Language Barrier

Language barrier dispositions occur in U.S. surveys when a household member or the sampled respondent does not speak or read English (or another target language) well enough to complete the interview. The language barrier disposition is used in all surveys, regardless of the mode (telephone, in-person, mail, and Internet). Language barrier dispositions in surveys in the United States are not common, but their frequency is growing. Approximately 20% of the U.S. population in 2005 spoke a language other than English in their home, according to the U.S. Census Bureau. Furthermore, the 2005 U.S. Census estimates show upwards of 5 million residents being “linguistically isolated,” in that they can speak little or no English. Language barriers are more likely to occur when data collection is conducted in central city areas and in rural areas of the Southwest.

The language barrier disposition functions as both a temporary and a final disposition. Cases may be coded temporarily with a language barrier disposition and then contacted again (in the case of a telephone or in-person survey) by an interviewer who speaks the same language as the household or the sampled respondent. In a mail or Internet survey, the survey organization may re-mail or resend a translated version of the questionnaire. However, cases with a language barrier final disposition often are considered eligible cases and thus are factored into survey nonre-sponse rates. The exception to this would be if the target population for a survey specified that respondents must speak English (or another specific language) to be eligible to complete the questionnaire.

Survey researchers may include a variety of categories within the language barrier disposition. One category of language barrier is used in telephone or in-person surveys for those cases in which there is no one in the household present at the time of contact who can speak or understand the language in which the introduction is spoken. Other categories of language barriers include cases in which the sampled respondent does not speak the language in which the interview is conducted or does not read the language in which the questionnaire is printed (for mail surveys) or displayed (for Internet surveys). Finally, a third category of language barrier occurs in in-person and telephone surveys when an interviewer fluent in the language spoken by the household is not available to be assigned to the case at the time of contact.

Because cases with a language barrier disposition increase the nonresponse rates in a survey, researchers fielding surveys in areas known to be multi-lingual often use practical strategies to ensure that households or respondents who do not speak the language in which the interview is to be conducted can complete the survey. In telephone and in-person surveys, these strategies usually include employing multi-lingual interviewers and arranging to have the survey questionnaire and all supporting materials translated into one or more additional languages. In mail and Internet surveys, these strategies usually include having the questionnaire and supporting materials translated into one or more additional languages and then re-sending [Page 414](via U.S. Postal Service or email, depending on the mode) the translated questionnaire and materials to respondents who reported that they are unable to read English.

...