Interpretation in cross-language research: tongues-tied in the health care interview?

Medication prescribing and use is a normative aspect of health care for the elderly, rendering medication taking by elderly persons problematic. In an earlier qualitative study, we examined how medicine-use is negotiated (used/refused/resisted, and assessed against expected outcomes) by older persons with limited fluency in English-the main language of health care in the study setting. In the present article, we describe a reflexive methodological review of that study's design, with a particular focus placed on interpreter-mediated data collection. We illustrate that what was heard in open-ended interviews (what became data) was influenced by not only what was asked and how, but also by how the interpreter 'heard' and conveyed dialogue to and from study participants. We illustrate differing accounts of the dialogue between an interviewer and participant provided via real-time interpretation and through a reflexive re-interpretation of talk-to-text transcripts, reflecting the different stakes in the research, and different capital available to study- and review-interpreters. Implications for research design and practice of cross-cultural and cross-language research are highlighted.