News

CityRoom Blog – By Fernanda Santos

The Vietnamese version of the 2010 census questionnaire used the words “dieu tra” to describe the population tally — but what the words really conveyed was something like a communist government investigation. On Korean forms, “county” was translated into “nation.” And the Chinese hot line offered information only in Mandarin at first, although elderly Chinese immigrants, who are less likely to be proficient in English, speak Cantonese.

Then there was this: In some questionnaire assistance centers in Queens, where people could get help filling out their forms, census workers instructed countless Bangladeshi to check “Indian” as their ethnicity.

“To have someone identified as something they’re not is not only offensive, but it also drives a misallocation of resources and representation,” said Glenn D. Magpantay, director of the democracy program at the Asian American Legal Defense and Education Fund, an advocacy group based in Manhattan.

So it goes, as the 2010 census tries to count and record New York’s richly and often complexly mixed neighborhoods. There have been mistakes.

Mr. Magpantay and his colleagues were prepared. They wanted to be sure that, among other things, Asians would be counted appropriately — Indians as Indians, Bangladeshi as Bangladeshi, for example — and that the language on the forms matched the nuances of the words used in immigrant households.

The Census Bureau is preparing to send out census takers to households that didn’t return the forms, and they are training the takers to be especially attuned to these cultural differences — and subtle word meanings.

It has tried to hire immigrants as takers and place them in the communities they represent. But the lines and the boundaries aren’t going to be 100 percent clear all the time. Case in point, Mr. Magpantay said, is the Guyanese man who has been assigned to knock on doors in Maspeth, an overwhelmingly white neighborhood in Queens, even though there is a huge Guyanese community in nearby South Richmond Hill.

Lester A. Farthing, New York’s regional census director, said the plan was to match heavily immigrant neighborhoods with census takers who speak the predominant languages there and understand its cultural nuances. That, he said, can go a long way into making people comfortable sharing personal information with the stranger at the door.

Should census takers encounter a person whose language they do not understand, they are to show this person a language card to see whether the language can be identified. Someone who is fluent in that language will contact the home later to finish the job.

Mr. Farthing acknowledged that, with 30,000 questionnaire assistance centers providing help in 60 languages nationwide, mistakes were bound to happen. Still, he took issue with the misinformation given to Bangladeshi in Queens, saying that the centers where the problem happened were staffed by people who receive two days of training at best.

“We tell them that the census is all about self-identification, but there are people who come in and ask, ‘What box do I check here?’” Mr. Farthing said. “The concept of picking a race or an ethnicity for non-American-born folks is not common.”

Mr. Magpantay said that no one knows how many Bangladeshi identified themselves as Indian in the questionnaires, but that the undercount could have meaningful consequences. “If we don’t know the real size of the Bangladeshi population in a neighborhood, how are we going to know if the local hospital needs to have signs in Bangla?” he asked.

He said his organization was exploring legal options to see whether the forms could be tracked and the mistakes corrected. And that they will continue to monitor the operation once the census comes knocking.

https://cityroom.blogs.nytimes.com/2010/04/25/census-outreach/