What’s lacking in the Pew Research Center Report on Asian Americans

George and Johnny Huynh’s story last year complicated the model minority myth.

I hate to get too academic on this blog, so please forgive me as I enumerate my issues with the Pew Research Center Report on Asian Americans that is circulating this morning. I will try to be as brief and clear as possible.

This report presents an overly optimistic picture of the state of Asian America. I think it’s fantastic that their research sample was better educated and happier with life than the typical American. However, we must look more closely at who they talked to in order to obtain these results.

The Pew Research Center survey was designed to contain a nationally representative sample of each of the six largest Asian-American groups by country of origin—Chinese Americans, Filipino Americans, Indian Americans, Vietnamese Americans, Korean Americans and Japanese Americans. Together these groups comprise at least 83% of the total Asian population in the U.S.

I am no expert on research survey design, but we might note that the Asian American groups with the highest poverty rates (namely Cambodians, Hmong, and Laotians) are not included in this survey. According to C. N. Le’s analyses of Census 2000 data, Cambodians, Hmong, and Laotians have poverty rates similar to those of blacks and Latinos. 22.5% of Cambodians, Hmong, and Laotians in the US lived in poverty, compared to 24.9% of blacks and 21.4% of Latinos.

Much of this difference, of course, is related to how these groups came to the US. As the Pew report points out:

About half of all Korean and Indian immigrants who received green cards in 2011 got them on the basis of employer sponsorship, compared with about a third of Japanese, a fifth of Chinese, one-in-eight Filipinos and just 1% of Vietnamese. The Vietnamese are the only major subgroup to have come to the U.S. in large numbers as political refugees; the others say they have come mostly for economic, educational and family reasons.

While the other groups were either stringently selected from the highly skilled, highly educated sectors of their native countries or came through family reunification, large swathes of Vietnam’s population fled the country during the Vietnam War. The same could be said about the Cambodians, Hmong, and Laotians. We must also note that the Cambodian elite was one of the primary targets of the Khmer Rouge genocide, and that many of the Cambodians who made it to the US were of low socioeconomic status.

Thus, by taking a sample of the largest Asian groups in the US, most of whom were highly skilled upon arrival and more or less easily moved into the US middle classes, Pew was able to find that Asian Americans are better off, better educated, and happier than the typical American. This report completely overlooks differences between the groups selected for the study and the smaller, poorer, less educated, and likely less happy groups that were excluded.

It also overlooks differences within groups. For example, as a result of family reunification, undocumented immigration, and the inclusion of refugees from the Vietnam War, Chinese Americans fall into two distinct, large clusters: the low-skilled, poorly educated working class and the high-skilled, highly educated middle class. Le’s analysis hints at this, showing that 23.6% of Chinese have less than a high school education, the largest proportion of the groups sampled for Pew’s study.

By looking only at the groups that are doing the best and not breaking down statistics within these groups, Pew’s study ends up reifying the model minority myth. Their biased aggregate measures say that Asian Americans are doing great. But just ignoring the significant differences between and within groups in terms of poverty, education, and other measures does not make these differences disappear! I worry that by hastily sweeping these differences under the rug, those groups that need help will not be able to get it. After all, if Asian Americans as a whole are doing well, then why bother helping them?