Measurement Equivalence of English Versus Native Language Versions of the Kessler 6 (K6) Scale: An Examination in Three Asian American Groups.

Abstract

The use of languages other than English in population-based surveys is necessitated by the linguistic diversities in the United States. However, inclusion of multiple languages in survey data collection raises concerns about whether an instrument administered in different languages functions equivalently across groups. Using the Kessler Psychological Distress Scale 6 (K6), the present study examined differential item functioning (DIF) between surveys conducted either in English or the native language of the groups of Chinese Americans (n = 622), Korean Americans (n = 471), and Vietnamese Americans (n = 513). DIF analyses using a series of multiple indicator multiple cause (MIMIC) models showed that there were substantial differences between English and non-English versions in the endorsement of the K6 items, with patterns that differed by ethnicity. The K4 (depressed) showed DIF in all three groups: non-English survey users consistently showed a higher degree of endorsement compared to their English using counterparts. It is speculated that its translated expression in Asian languages may carry less associations with illness/disorder than the English word, thereby making it easy to endorse among Asian language survey users. Findings suggest a lack of measurement equivalence between the K6 administered in English and Asian languages and call for caution in cross-linguistic contexts.