Multilingual textual data: an approach through multiple factor analysis

Abstract

This paper focuses on the analysis of open-ended questions answered in different languages. Closed-ended questions, called contextual variables, are asked to all respondents in order to understand the relationships between open-ended and closedended responses across samples, as the latter are likely to influence word choice. We have developed "Multiple Factor Analysis on Generalised Aggregated Lexical Tables" (MFAGALT) to examine together open-ended responses in different languages through the relationships between word choice and the variables that drive that choice. MFA-GALT investigates whether the variability between words is structured in the same way as the variability between variables, and vice versa, from one sample to another. An application
to an international satisfaction survey shows the easy-to-interpret results proposed.