LANGUAGE ISSUES IN CROSS-CULTURAL TESTING

Cross-cultural researchers typically assume that administering verbal tests across cultural groups speaking a common language allows comparability of test scores, but this assumption is rarely examined. In this study, two forms (sentence and adjective) of a self-report measure of wellbeing were administered to Australian and Indian samples. Results indicated that for the Australian sample, both forms showed significant item-total and interitem correlations. However, Indian participants did not respond to sentence items in the same way as they did to adjective items. These findings highlight the importance of investigating the linguistic equivalence of measures used in cross-cultural assessment. They indicate that shorter, less complex ''linguistic stimuli'' may be more appropriate in cross-cultural contexts, and support the argument that, even with samples speaking a common language, cultural equivalence cannot be assumed.