Multilingual textual data: an approach through multiple factor analysis

  1. Belchin Kostov 3
  2. Ramón Alvarez-Esteban 2
  3. Mónica Bécue-Bertaut 3
  4. François Husson 1
  1. 1 Institut Agro, Univ Rennes 1, CNRS, IRMAR, 35000, Rennes, France
  2. 2 Department of Economics and Statistics, Universidad de León, Campus de Vegazana s/n, 24071 León, Spain
  3. 3 Department of Statistics and Operational Research, Universitat Politècnica de Catalunya, C/ Jordi Girona 1-3, 08034 Barcelona, Spain
Show affiliations +
Journal:
Statistica Applicata - Italian Journal of Applied Statistics

ISSN: 2038-5587

Year of publication: 2024

Pages: 339-357

Type: Article

DOI: 10.26398/IJAS.0035-015 GOOGLE SCHOLAR lock_openOpen access editor

More publications in: Statistica Applicata - Italian Journal of Applied Statistics

Sustainable development goals

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 applicationto an international satisfaction survey shows the easy-to-interpret results proposed.