Periodically our readers ask us why we don’t provide individual data for Asian Americans/Pacific Islanders as part of the standard demographic comparisons in our reports. We hope this article sheds light on some of the challenges of reporting survey results from some populations and helps explain why we do not report Asian-American figures in our stand-alone demographic reports.
What are the standard racial/ethnic groups that you include in your demographic tables?
In general, all of our reports contain demographic comparisons for whites and African Americans. We also include findings for Latinos on surveys with a Spanish-language interview option (all of our standard tracking surveys now include Spanish-language interviews). For a time in the early- and mid-2000s we reported figures for English-speaking Latinos, but in recent years we have only reported on Latinos for when we conduct surveys in English and Spanish. From time to time we participate in omnibus commercial surveys where we purchase questions and we do not report results for Latinos because those surveys are only conducted in English.
Why don’t you report on Asian Americans within your standard demographics?
Several years ago we asked our lead pollster, Evans Witt (principal and CEO of Princeton Survey Research Associates International) to provide a survey methodologist’s take on this question. His response hopefully sheds some light on the challenges associated with polling the Asian population in the U.S.:
“The short answer is that Asian Americans make up a very small slice of the population, 3.7 percent in the 2000 Census (Editor’s note: In the more recent 2010 Census, Asian Americans make up around 5.6 percent of the national population). In addition, for a good portion of that population, there are complex language barriers…and language barriers reduce the number of completes with the non-English speaking minorities (Editor’s note: A recent Pew Research survey found that 64% of all Asian Americans—and 53% of those not originally born in the United States—speak English “very well”). The diversity of the Asian American population and the languages they speak makes offering interviews in those native languages very difficult and very, very expensive.”
To use a specific recent example, our November 2012 national tracking survey contained 54 individuals who identified themselves as Asian or Pacific Islander out of a total of 2,261 respondents. The margin of error for a group of that size is a whopping +/-15.5 percentage points, which is far too large for us to have confidence in the accuracy of our findings. Moreover, none of the 54 respondents include non-English-speakers, meaning that our findings would not be representative of the broader Asian-American population as a whole.
So are Asians included in your findings?
Yes, Asian Americans are included in our findings. Our surveys are representative of the entire adult population of the United States. They accurately account for the full population’s full diversity by age gender, racial and ethnic groups, regional representation, and socio-economic factors such as education levels, household income, employment status and even the differences tied to landline telephones and cell phones. We do not exclude anyone from our analysis based on their demographic characteristics. The 54 Asian Americans in the November survey cited above were included in the overall analysis of all the questions we asked. We simply are not able to report on Asian-Americans as a distinct group due the limitations noted above. Still, their responses (indeed, all of the responses from our surveys) are incorporated into the general population figures we report.
Do you count Asians as white, or include them in some other racial/ethnic group?
No, Asian Americans are not folded into any other racial/ethnic group when we report on those groups individually. The same applies to Native Americans, those who identify as mixed race, and those who either don’t know or refuse to answer our race/ethnicity questions. All of these groups are included in our general population figures, but are not reported individually.
Do you have the same issue related to reporting statistics among other groups?
Any small or low-incidence group is challenging to isolate in a national survey with our standard sample size. That is especially true when those groups have complex language barriers, or other issues (such as hearing impairments) that make it difficult to complete a telephone survey.
Where can I find other information on Asian-American technology use?
In the past we have combined multiple surveys to produce a pool of Asian-American respondents large enough to report individually on this group (although this analysis is limited to the English-speaking portion of the Asian population). A recent example is this presentation by Lee Rainie: http://www.pewinternet.org/Presentations/2011/Jan/Organization-for-Chinese-Americans.aspx.
A U.S. Commerce Department division called the National Telecommunications & Information Administration has sponsored a special internet use supplement to the Census Bureau’s Current Population Survey. That survey is large enough to provide robust national data for both Native Americans and Asian Americans. Their most recent findings are available at http://www.ntia.doc.gov/files/ntia/publications/exploring_the_digital_nation_computer_and_internet_use_at_home_11092011.pdf
Another potential resource is a recent survey by the Public Policy Institute of California. While not national in scope, it does provide racial/ethnic comparisons for Asians in California. The PPIC survey can be found at http://www.ppic.org/main/publication_show.asp?i=263
For researchers with a specific interest in young adults, Eszter Hargittai's surveys of incoming University of Illinois-Chicago students offer another valuable resource. Conducted in 2007, 2009 and 2010, these surveys also offer a large enough Asian-American respondent pool to enable robust analysis within this group. For more information, see http://webuse.org/pubs/.