Lydia Shi
Research Mentor: Stephanie Morales
Mentor Department: Michigan Program in Survey and Data Science, ISR
Author(s): Not Available
Session: Session 2 (10:00 AM – 10:50 AM)
Presentation Type: Poster 43
Abstract
Language and culture shape health experiences, access to care, and participation in research, yet many national health surveys are not designed to measure health equivalently across linguistically diverse populations. Self-rated health (SRH) is a commonly used question due to its ability to predict mortality. However, researchers do not fully understand what SRH measures or why it predicts mortality for some cultural groups but not others. Previous research suggests that cultural and linguistic differences in how health is conceptualized may underlie this inconsistency. This study aims to address this gap by exploring how respondents conceptualize health, both within and outside of a health context, using web probing techniques. We conduct a multilingual web-based health survey examining health perceptions, behaviors, and outcomes among diverse linguistic and cultural groups in the United States, including Latino and Chinese American populations. An experimental design is employed in which half of respondents receive the SRH question within a health context while the other half receive it outside a health context, allowing us to examine whether question placement influences health conceptualization. The questionnaire was developed, translated into Spanish and Chinese, and programmed in Qualtrics. Respondents are recruited using respondent-driven sampling (RDS), a method commonly used to target hard-to-reach populations. Responses to probing questions are coded and analyzed using a previously developed scheme that captures multiple health attributes (e.g., health behaviors, illness) and their associated tone (positive, negative, or neutral), allowing comparisons across language groups and experimental conditions. This project is expected to reveal differences in how cultural and linguistic groups conceptualize self-rated health. The resulting multilingual dataset will support comparative analyses and contribute to more inclusive survey design, improving how health is measured across diverse populations and reducing language-based disparities.
