Weidmer B, Brown J, Garcia L
RAND, Santa Monica, CA 90401, USA.
Med Care. 1999 Mar;37(3 Suppl):MS89-96. doi: 10.1097/00005650-199903001-00010.
Collecting accurate health data on the growing number of ethnic minorities in the United States has increased in policy relevance in recent years. Today, most general population sample surveys conducted in the United States, including the CAHPS 1.0 Surveys, require translation into Spanish and often other languages as well. This article discusses the process used to translate the CAHPS 1.0 survey instruments into Spanish and the techniques used to evaluate these instruments.
The CAHPS team used a technique of translation-backtranslation to translate the survey instruments into Spanish and conducted cognitive testing and pretesting of preliminary versions of the 1.0 Survey Instruments in both English and Spanish.
The translation-backtranslation method produced Spanish-language versions of the survey instruments that were adequate for more educated respondents but were inadequate for less educated respondents and respondents who seemed to be less acculturated.
Adept translation of a survey instrument is an integral part of the instrument-development process, but it alone does not ensure that a culturally appropriate survey instrument will result. Producing a survey instrument that is culturally appropriate for Latinos in the United States may require modifying the English versions of instruments as well as subjecting Spanish-language instruments to more rigorous testing that includes cognitive testing, pretesting, and an evaluation of the reading level by a literacy expert.
近年来，收集美国日益增多的少数族裔的准确健康数据在政策方面的相关性有所增加。如今，在美国进行的大多数普通人群抽样调查，包括CAHPS 1.0调查，都需要翻译成西班牙语，通常还需要翻译成其他语言。本文讨论了将CAHPS 1.0调查工具翻译成西班牙语所采用的过程以及评估这些工具所使用的技术。
CAHPS团队采用回译技术将调查工具翻译成西班牙语，并对1.0调查工具的英文和西班牙文初步版本进行了认知测试和预测试。
回译方法产生的调查工具西班牙语版本对于受教育程度较高的受访者来说足够，但对于受教育程度较低的受访者以及文化适应程度较低的受访者来说不够。
熟练翻译调查工具是工具开发过程的一个组成部分，但仅靠翻译本身并不能确保产生一个符合文化背景的调查工具。要制作一个适合美国拉丁裔文化背景的调查工具，可能需要修改工具的英文版本，并对西班牙语工具进行更严格的测试，包括认知测试、预测试以及由识字专家评估阅读水平。
