Survey of Americans on Race

 The CNN/Kaiser Family Foundation  Survey of Americans on Race was conducted August 25-October 3, 2015, among a nationally representative, probability-based random digit dial telephone sample of 1,951 adults ages 18 and older, living in the United States, including Alaska and Hawaii (Note: persons without a telephone could not be included in the random selection process). Computer-assisted telephone interviews conducted by landline (715) and cell phone (1,236, including 771 who had no landline telephone) were carried out in English and Spanish by SSRS.  CNN and the Kaiser Family Foundation both contributed financing for the survey, and teams from both organizations worked together to develop the survey questionnaire and analyze the data.  Each organization is responsible for its content.

To ensure there were enough respondents to capture the views and experiences of Black and Hispanic Americans, the full sample includes additional interviews with randomly selected African Americans and Hispanic Americans (commonly referred to as an “oversample”), for a total of 500 Black and 501 Hispanic respondents, although results for all groups have been adjusted to reflect their actual national distribution (See weighting description below). To oversample Blacks and Hispanics, the sample plan consisted of three elements: 1) general population respondents were reached by random digit dialing landline telephones or cell phones (1401 total, including 215 Blacks, 236 Hispanics); 2) respondents were reached by RDD landline or cell phone, and interviewed only if they were Black (186) or Hispanic (187); 3) prescreened respondents were reached by calling back phone numbers where respondents previously interviewed indicated that a person answering that number was either Black (100) or Hispanic (77). Both the RDD landline and cell phone samples were designed to target areas with high incidences of Black and Hispanics, and all phone numbers matched to a directory listing with a Hispanic surname were defined as a separate sampling group and oversampled. Both the RDD landline and cell phone samples were provided by Marketing Systems Group (MSG).

A multi-stage weighting process was applied to ensure an accurate representation of the national adult population. The first stage of weighting involved corrections for sample design, including a correction for the oversampling of telephone exchanges known to have higher incidences of Blacks and Hispanics, the likelihood of non-response for the re-contacted sample, and an adjustment to account for the fact that respondents with both a landline and cell phone have a higher probability of selection. The second weighting stage was conducted for each racial group separately, weighting to match estimates for the national population using data from the Census Bureau’s 2014 March supplement of the Current Population Survey (CPS), and a distribution of phone use was estimated based on the CDC’s National Health Interview Survey (NHIS). The weighting parameters used were age and gender, race/ethnicity, education, marital status, census region, population density of the respondents’ county, and telephone use. The sample of Hispanic respondents was weighted by nativity and national family heritage (Mexican, Puerto Rican, Cuban, and all other countries).  Each group was weighted to reflect the population distribution of race among the U.S. adult population.  All statistical tests of significance account for the effect of weighting.

The margin of sampling error including the design effect for the full sample is plus or minus 3 percentage points; for results based on African American and Hispanics it is plus or minus 6 percentage points. Numbers of respondents and margins of sampling error for key subgroups are shown in the table below. For items based on half samples, including half samples of only Whites or half samples of those of races other than Black or Hispanic, the MOSE for the total is +/- 4 percentage points and the MOSE for Whites is +/- 6 percentage points. For half samples of Blacks or Hispanics, the MOSE is typically +/- 8 percentage points for each group. For results based on other subgroups, the margin of sampling error may be higher. Sample sizes and margins of sampling error for other subgroups are available by request. Note that sampling error is only one of many potential sources of error in this or any other public opinion poll. Kaiser Family Foundation public opinion and survey research is a charter member of the Transparency Initiative of the American Association for Public Opinion Research.

The full CNN/KFF poll results are representative of the U.S. adult population, including people of all races. The reported results focus on the total sample, reflective of all adult people in the U.S., as well as Whites, Blacks, and Hispanics. While the responses for some smaller groups of the US population, for example, Asian-Americans and people of mixed race, are counted in the “total” poll responses, because adults in this poll were sampled randomly, the number of respondents who identify as Asian or as mixed race was fewer than 100, and therefore too small to report separately. With fewer than 100 respondents, the margin of sampling error around any poll result would be so large that the result would be unreliable and potentially misleading.