Objective: To assess the impacts of survey languages on participation and representativeness of the study subjects in a health survey in a Chinese community in Australia.
Method: A random sample of 500 ethnic Chinese in Brisbane, Queensland, Australia was surveyed during November 2005 to February 2006 by using a bilingual survey questionnaire in their preferred languages, i.e. English or Chinese.
Results: 210 questionnaires were returned. Two-thirds of the participants chose to answer the questionnaires in Chinese. Besides being older with relatively lower income, they were more likely to be married, have a Chinese family doctor, and visit a Chinese medicine practitioner. Fewer of them have visited the Diabetes Australia website or read any educational information materials about diabetes.
Conclusions & Implication: The multilingual approach is crucial to improving participation and representativeness of samples from ethnic populations.