan Native, boxes for Asian, Pacific Islander groups, Black, white. None of those boxes really fit them, they said. And so they ended up checking off some other race. A lot of Latinos just feel like the census race question, the categories that are used just do not reflect how they see their racial identity. This has been a long-running issue. In fact, it's a major data problem...

DEMBY: OK.

WANG: ...Because we really don't know who this second-largest racial group in the United States is and how they really identify.

DEMBY: Right.

WANG: It is in many ways now because it is now the second-largest racial group - some other race - a sign of the flaws in how this census captures data about race and ethnicity. And it raises the question, how full, accurate, thorough of a portrait of the country did the 2020 census capture?

(SOUNDBITE OF MUSIC)

DEMBY: Hansi Lo Wang is an NPR reporter trying to make sense of the census and is a CODE SWITCH OG. Thank you for coming back on the show. We always love having you here, man.

WANG: You're very welcome.

(SOUNDBITE OF MUSIC)

DEMBY: That was NPR's census expert, Hansi Lo Wang.

All right, y'all, that's our show. We want to hear from you. You can email us at codeswitch@npr.org or follow us on Twitter and IG - @NPRCodeSwitch. You can subscribe to our newsletter by going to npr.org/codeswitchnewsletter. This episode was produced by Alyssa Jeong Perry. It was edited by Leah Donnella. And a shout out to the rest of the CODE SWITCH squadron - Karen Grigsby Bates, Kumari Devarajan, Christina Cala, Jess Kung, Sam Yellowhorse Kesler, Summer Thomad and Steve Drummond. Our art director is LA Johnson. Our intern is Aja Drain.

I'm Gene Demby. Be easy, y'all.

Copyright © 2021 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.