Interviewing Linguistic and Cultural Minorities: Insights from Ethnographic Observation of 2010 Census InterviewsView Presentation
*Yuling Pan, U.S. Census Bureau
Keywords: survey interviews, linguistic and sociocultural factors, communicative interaction, questionnaire design
As immigration continues to diversify the US population, linguistic and cultural differences pose growing challenges to survey methodologists. This paper highlights how the standardized in-language survey interview fails to fully address the ways in which minority groups interpret the survey encounter as a particular form of social interaction within different minority communities; or how these culturally-specific understandings influence communication within survey encounters. This paper draws upon a unique body of ethnographic data, consisting of 546 observations of Non Response Follow Up interviews conducted during the 2010 Census amongst Spanish, Chinese, Korean, Russian, Vietnamese, Arabic, Portuguese, and English-only speakers, to analyze how linguistic and socio-cultural differences influenced communicative interaction during the survey process. Following these initial observations additional debriefing interviews were conducted with the respondents and their enumerators. Our discussion of findings focuses on: 1) linguistic factors that affected communication —including forms (e.g. verbal vs. written, on-the fly vs. pre-scripted) and quality (e.g. verisimilitude to authorized scripts) of translations and interpretations, as well as issues of conceptual equivalence (or lack thereof) across different languages that affected how particular questions were understood; and 2) socio-cultural factors that shaped social interaction during the survey encounter—including gender, socio-political identity (e.g. national, regional, ethnic), and views of government and surveys based on previous experience.
Our conclusions highlight several fundamental methodological implications for standardized surveys conducted within linguistically and culturally diverse populations and outline key issues that frameworks concerned with conceptual equivalence must account for in the design of questionnaires and of survey protocols for such populations.