The KFF/LA Times Survey of Immigrants was conducted April 10 through June 12 online, by telephone and by mail among a nationally representative sample of 3,358 American adults born outside the U.S. It was conducted in 10 languages: English, Spanish, Chinese, Korean, Vietnamese, Tagalog, Haitian-Creole, French, Portuguese and Arabic.
Respondents were reached through one of three samples: an address-based sample, a random digit dial (RDD) telephone sample of prepaid (pay-as-you-go) cellphone numbers, and callbacks to telephone numbers that were previously randomly sampled for RDD surveys and identified with speakers of a language other than English or Spanish.
The address-based sample was invited to participate through multiple mail invitations. In addition, interviewers attempted calls to telephone numbers that were matched to sampled addresses. The sample was divided into areas (strata), defined by census tract, based on the incidence of immigrants among the population overall and by countries of origin. To increase the likelihood of reaching households with immigrant adults, strata with higher incidence of immigrants overall, and of certain countries of origin in particular, were oversampled.
Respondents from all three samples were asked to specify their country of birth and qualified for the survey if they were born outside the U.S.
Respondents received an incentive of $10-$25 via check or gift card for completing the survey. The online questionnaire included two questions designed to establish that respondents were paying attention, and cases were monitored for data quality.
The combined sample was weighted by sex, age, education, race/ethnicity, census region, number of adults in the household, presence of children in the household, homeownership, time living in the U.S., English proficiency and U.S. citizenship to match the characteristics of the U.S. adult immigrant population, based on data from the Census Bureau’s 2021 American Community Survey (ACS).
Weighting was done separately for each of 11 groups defined by country of origin (Mexico, China, Other East/Southeast Asia, South Asia, Europe, Central America, South America, Caribbean, Middle East/North Africa, Sub-Sahara Africa, all others). The overall sample was also weighted to match the share of U.S. adult immigrants from each country/region of origin group.
The margin of sampling error including the design effect for the full sample is plus or minus 2 percentage points. Margins of error are larger for sub-samples.
Discussion about this post