Cultural Context and Effect on Survey Responses: Findings and Recommendations from the 2010 Census Bilingual Questionnaire Cognitive Testing
Sha, M., Quiroz, R. S., Caspar, R. A., & McAvinchey, G. (2007, May). Cultural Context and Effect on Survey Responses: Findings and Recommendations from the 2010 Census Bilingual Questionnaire Cognitive Testing. Presented at American Association for Public Opinion Research Conference, Anaheim, CA.
In preparation for the 2010 Census, the U.S. Census Bureau has developed plans to test a self-administered bilingual “swimlane” questionnaire (Questions are grouped visually by placing them in vertical lanes and divided by parallel lines, with one lane for English questions and the other lane for Spanish). The latest effort involves structured in-depth cognitive interviews with monolingual and bilingual Spanish speakers, as well as monolingual English-speaking respondents. The primary objective of this research is to conduct cognitive testing of the content as related to the Spanish translation contained in the questionnaires and provide empirical information on how respondents of varying language abilities interpret, comprehend, and respond to the questions. The bilingual questionnaire may be revised based on the research results.This paper presents the main findings and recommendations using this rigorous approach to conduct cognitive tests of a bilingual questionnaire. Discussion will be focused on cultural contexts and effect on survey responses, as well as factors such as form-taking literacy (ability to follow the instructions and design of a questionnaire). Because of the magnitude of the 2010 Census, this paper merits the attention of the consumers of the decennial Census and scholars of cross-cultural survey methods.