New York Times Piece on Hispanics and Census Based on Study Not Yet Finalized or Public

May 22, 2014
5:24 PM

Yesterday, The New York Times’ “Upshot” blog announced the following to the world: “More Hispanics Declaring Themselves White.” The piece, written by Nate Cohn, included the animated image of a figure shifting from silver grey to ivory white and this lede paragraph: “Hispanics are often described as driving up the nonwhite share of the population. But a new study of census forms finds that more Hispanics are identifying as white.”


Such a definitive statement in the opening paragraph of a NY Times piece with no clear attribution plus sweeping generalizations about Latinos made throughout Cohn’s post raised several eyebrows, especially from our social community and specifically from Julio Ricardo Varela, this site’s founder. Granted, Cohn’s next paragraph made mention of “research at an annual meeting of the Population Association of America [PPA] and reported by Pew Research” which had yet to be published, but to many, the journalistic damage had already been done.

Now emails and tweets about the story confirm that Cohn summarized the summary that Pew wrote about a study that does not yet exist, and he never was able to talk with researchers who are still working on their findings—data that is still getting reviewed, analyzed and possibly changed.

When asked today by Varela via email whether there was a public version of the study/presentation given at PPA, one of the study’s co-authors, Assistant Professor Carolyn Liebler of the University of Minnesota, wrote: “At this point, there is not a public version of the study or presentation that we are ready to share with the press. This is because it is an ongoing study and we expect the numbers to change as we refine our measures and data selection criteria.”

Furthermore, Liebler said that even though the math cited in Cohn’s piece was correct (“2.5 million Americans of Hispanic origin… changed their race from “some other race” in 2000 to “white” in 2010; “an additional 1.3 million people switched in the other direction.”), the rest of the conclusions in his post did not accurately represent the study’s initial findings nor did Cohn present a complete picture of her group’s current research.

The New York Times post included two numbers from our study that are correct: 2.5 million and 1.3 million. Our paper does not involve any interpretation of why people are changing and the inferences about that are entirely his. Also, the New York Times post focuses only on one aspect of the paper, rather than the full set of results. We plan to post a Census Bureau blog with a few more details soon and then the public paper will be coming out in late summer, we expect. We will send you both when they are ready.

On Twitter, the author of the Pew piece, D’Vera Cohn, tweeted clarifications about the census study:

Liebler did say via email that “D’Vera Cohn of Pew attended PAA and correctly reported quotes from the PAA presentation, which she attended in person. She did not receive a copy of the presentation.” In addition, Liebler also added: “New York Times did not attend PAA (as far as I know) and we have not given them any materials (presentation, paper, etc) or done an interview with them.”

On May 6, D’Vera Cohn said this to journalist Richard Prince: “This is preliminary data, so it’s a heads-up to watch for more numbers—and a more complete narrative explanation—in the next few months. These are numbers, but each one is about a person with a story to tell. So the data could be a jumping-off point for journalists to interview folks about how they identify, and why.”

Varela also sent an email to “Upfront” editor David Leonhardt asking whether Cohn had contacted Liebler or any of the other researchers of the project before writing his piece. Later in the day, Varela shared Liebler’s comments about Cohn’s piece with Leonhardt. As of this posting, Leonhardt has yet to respond.

This tweet to Cohn has also gone unanswered:

Interestingly enough, Cohn’s post cited a 2013 NYTimes op-ed called “Hispanics, the New Italians,” written by Leonhardt, identified on the “Upfront” post as the editor for Cohn.

Two years before Leonhardt’s opinion piece, the Times published a story that easily challenges what both Cohn and Leonhardt had written. The first one, “Hispanics Identifying Themselves as Indians,” stated the following:

The trend is part of a demographic growth taking place nationwide of Hispanics using “American Indian” to identify their race. The number of Amerindians — a blanket term for indigenous people of the Americas, North and South — who also identify themselves as Hispanic has tripled since 2000, to 1.2 million from 400,000.

In 2012, another Times piece, “For Many Latinos, Racial Identity Is More Culture Than Color,” stated this:

More than 18 million Latinos checked this “other” box in the 2010 census, up from 14.9 million in 2000. It was an indicator of the sharp disconnect between how Latinos view themselves and how the government wants to count them. Many Latinos argue that the country’s race categories — indeed, the government’s very conception of identity — do not fit them.

The main reason for the split is that the census categorizes people by race, which typically refers to a set of common physical traits. But Latinos, as a group in this country, tend to identify themselves more by their ethnicity, meaning a shared set of cultural traits, like language or customs.

So when they encounter the census, they see one question that asks them whether they identify themselves as having Hispanic ethnic origins and many answer it as their main identifier. But then there is another question, asking them about their race, because, as the census guide notes, “people of Hispanic, Latino or Spanish origin may be of any race,” and more than a third of Latinos check “other.”

As for the Times? One Latino academic has questions.

“How does the New York Times writing an article based on a blog post of a presentation of a preliminary study no one has seen and without citing any other experts (or the authors themselves) help advance any science, any debate, or any public understanding of anything? The NYT article ended up obfuscating much more than it helped clarify,” said Hector Cordero-Guzman, Ph.D. from Baruch College-CUNY. Cordero-Guzmán was one of the first people on Twitter to ask Cohn the question about whether he had the actual research with him when he wrote his piece.

UPDATE, 8:15pmET, May 22, 2014: Tonight Liebler sent another email, “Nate Cohn did reach out to us to talk, but we were not able to talk as quickly as he needed to get his article to press, so it didn’t happen.” Liebler then added that Cohn had contacted her on May 8 but one of the key principals of the study was sick and the group decided to “not to go forward with interviews about this preliminary research” until that individual was available.